Measuring Airport Landside Capacity: Special Report 215 (1987)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Apecial Report 215 MEASURING AIRPORT LANDSIDE CAPACITY ~~ - I Transportation Research Board National Research Council i* ••;._ I... . Ff &%

1987 TRANSPORTATION RESEARCH BOARD EXECUTIVE COMMITTEE OFFICERS Chairman: Lowell B. Jackson, Executive Director, Colorado Department of Highways, Denver Vice Otair,nan. Herbert H. Richardson, Deputy Chancellor and Dean of Engineering, Texas A&M University System, College Station Executive Director: Thomas B. Deen, Transportation Research Board MEMBERS Ray A. Barnhart, Administrator, Federal Highway Administration, U.S. Department of Transportation (ex officio) John A. Cements, Principal for Transportation and Public Works, Sverdrup Corporation, Boston, Massachusetts (ex officio, Past Chairman, 1985) Donald D. Engen, Vice Admiral, U.S. Navy (retired), Administrator, Federal Aviation Administration, U.S. Department of Transportation (ex officio) Francis B. Francois, Executive Directo American Association of State Highway and Transportation Officials, Washington, D.C. (ex officio) E. R (Vald) Heiberg Ill, Chief of Engineers and Commander, U.S. Army Corps of Engineers, Washington, D.C. (ex officio) Lester A. Hoel, Hamilton Professor and Chairman, Department of Civil Engineering, University of Virginia, Charlottesville (ex officio, Past Chairman, 1986) Ralph L Stanley, Administrator, Urban Mass Transportation Administration, U.S. Department of Transportation (ex officio) Diane Steed, Administrator, National Highway Traffic Safety Administration, U.S. Department of Transportation (ex officio) George H. Way, Jr., Vice President, Research and Test Department, Association of American Railroads, Washington, D.C. (ex officio) Alan A. Altshuler, Dean, Graduate School of Public Administration, New York University, New York John R. Borchert, Regents Professor, Department of Geography, University of Minnesota, Minneapolis Robert Q. Bugher, Executive Director, American Public Works Association, Chicago, Illinois Dana F. Connors, Commissioner, Maine Department of Transportation, Augusta C. Leslie Dawson, Secretary, Kentucky Transportation Cabinet, Frankfort Thomas E. Drawdy, Sr., Secretar) Florida Department of Transportation, Tallahassee Paul B. Gaines, Director of Aviation, City of Houston Aviation Department, Texas Louis J. Gambaccini, Assistant Executive Director/Trans-Hudson Transportation of The Port Authority of New York and New Jersey, New York Jack R. Gilstrap, Executive Vice President, American Public Transit Association, Washington, D.C. William J. Harris, Snead Distinguished Professor of Transportation Engineering, Department of Civil Engineering, Texas A&M University, College Station Raymond H. Hogrefe, Director-State Engineer, Nebraska Department of Roads, Lincoln Thomas L. Mainwaring, Consultant, Trucking Industry Affairs, Ryder System, Inc., Miami, Florida James E. Martin, President and Chief Operating Officer, Illinois Central Gulf Railroad, Chicago Dennian K. McNear, Chairman, President and Chief Executive Officer, Southern Pacific Transportation Company, San Francisco, California Leno Meoghini, Superintendent and Chief Engineer, Wyoming Highway Department, Cheyenne William W. Millar, Executive Director, Port Authority of Allegheny County, Pittsburgh, Pennsylvania Milton Pikarsky, Distinguished Professor of Civil Engineering, The City College of New York, New York James P. Pitz, Director, Michigan Department of Transportation, Lansing Joe G. Rideoutte, Chief Commissioner, South Carolina Department of Highways and Public Transportation, Columbia Ted Tedesco, Vice President, Resource Planning, American Airlines, Inc., Dallas/Fort Worth Airport, Texas Carl S. Young, County Executive, Broome County, Binghamton, New York

Special Report 215 MEASURING AIRPORT LANDSIDE CAPACITY Transportation Research Board National Research Council Washington, D.C. 1987

Transportation Research Board Special Report 215 mode 4 air transportation subject areas 12 planning 55 traffic flow, capacity, and measurements Transportation Research Board publications are available by ordering directly from TRB. They may also be obtained on a regular basis through organizational or individ- ual affiliation with TRB; affiliates or library subscribers are eligible for substantial discounts. For further information, write to the Transportation Research Board, Na- tional Research Council, 2101 Constitution Avenue, N.W., Washington, D.C. 20418. . Printed in the United States of America NOTICE: The project that is the subject of this report was approved by the Governing Board of the National Research Council, whose members are drawn from the councils of the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine. The members of the committee responsible for the report were chosen for their special competences and with regard for appropriate balance. This report has been reviewed by a group other than the authors according to procedures approved by a Report Review Committee consisting of members of the National Academy of Sciences, the National Academy of Engineering, and the In- stitute of Medicine. This study was sponsored by the Federal Aviation Administration of the U.S. Department of Transportation. Library of Congress Cataloging-in-Publication Data National Research Council. Transportation Research Board. Measuring airport landside capacity. (TRB special report ISSN 0360-859X ; 215) Includes bibliographies. 1. Airports—Planning. I. National Research Council (U.S.). Transportation Re- search Board. H. Series: Special report (National Research Council (U.S.) Transporta- tion Research Board) ; 215. TL725.3.P5M43 1987 387.7'36 87-22021 ISBN 0-309-04457-X

Committee for the Airport Capacity Study MARJORIE BRINK, Chairman, Peat Marwick Airport Consulting Services, San Francisco, California MARGARET M. BALLARD, Sverdrup Corporation, Fairfax, Virginia GEORGE J. BEAN, Hillsborough County Aviation Authority, Tampa, Florida FRANK T. BISHOP, Houston Intercontinental Airport, Houston, Texas GEORGE W. BLOMME, Port Authority of New York and New Jersey, Flush- ing, New York THOMAS H. BROWN, United Airlines, Chicago, Illinois WILLIAM C. COLEMAN, Smith Barney, Boston, Massachusetts KENNETH McK. ELDRED, Ken Eldred Engineering, Concord, Massachusetts JOHN GLOVER, Port of Oakland, Oakland, California ADIB KANAFANI, University of California, Berkeley, California PETER B. MANDLE, Peat Marwick Main & Co., San Francisco, California DORN CHARLES MCGRATH, JR., George Washington University, Wash- ington, D.C. FRANCIS X. MCKELVEY, Michigan State University, East Lansing, Michigan ROBERT S. MICHALL, Regional Airport Authority of Louisville and Jefferson County, Louisville, Kentucky RAY A. MUNDY, University of Tennessee, Knoxville, Tennessee E. WAYNE POWER, Lester B. Pearson International Airport, Toronto, Canada J. DONALD REILLY, Airport Operators Council International, Washington, D.C. JAMES W. SPENSLEY, Stapleton International Airport, Denver, Colorado Liaison Representatives GARY L. OLIN, Federal Aviation Administration, Washington, D.C.

Transportation Research Board Staff ROBERT E. SKINNER, JR., Director for Special Projects ANDREW C. LER, Project Director NAOMI C. KASSABIAN, Associate Editor

Preface Congestion at airport terminal buildings, access roads, and parking areas increasingly threatens the capability of airports to serve additional passengers and air cargo. Measuring the capacity of these airport landside facilities and services is becoming as critical to operations of major airports as capacity measurements for the airside taxiways and runways that serve aircraft. Yet the analysis procedures for assessing airport landside capacity are less developed than techniques for measuring airside capacity, and no generally accepted standards exist for gauging the level of service provided by landside facilities and their operations. The Federal Aviation Administration, in recognition of this situation, commissioned the Transportation Research Board to review existing capacity assessment techniques and recommend guidelines that can be used by airport operators, planners, and others who must measure airport landside capacity. To undertake this study, TRB convened a committee that included airport executives, planners, researchers, consultants, the airlines, and aviation indus- try organizations. Meeting over a period of about one year, the committee concluded that current knowledge about the performance of various airport landside components is inadequate to support airport landside service stan- dards at this time. Instead, the committee proposed a process for measuring airport landside capacity that takes an important first step toward developing such standards. Part I of the report contains a general description of the airport landside, how landside capacity is defined and measured, a process for landside capac- ity assessment, and research needs. In Part II guidance is presented for

vi PREFACE applying these definitions and the process to specific landside functional components likely to represent constraints on the airport's ability to satisfy demand. The report represents the collective efforts of the committee, TRB staff, and a host of airport operating agencies who provided valuable data and assis- tance. The study was performed under the overall direction of Dr. Damian J. Kulash and Robert E. Skinner, the former and current Directors for Special Projects. Dr. Andrew C. Lemer, the project director, conducted the review of existing capacity measurement techniques and drafted much of the report under the direction of the committee. Special appreciation is expressed to Nancy A. Ackerman, TRB Publications Manager, for publication of the final report, and to Marguerite Schneider and Frances E. Holland for typing drafts and the final manuscript.

Contents ExEcUTIvE SUMMARY .1 Part I Definitions and Process 1 BACKGROUND ..............................................9 Motivation for a Landside Capacity Study, 10 Current Issues Affecting Landside Capacity, 12 Scope of the Report, 15 BASIC DEFINITIONS ........................................17 Airport Landside, Airside, and Community, 17 Functional Components and Capacity Determinants, 19 Capacity Analysis Period, 28 LANDSIDE CAPACITY ASsEssNrr.r'r PROCESS ...................33 Step 1: Identify Goals and Objectives of the Assessment or Problem To Be Solved, 34 Step 2: Specify Landside Components for Assessment, 36 Step 3: Describe Each Component, 36 Step 4: Describe How Components Relate, 36 Step 5: Collect Data on Demand Characteristics and Operating Factors, 37 Step 6: Collect Data on Community Factors, 40

Step 7: Estimate Component and Total Service Levels, 41 Step 8: Estimate Current and Maximum Service Volumes, 42 Intermediate Decision: Is Landside Capacity Adequate? 42 Step 9: Examine Trade-Offs in Service Among Components, 43 Step 10: Identify Short-Term Measures To Improve Capacity, 43 Step 11: Review Long-Term Planning and Management Implications, 44 4 COMMUNITY FACTORS .....................................45 Sources of Concern, 46 Types of Restrictions, 49 Assessing Community Factors, 50 5 RESEARCH NEEDS .......................................... 56 Part II Assessing Capacity and Service Levels of Functional Components 6 AIRc1Fr PARKING PosrrIoN AND GATE .....................61 7 PASSENGER WAITING AREA.................................72 8 PASSENGER SECURITY SCREENING ...........................81 9 TERMINAL CIRCULATION .......... . ........................ 88 10 TIcKEr COUNTER AND BAGGAGE CHECK.....................94 11 TERMINAL CURB.........................................102 12 PARKING AREA ..........................................112 13 GROUND ACCESS ........................................120 14 BAGGAGE CLAIM ........................................127 15. Cusroits AND IMMIGRATION..............................135 16 CONNECTING PASSENGER TRANSFER ........................141 17 LANDSIDE SYSTEM AS A WHOLE ...........................148

GLOSSARY . 155 APPENDIX A FRAMEWORK FOR DEFINING AIRPORT LANDSIDE SERVICE-LEVEL TARGETS ........................162 APPENDIX B AIRPORT LANDSIDE CAPACITY ANALYSIS METHODS . .165 STEERING COMMITrEE BIOGRAPHICAL INFORMATION .............186

Executive Summary The Federal Aviation Administration (FAA) expects the annual number of passengers using the nation'S airports to grow more than 70 percent above 1986 levels in the next 10 years. To accommodate this growth, many airports must add new facilities or make better use of existing facilities or do both. Airport operators and local officials must make decisions about airport use and expansion against a backdrop of local economic, environmental, and social consequences. Airlines and the FAA, on the other hand, view the operation and expansion of individual airports from the standpoint of the role they play in providing an efficient national air transport system. Regardless of their different perspectives, each needs reliable estimates of airport capacity to make decisions about airport expansion and operations. An airport may be divided into two parts: the airside—runways, taxiways, and air traffic control systems used by aircraft and pilots—and the landside— aircraft parking positions and gates, terminal buildings, baggage services, access roadways, and automobile parking structures used by passengers. Capacity problems may occur in either airside or landside facilities and services. Extensive research and practical experience have produced widely accepted procedures for assessing an airport's airside capacity. The FAA sanctions these procedures, and they are applied throughout the United States and in many other countries. However, similar guidelines are not available for assessing landside capacity. As a first step toward developing such guidelines, a special 18-member Transportation Research Board committee reviewed current practice and recommended a process for assessing the capacity of airport landside facilities and services. The committee concluded that gener- ally accepted definitions and procedures for capacity assessment are needed.

2 MEASURING AIRPORT LANDSIDE CAPACITY Data collection and research will be required to produce more definitive guidelines that will yield consistent capacity measures for the many diverse conditions found in airports throughout the United States. MEASURING CAPACITY Landside capacity refers to the capability of the airport's landside facilities and services to accommodate passengers, visitors, air cargo, ground access vehicles, and aircraft. Of these, the capability to serve air passengers is the greatest concern. Other aspects of landside operations such as cargo shipments and aircraft maintenance are included in this report only to the extent that they directly influence passenger service. Passengers impose a variety of demands on parking facilities, ticketing, baggage claim, and other landside compo- nents. These demands are influenced by when and how passengers arrive at the airport, the number of bags they carry, their age and trip purposes, the number of people accompanying or meeting them, and myriad other characteristics. Estimates of passenger capacity are meaningful only when they are refer- enced to the service level provided to passengers. Service level includes factors such as waiting time, processing time, walking distance, crowding, and availability of passenger amenities for comfort and convenience, many of which may be difficult to measure. Passenger demands and facilities opera- tions interact to determine service level and capacity, making the measure- ment of capacity an iterative process. When passenger demands are large, landside components may operate at maximum throughput rates, which reflect the greatest number of passengers that can be processed in a given time by a component or group of components. Typically, however, maximum throughput can be sustained only during peri- ods when demand is high and then only briefly, because significant passenger delays and crowding usually develop and disrupt operations. The recom- mended measure of landside capacity is service volume, which is the number of passengers that can be accommodated by a functional component or group of components in a particular time period relative to a particular demand at a given service level. Service volume is typically measured for periods of 15 mm, 1 hr, 2 hr, or sometimes a full day. Although there are a number of landside service-level indicators that may be important at a particular airport, total passenger processing time (including service time and delay) and crowd- ing are the most important. Many airports could accommodate a greater number of passengers if new demands occurred during quiet periods of the day or if passenger processing times and crowding were allowed to increase. The level of acceptable delay

EXECUTIVE SUMMARY 3 and crowding that may limit capacity will in general vary from airport to airport. Airport operators may set explicit service-level targets that can be used as guidelines to assess a particular airport's landside capacity and guide decision making about management and development of landside facilities and services. The FAA, airlines, and airport operators need to establish common service-level targets that can be used when comparing conditions among airports and discussing issues of significance to national airport system operations. CAPACITY ASSESSMENT PROCESS Landside capacity assessment must respond to a variety of questions and issues about airport operations and planning, ranging from how existing gates and ticketing facilities should be used to when a new terminal building may be required. The specific problems a capacity assessment must address determine which component facilities and services must be considered. An airport's landside capacity can be measured only in terms of its individual functional components. The recommended assessment process is based on this premise that land- side capacity cannot be assessed without estimating service levels and service volumes for individual components (Figure ES-i). Specific passenger demand must be known or assumed, as well as relevant airline and airport operating policies and procedures. Interaction and feedback among particular steps in the process are critical. Service levels and capacity measures can be assessed only with reference to one another. Performance of any single landside component may depend on performance of other components with which that component interacts. Service-level targets set for current conditions may have long-term implications for the airport's future development. Only by suc- cessively considering how individual components perform and how they interact with one another and with demand can their potential service levels and service volumes be determined and a meaningful estimate of landside capacity be derived. All landside components are important, but not all are likely to cause passenger delay and crowding or become significant to determining an air- port's landside capacity. Newsstands, public telephones, restaurants, and rest rooms are essential amenities, yet they are seldom a basis for estimating landside capacity. The committee identified critical landside components or component groups and assembled guidelines for assessing the capacity of each:

C w E Ce Ce 0 Ce Ce 4 (C 0. cx 0 4 MEASURING AIRPORT LANDSIDE CAPACITY Aircraft parking position and gate Passenger waiting area Passenger security screening Terminal circulation (corridors, stairs, etc.) Ticket counter and baggage check Terminal curb Parking area Ground access 1. Identify Goals and ObjectIves of the Assessment or Problem To Be Solved 2. SpecIfy Landside Components for Assessment I 3. Describe Each Component 4. Describe How Components Relate (functionally and physically) S. Collect Data on Demand 6. Collect Data on Community Characteristics and Operating F Factors actors 7. Esilmate Component and S. Estimate Current and Total Service Levels Maximum ServIce Volumes INTER ME EC1SION: IS LANDSIDE CAPACITY ADEOUATE? ExamIne Trade-Offs In Service Among Components IdentIfy Short-Term Measures To improve Capacity ii. Review Long-Term Planning and Management implications I Long-Range Planning FIGURE ES-i Landside capacity assessment, management, and planning proCess.

EXECU71VE SUMMARY 5 Baggage claim Customs and immigration Connecting passenger transfer The committee also described how capacity analyses for individual compo- nents may be used to estimate capacity of the landside system as a whole, particularly within a context of strategic management and longer-range planning. COMMUNITY FACTORS Airport planning and operational studies must take into account how the surrounding community may influence capacity. In addition to air passengers, this community includes cargo shippers and other airport users, neighboring residences and businesses, and local and state government. Airport operators must work with this community, the airlines, and the FAA to operate and develop the airport to meet the demand for aviation services. If the community perceives that the benefits of these services are outweighed by such problems as noise and highway traffic associated with the airport, they may seek to restrict the airport's operations. Restrictions may take the form of policies that limit expansion of airport landside facilities or they may curtail permissible aircraft operations. The community's influence may also include promoting development of new landside facilities to attract users and investment. In either case, the demand on landside facilities or the ability to provide new facilities will be affected. NEED FOR FURTHER RESEARCH Airport operators and others cannot conduct consistent landside capacity assessments at different airports without uniform service-level targets that distinguish between excellent, adequate, and poor service. Although the study identifies which landside services should be covered and how service-level targets can be related to different types of air travel, the committee concluded that available data are inadequate to support proposal of firm service-level targets. Research to assemble a data base on landside operating and service conditions is needed: specific data are needed to describe typical crowding, delay, and other relevant service-level indicators for all types of airports in the United States. The FAA should work with airport operators and airlines to develop a coordinated research program to collect such data, which industry could use to establish service-level targets.

Part I Definitions and Process Although many people have some intuitive understanding of what conditions occur when demand on an airport's landside facilities and services is close to that airport's landside capacity, there appears to be little general agreement among airport and aviation professionals on how landside capacity should be defined and assessed. In Part I of this report major landside capacity issues being faced by operators of many airports in the United States are reviewed, the terms needed to establish a common understanding of landside capacity are defined, a process for assessing the landside capacity of a particular airport is described, and research is recommended that is needed to move the airport and aviation industry toward generally accepted standards for measuring and judging whether the landside capacity of an airport is adequate to meet the demand on its landside systems.

1 Background Airports serve a broad and complex range of needs related to the movement of people and goods. Passengers and cargo shippers gain access to national and international air transportation through airports. Airlines and other operators of aircraft use airport facilities to serve passengers and shippers and to operate, maintain, and store their aircraft. Concessions offer a variety of products and services to passengers, visitors, and airport and airline em- ployees. The community served by an airport may depend on the airport for transportation, jobs, business opportunities, and recreation, but may be ex- posed to aircraft noise and to the growing need for land to expand airport facilities. Although it is difficult to draw a precise line separating the two areas, airport and aviation professionals speak of airports in terms of "airside" and "landside" components. The Federal Aviation Administration (FAA) defines the airside as "the airfield and its components (i.e., runways, taxiways, and apron-gate areas)" (1). Aircraft operate within this airside system and the accompanying airspace under the federal government's air traffic control procedures and regulations. In this report, apron-gate areas are included in the landside and described as aircraft parking positions and gates.1 The airport landside includes terminal buildings, access roads, and parking areas and the services provided for users of these facilities. Passengers, employees, cargo, and aircraft maintenance activities use an airport's landside facilities and services. However, this report addresses only those aspects of an airport's landside system that relate directly to air passenger capacity and available methods for assessing that air passenger capacity. The remainder of

10. MEASURING AIRPORT LANDSIDE CAPACiTY this chapter is devoted to discussion of the need for landside capacity assess- ment procedures and key issues currently affecting airport landside capacity. MOTIVATION FOR A LANDSIDE CAPACITY STUDY The airport landside is controlled to a great extent by the local community that owns and is served by the airport. This community includes airport users, airport neighbors, and local and state governments. In addition to this airport- related community, the airport operators must also work cooperatively with airlines and the FAA. Each of these groups may deal directly with any of the others on matters affecting the airport (see Figure 1-1). Airlines operate at an airport generally under terms of leases on terminal building space and gates. The FAA administers programs to support airport planning and development of airport facilities and to ensure an effective and safe national air transporta- tion system. The interaction among these groups is the context within which airport operating and development decisions are made—decisions that influ- ence and are influenced by landside capacity. Airport-Related Community - Airport Users - Airport Neighbors - Local and State Governments , Airport Operators Airlines 1 ( FAA FIGURE 1-1 Groups participating in airport landside management.

BACKGROUND 11 Extensive research in the United States and abroad has produced methods for airside capacity assessment and facilities planning and design. The FAA, responsible for airside safety and for development of the national air transpor- tation system, has published recommended procedures for airside capacity assessment (2). In comparison with the airside analysis, procedures for landside capacity assessment, planning, and design are less clearly defined (3), and although several professional organizations and the FAA have published guidelines, there are few generally accepted procedures and practices. A representative of the U.S. Department of Transportation, participating in a 1975 conference on airport landside capacity sponsored by the FAA and the Transportation Sys- tems Center, wrote (4): The problem of improving landside capacity of airports is an elusive one. At present, we do not even have a standard against which to measure existing levels of airport service or by which to outline desirable levels of service. In the decade since that conference and in a climate of substantial change in the aviation industry, airport professionals have been able to shed little light on the problem. The FAA expects the number of passengers using the nation's airports in 1998 to increase more than 70 percent over 1986 levels, from 409.6 million annual enplanements in 1986 to 696.8 million by 1998. During this same period, annual air passenger carrier operations are forecast to grow 32.5 percent, from 12.3 million to 16.3 million. Air taxi and commuter operations are projected to grow from 6.9 million to 10.9 million, or nearly 58 percent over the period (5). Such growth will present challenges to the nation's air transportation system. The FAA forecasts that larger and perhaps more fully loaded aircraft will be operating at the nation's airports. These aircraft and their passengers will place increasingly greater demands on airport landside facilities. Al- though the technical issues of landside capacity may be elusive, passengers know by the long delays and crowded conditions when landside facilities and services are being strained. Although the majority of the nation's airports generally have few landside problems, the need for procedures to assess and manage landside capacity will continue to grow. Some efforts have been made by international organizations to provide guidance on landside capacity analysis. The International Air Transport Asso- ciation (IATA) and International Civil Aviation Organization (ICAO) have published manuals (6-8). However, practice in the United States depends on a small number of typically older, key technical publications (9-11) and the accumulated experience of individual practitioners.

12 MEASURING AiRPORT LANDSIDE CAPACITY Recognizing this situation and the importance of the landside in efficient operation of the nation's air transportation system, the FAA requested the Transportation Research Board (TRB) to conduct a study in which technical guidance to support assessments of airport landside capacity would be developed. CURRENT ISSUES AFFECTING LANDSIDE CAPACITY Many aspects of landside capacity depend on the unique character of an individual airport and the community values within which the airport operates. Broad organizational, legal, and political issues influence the judgment of the practical limits of landside capacity. Although the details of facility sizes and operating practices that influence landside capacity may differ from one airport to another, the broad issues are often similar. Four major issues in particular influence landside capacity at many airports—airline leases on airport facilities, shifting patterns of airline operation, changing procedures for passenger processing, and a range of community effects on airport opera- tions. The landside capacity assessment process must be able to respond to such institutional and management issues as well as to the details of facilities and services of airport and airline operations. Airline Leases on Airport Terminal Facilities Airlines typically lease airport terminal facilities from the airport operators. In the majority of cases, these leases give each airline exclusive rights to use particular gates and terminal areas. The passengers of one airline may some- times experience crowding and delay associated with inadequate capacity whereas at the same time another airline's facilities at the same airport stand relatively empty. Each airline is responsible for the complex task of using its leased facilities efficiently, taking into consideration its operating costs, customer relations, and competitive position in scheduling this use. Until the early 1980s, airline leases historically had long terms. Such long- term lease agreements have traditionally given greater security to the airport operators and supported the sale of reasonably priced long-term revenue bonds to fund capital improvement programs.2 Three-fourths of the major airports in a 1975 survey had lease terms exceeding 10 years. Of those airports at which airlines were committed to pay the difference between actual airport operating costs and revenues in a given

BACKGROUND 13 year, 89 percent had lease terms longer than 10 years.3 At some airports long- term ground leases permit the airlines to construct their own terminals, to which they retain ownership rights. Under such long-term and legally binding arrangements, airlines can control much of an airport's facilities, and the airport operator may have a limited ability to respond to changes in air passenger demand and airline route structures, competitive practices, and commercial health. Airline deregulation appears to be encouraging some airport operators to adopt lease terms that afford greater flexibility in facilities management. Some airlines as well appear to prefer the reduced commitment associated with shorter lease terms, although other airlines recognize the influence of lease terms on an airport's ability to raise capital and continue to enter into long- term leases. Airports with strong local markets—rather than a great deal of transfer traffic—are typically in a stronger financial position and may there- fore operate more independently of the airlines. At some U.S. airports and most foreign airports, airlines share gate and other terminal facilities. Under joint gate use, gates are assigned to aircraft as they arrive. Preference is given so that airlines can expect to have all their gates grouped in a particular area, but no airline holds exclusive rights to any gate. Joint gate use may allow.the airport to serve a greater number of aircraft and passengers at any given time with a smaller number of gates. At Hawaii's major airports, for example, where gates are shared in this manner, numbers of annual aircraft operations per gate are higher than might otherwise be ex- pected, and are comparable with those at the largest and busiest mainland airports. Airports at which a large airline hub-and-spoke operation is based may have very high gate utilization as well under an exclusive gate use strategy, as explained later. Changes in Airline Operating Practices and Market Behavior Many airlines have tried to attract customers and improve efficiency by assigning seats and issuing boarding passes to passengers before their arrival at the airport. Machines are now available in many airports (and even down- town or at local shopping centers) that enable a passenger to purchase a ticket and receive a boarding pass by using a credit card. This practice reduces both the number of people who must stop at the ticket counter before going to the gate and the need for ticket counter space in the terminal. The effective capacity of the terminal gets a small boost. The airlines' continuing efforts in an open competitive environment to fly their aircraft fully loaded have generated such practices as discounted standby

14 MEASURING AIRPORT LANDSIDE CAPACiTY and bulk sales of available seats. New passenger traffic may be attracted by discounted fares. Even when total passenger traffic does not grow, increasing crowds of passengers waiting in areas adjacent to gates for changes in their flight, seating, or class of service may be the sign of a terminal area experienc- ing a landside capacity problem. Particularly significant for landside management are the effects of airline hub-and-spoke operations. For such an operation an airline selects an airport as a central point for many of its routes and uses that airport (the "hub") as a transfer point to allow passengers a wide combination of origins and destina- tions by using a relatively small number of direct flights (the "spokes"). Gate areas may be very crowded for one hour when many flights converge and then may be empty the next hour. Requirements for terminal facilities serving such an operation are quite different from those for an operation serving primarily origination and destination traffic. Because there are several such peaks of activity during the day, gate utilization is much higher than is usually the case at an airport without such an airline hub-and-spoke operation. Airlines also try to serve demand most profitably by using aircraft that match seating availability closely to route demand. High-density routes fed by hub-and-spoke operations call for larger-capacity aircraft. When the Boeing 747 and other widebody aircraft were first introduced, airport gate lounges, concourses, and baggage claim areas had to contend with a larger surge of passengers arriving at one time than had been previously experienced. More gate lounge and baggage claim space was required to handle one widebody aircraft arrival than was needed for several small aircraft. This is still a problem at some airports when widebody service is introduced. Anticipated introduction of the 500-seat Boeing 747-500 is caus- ing concern among airport operators, and even larger aircraft may be de- veloped (13). Changes in Passenger Processing Curbside baggage check operations coupled with advance ticketing and seat assignment reduce the number of passengers who must stop at the ticket counter. But some airport operators believe that concerns for airport security in the 1980s could lead to broader restrictions on curbside baggage checking. Such checking has already been eliminated at many airports in those areas serving international traffic. The effective capacity of some terminal facilities might then be reduced as more space is required for passenger and baggage processing. The situation could be further aggravated if proposals to curtail sharply the practice of carrying luggage onto the aircraft are adopted. At the destination, reduced interline transfers (a result of the increased airline hub-and-spoke operations mentioned earlier) may improve the speed of

BACKGROUND 15 baggage arrival and subsequent passenger departure from the airport by reducing the number of passengers and baggage being transferred from one airline to another. Increased use of computers for customs and immigration inspection of arriving international passengers may have a similar effect. Community Effects on Operations Although some communities actively seek to expand their airport business activities and to attract new service, a few jurisdictions have imposed restric- tions on aircraft operations to reduce the levels of airport noise to which neighbors of the airport are exposed or to limit demands on already inadequate facilities. These restrictions may limit flights scheduled for nighttime opera- tions or may limit the number of passengers an airport can serve annually. Sometimes restrictions may be placed on the types of aircraft operating at the airport. Such restrictions may have a direct effect on how gates, roadways, and other airport facilities can be used, and thereby influence capacity. SCOPE OF THE REPORT This report is written to support the assessment of landside capacity at a particular airport. Because of the great variety of conditions at different airports, a single set of simple numerical parameters and procedures applica- ble to all airports is not feasible. Instead, assessment should be tailored to the specific problem at hand and conducted at a level of complexity appropriate to the type of decision being made. Airport and aviation professionals may use landside capacity assessment to decide whether a new terminal building is needed or to estimate the number of x-ray devices to be staffed and in operation during a midday period or how much space to cordon off for baggage claim for a special charter flight. Guidelines on how to conduct landside capacity assessments are given in Part I of this report by providing basic definitions (Chapter 2) and an assessment process (Chapter 3) and by describing the influence of community factors on available landside capacity (Chapter 4). In Part II the capacity assessment process is applied to key landside components. NOTES See the glossary for definitions of terms. A study by the Office of Technology Assessment (OTA) (12) found that in 1975-1976 approximately 30 to 50 percent of the revenues of commercial airports came from concessions and rentals of terminal facilities. Airside-based charges such as aircraft landing fees and aviation fuel sales compose another major source of airport revenues.

16 MEASURING AIRPORT LANDSIDE CAPACITY 3. This basis for leasing is termed the "residual-cost" approach to airport financial management and was used at an estimated 58 percent of larger airports in 1983. REFERENCES Techniques for Determining Airport Airside Capacity and Delay. Report FAA- RD-74-124. Federal Aviation Administration, U.S. Department of Transportation, June 1976. Airport Capacity and Delay. Advisory Circular 150/5060-5. Federal Aviation Administration, U.S. Department of Transportation. Sept. 23, 1983. W. Hart. The Airport Passenger Terminal. Wiley, New York, 1985. J. W. Barnum. Importance of Airport Landside Capacity. In Special Report 159. Airport Landside Capacity, TRB, National Research Council, Washington, D.C., 1975. FAA Aviation Forecasts-Fiscal Years 1987-1998. Report FAA-APO-87-1. Federal Aviation Administration, U.S. Department of Transportation, Feb. 1987. Air Terminal Reference Manual, 6th ed. International Air Transport Association, Montreal, Quebec, Canada, Sept. 1978. Master Planning. In Airport Planning Manual, Part 1, Document 9184, Interna- tional Civil Aviation Organization, Montreal, Quebec, Canada, Dec. 1979. Airport Terminal Capacity Analysis. International Air Transport Association, Montreal, Quebec, Canada, Jan. 1982. Planning and Design Considerations for Airport Terminal Building Development. Advisory Circular 150/5360-7. Federal Aviation Administration, U.S. Department of Transportation, Oct. 5, 1976. Planning and Design Guidelines for Airport Terminal Facilities. Advisory Circu- lar 150/5360-7A. Federal Aviation Administration, U.S. Department of Transporta- tion, in preparation. Ralph M. Parsons Co. The Apron and Terminal Building Planning Report. Report FAA-RD-75-191. Federal Aviation Administration, U.S. Department of Transporta- tion, July 1975. Airport System Development. Report OTA-STI-231. Office of Technology Assess- ment Washington, D.C., Aug. 1984. Too Large and Too Soon. Airports International, Nov. 1986, pp. 9-10.

2 Basic Definitions Despite general agreement among airport and aviation professionals on major principles, precise meanings of many terms related to landside capacity may vary. To establish a basis for developing a landside capacity assessment process, definitions for the key terms that are used throughout this report are presented in this chapter. The glossary at the end of the report includes these terms and others that have particular importance in the understanding of landside capacity. The main focus in this report is on air passenger capacity at commercial service airports. Nevertheless, any discussions of landside and landside capac- ity must also consider air cargo and general aviation users as well as airport employees and visitors who use the airport's landside facilities and services. AIRPORT LANDSIDE, AIRSIDE, AND COMMUNITY For this report, the airport landside is defined to include an facilities and services associated with an air passenger's ground-based journey from trip origin to the aircraft and from the aircraft to the destination of the trip. The landside includes ground access, terminal buildings, roadways and parking, ticketing, baggage handling, and aircraft parking positions and gates (Figure 2-1). Landside capacity depends on the type, sie, configuration, and condition of such facilities and on how they are staffed, operated, and regulated. In broad terms, the landside includes the, entire ground transportation system linking an airport to the region it serves. Except for special airport access services such as those to remote terminals, mass transit stations, and 17

I Access _______ Ground Access Terminal Terminal II Gates and Taxiways I I Regional Roads I Roads Curb Buitding Aircraft and I Transit and Parking H Parking Runways Transit I I I Shunte seice Positions II 'I I II : II - Residential and Commercial Interests Affected by Airport Economic, - - - - Social, and Environmental Impact (employment, noise, etc.) I Airport Boundary I AIRPORT-RELATED COMMUNITY L---------------- FIGURE 2-1 Functional view of an airport.

BASIC DEFINITIONS 19 central business districts or limousine services, the dependence of an airport on this system and on other municipal services decreases with their distance from the airport. The airport operator's ability to control these facilities and services decreases with increasing distance as well. As a practical matter, the boundary of the landside may often be drawn at the major links between the region's ground transport and municipal service systems and the airport, along the airport's property boundary, and at the door of the aircraft. Although many components of the airport landside influence the air pas- senger's comfort, safety, and convenience, landside capacity is determined primarily by those components that are or may become real constraints to passenger movement to and from the aircraft. The airside includes the facilities and services used by aircraft to transport passengers and cargo. In this report, taxiways and runways, airspace, and air traffic control systems are included in the airside. Passengers are in effect part of the airside while they are on board an aircraft and making no demands on landside facilities (although the aircraft itself is still parked and therefore on the landside). The airport-related community includes airport users, neighbors, and local and state government. This community provides land, labor, financial support, infrastructure, and municipal services to the airport and receives a variety of benefits (e.g., air service, employment) and some adverse effects (e.g., noise, traffic congestion). The community's perceptions of and response to these benefits and adverse effects can influence an airport's capacity (see Chapter 4). FUNCTIONAL COMPONENTS AND CAPACITY DETERMINANTS An airport's landside is a complex collection of individual functional compo- nents such as the following (see glossary for definition of terms). Those components in italics are considered potentially critical to landside capacity and are discussed in detail in Part II of this report. Environs Ground access Remote terminals Transit links Highway links Remote parking and shuttle Access roads/interchanges

20 MEASURING AIRPORT LANDSIDE CAPACITY Air freight and air-related industrial land and buildings Airport grounds Approach roads Remote processing facilities and services Parking areas Taxis Private vehicles Rental cars Circulation/distribution roads Cargo docking area Terminal building General configuration Pier Satellite Linear Transporter Terminal curb Departures Arrivals Terminal transition Entryways and foyers Lobby area Airline facilities Office Ticket counter Bag gage check Baggage claim Circulation Corridors, stairs, escalators Security screening Passenger amenities Food/beverage, news/tobacco Drugs, gifts, clothing, florists Barber and shoeshine Car rental and flight insurance Public lockers and telephones Post office

BASIC DEFINITIONS 21 Amusement arcades, vending machines Restrooms and nurseries Showers and health club Chapels VIP waiting areas Departure lounges (passenger waiting areas) International facilities/Federal Inspection Services (FIS) Immigration and Naturalization Customs Plant and Animal Health (Agriculture) Public Health Airline operations Flight operations/crew ready rooms Valuable/outsized baggage storage Air freight and mail Administrative offices Airport operations and services Offices Police Medical and first aid Fire fighting Building maintenance Building mechanical systems Communications facilities Electrical equipment Government offices Air traffic control Weather FIS and public health Conference and press facilities Apron-gate system Aircraft parking positions and gates Passenger enplanement/deplanement Waiting areas Bridge Stairs Mobile conveyance

22 MEASURING AIRPORT LANDSIDE CAPACITY Apron utilities Fuel. Electric power Aircraft electrical grounding Apron lighting and marking Cabin services, aircraft maintenance Aircraft parking and circulation Support systems Power, water, and sewer Fuel storage Development restricted areas Clear zones Noise exposure zones Individual functional components of an airport interact to provide service to air passengers. Some of these components can become bottlenecks or "choke points" in the processing of passengers, causing delays and crowding. Often the symptoms of inadequate capacity may be apparent primarily to those passengers actually using the affected component. Whether or not such choke points occur depends on the design and operation of the functional compo- nents and on the demands placed on these components. Any of these func- tional components may become a constraint on landside capacity. The term "demand characteristics" refers to the number of air passengers and those aspects of their behavior that materially affect the ability of a functional component or gmup of components to accommodate passengers. Demand characteristics include distribution of passenger arrivals over time, modes of travel to and from the airport, number of bags carried and checked, passenger's age, trip purpose, number of people accompanying the passenger, and whether the passenger has a ticket and boarding pass. Airlines often try to tailor their services to their passengers' demand characteristics. Landside capacity refers to the capability of an airport's landside or its functional components to accommodate passengers, airport visitors, air cargo, ground access vehicles, and aircraft. This report focuses on air passenger capacity, the number of air passengers who can be served by airport landside components in a given period of time. A variety of passenger capacity indicators exist, reflecting either flow rates (passengers per unit of time) or crowding (the number of passengers within a specific area during a given time period). Because those waiting at the airport are only temporary occupants and waiting areas may fill and empty several times during a typical day, even crowding indicators are a reflection of passenger flow through the landside.

BASIC DEFINITIONS 23 Flow-rate capacity indicators vary between the maximum throughput' and a lower service volume that results from service-level consideration, demand characteristics, and other limitations. Crowding capacity indicators at max- imum throughput may signal crush conditions or reflect a lower service volume that maintains service levels consistent with passenger safety, health, comfort, and convenience. For both flow rates and crowding, service volume is the principal capacity indicator used throughout this report. Capacity can be evaluated for an individual functional component of the airport landside, such as a ticket counter or aircraft gate, or for groups of components. Those components that are most likely to become constraints on landside capacity (Figure 2-2) are described in detail in Part II. For purposes Enpianing Deplaning Passengers Passengers Aircraft Parking Positions and Gates Waiting Area and Hoidroom Customs and immigration Passenger Security Screening interline transfers Terminal Circuiation (corridors, stairs, escalators, etc.) Ticket Counter/Baggage Check Baggage Claim Terminal Curb Terminal Curb Parking Areas Parking Areas Airport Ground Access (road and transit operations) 500t,• indialdani p0000ng.ru may tallow a va1.ty of patty, mttlng some components and avoiding others entirely. FIGURE 2-2 Landside components most likely to determine capacity. of this report, components such as concession areas, rest rooms, and tele- phones, although important passenger amenities, are not a basis for defining airport landside capacity. Maximum Throughput Maximum throughput is the maximum rate at which passengers can be processed by a functional component or group of functional components. In

24 MEASURING AIRPORT LANDSIDE CAPACiTY practice this rate is actually observed only when demand equals or exceeds the component's processing capability, and is typically sustained for only brief periods of time, because excess passenger demand usually produces signifi- cant passenger delays and crowding that disrupt operations. As an example of what happens when a component operates at maximum throughput, consider the case of a single ticket counter staffed by a single agent. Passengers arrive at the ticket counter to check baggage, purchase a ticket, receive boarding passes, or simply ask a question. If there is a queue at the counter, a new arrival must await his or her turn before moving onward through the airport toward his departing flight. If everyone using the ticket counter requires exactly the same service and if the agent at the counter maintains consistent performance, then each person might be served in exactly the same amount of time—say 5 min per person. The number of passengers who could be served by this agent would then be 12 per hour. This rate is the ticket counter's maximum throughput. If this rate continued throughout a 24-hr day, the ticket counter could in theory serve up to 288 passengers per day (Figure 2-3). 300 0 UI 240 Cr U) UI (2 cn z UI 180 a. U- 0 UI a 120 UI > I- a o 60 1 4 8 12 16 20 24 TIME (HOURS) FIGURE 2-3 Maximum throughput for an example ticket counter.

BASIC DEFINITIONS 25 Service Level The quality and conditions of service of a functional component or group of functional components as experienced by passengers constitute the service level. Factors such as waiting time, processing time, walking time, crowding, and availability of passenger amenities for comfort and convenience are measures of the service level of components. Many of these factors are interrelated, and there may be others of importance at a particular airport. There are a variety of ways in which some of these factors may be measured, whereas other factors may be difficult to quantify. An analyst may choose any number of specific measures for capacity assessment. As long as passengers arrive at the example ticket counter, manned by a single agent, at a rate no greater than the rate at which they can be served (in this case, one passenger arrival every 5 mm), there will be no queue and no waiting for service. If a passenger arrives at the counter before the previous arrival has been served and has departed, this new passenger will have to wait. As more passengers arrive, a queue may begin to grow. When the rate of arrivals drops, the agent at the counter may begin to catch up, and the length of the queue will begin to decline (Figure 24).2 In this case, after 2 hr, 24 passengers had arrived, 5 were in queue, and 19 had been served. The length of time that each person will have to wait for service is directly related to the length of the queue when he arrives. Because the arrival of passengers at the counter is random, service level fluctuates frequently. As a practical matter, service volume is measured as an average or expected maximum over a period of 15 min, 1 hr, or sometimes several hours. In theory if the rate of arrivals stays higher than the service rate, the queue can keep growing and service level can keep declining. However, in a real situation, passengers would begin to miss their flights and the terminal lobby would be too full to hold more people: Even before conditions become that severe, passengers would complain. Typically an airline or airport operator will decide that there is some level of delay or queue length that is the maximum acceptable. As an example, the airline using the single ticket counter might decide that maximum passenger processing time (including time waiting in line) should be 15 mm. Delays exceeding that will trigger assignment of a second ticket agent to work at the counter (Figure 2-5). This maximum expected wait time of 15 min is an example of a service- level target, which is the minimum tolerable service level set for a functional component or group of components in a specific analysis period. The analysis period used may be a peak hour, a particular holiday season, or some other period of time during which capacity problems may develop. Airport opera- tors may set service-level targets to guide decision making affecting the development, operation, or management of the airport landside.

26 MEASURING AIRPORT LANDSIDE CAPACiTY Service—Level Indicators Queue Status A = no queue B = queue building C = queue clearing T = expected wait time in queue L = expected queue length A B / 20 Passengers Arrived 10 r Slope a service rate Passengers Served 1 2 TIME (HOURS) FIGURE 2-4 Service-level changes at the example ticket counter. Each airport, and each component of an airport's landside, has unique operating characteristics and demands placed on it. It is thus extremely difficult to suggest service-level targets that can be used at all airports under all conditions. However, experience can suggest how to define service levels and targets, and with further research and development it may be possible to

BASIC DEFINITIONS 27 Queue Exists Queue Exists U) cr II 20 Passengers Arrived (flW I 4 > Wait Exceeds IS mm uUJ 00) I I Passengers Served I .Vmt• / w / z 10 / Lu Cr I / j_4 / / Queue Wait Time TIME (HOURS) FIGURE 2-5 Definition of unacceptably low service level for the example ticket counter (dotted area indicates time during which service level not met unless second ticket agent assigned to work). adopt generally useful guidelines for such definitions. Examples of how such service-level targets have been defined in Canada and Europe are included in Part II of this report, and research to provide a basis for adopting common guidelines in the United States is recommended in Chapter 5. In a study made for this report, an attempt was made to develop service- level targets for U.S. airports. However, the study committee found that available data on service levels are inadequate for development of defensible and valid targets. Nevertheless, the framework devised by the committee for developing targets is a useful starting point; it is presented in Appendix A. Service-level targets have financial and political implications. Individual communities may wish to maintain particular service levels at their airport to attract new business or simply as a matter of community goals. Maintaining targets may require development of new facilities or new operating and management strategies.

28 MEASURING AIRPORT LANDSIDE CAPACiTY Service Volume Service volume—the principal measure of capacity—is the number of pas- sengers that can be accommodated by a functional component or group of components at a given service level given the demand placed on that compo- nent. For components where passenger processing takes place, such as the ticket counter or security screening, service volume may be measured as a rate (passengers per unit of time). For components where passengers wait or stand in queues, service volume may be measured as the number of passengers accommodated at any given time. For components that involve both passenger waiting and processing, both measures may be appropriate. The demand on the example ticket counter—the number of passengers who arrive in a given period of time—fluctuates from time to time, as was shown in Figure 24. Arrival rates tend to increase as the time of a flight departure approaches. If there is only one flight, the rate of arrivals will decline close to the scheduled departure time (and should of course drop to zero once the flight has departed). The service volume that can be accommodated at the example ticket counter manned by a single agent is only 19 passengers over the 2-hr period (Figures 2-4 and 2-5), even if 24 passengers arrive and a waiting time in excess of 15 min is acceptable. If a service-level target is set—for example, that the expected wait at the ticket counter should never exceed 15 mm—then even fewer passengers can be served. If the average service rate is s min per person, setting this target means that if there are more than three passengers at the counter, the service-level target is not met (Figures 2-5 and 2-6). To maintain this target service level, airlines might assign additional agents to counter positions when passenger arrivals are expected to be greatest. Be- cause demand characteristics at any component are seldom exactly matched to the component's service rate, over longer periods of time achievable service volume is generally less than maximum throughput (Figure 2-7).3 Airlines adjust to the patterns of demand by assigning additional personnel and by allowing service levels to decline during busy periods. The fixed physical facilities of the airport are often designed to allow for some variation in demand and growth in traffic. Capacity problems can arise from the way in which facilities are operated as well as from a shortage of these facilities. CAPACITY ANALYSIS PERIOD Capacity becomes a problem during those periods when demand is high. Although these periods may be limited at some airports, at others demand is high during much of the day. Capacity assessment must estimate service

BASIC DEFINITIONS 29 CI) I cc Passengers Arrived Ui 0 I and In Queue 20j- (n U_w I 0( I wz I / /Passen:rvsery c4 I ed w 10 1 z r > cc I / Passengers not Accommodated at Target Service Level 0 I TIME (HOURS) FIGURE 2-6 Influence of service-level targets on service volume at the example ticket counter. volume and service level for a landside functional component or group of components during a specific analysis period. The appropriate analysis period depends on the type of component or components and the characteristics of the demand placed on that component. For some components the appropriate analysis period may be an hour or less, for others a series of hours, and for some perhaps as long as a day. Usually the analysis period is selected to coincide with a period of concentrated demand, for example, a peak hour or busy hour. Service volumes may be set for longer periods of time, such as a month or a full year, but the study committee believes that capacity cannot be meaningfully assessed over these longer periods. The busiest times at many U.S. airports are the few days immediately before Thanksgiving and Christmas. Airports serving southern beach resort areas will often have more traffic in the winter months. Summer vacations make August a busy month at many airports. Regardless of the time of year and the flight schedule, passenger demand varies from month to month, from day to day in a week, and during a day in a pattern that is relatively predictable from historic records. The analysis period should reflect realistic recurring conditions and not the worst possible case at an airport. It is uneconomical to try to maintain the highest service levels under all demand conditions.

30 MEASURING AIRPORT LANDSIDE CAPACiTY Target Service Level 3 -J Iii 0> zuJ Target Service Level 2 ------------- 'U> 'U U) Target Service Level I Maximum Throughput Service Volume ¶ - _ / / iNCREASING Service Volume 2 PASSENGER DEMAND Service Volume Determlned by service-level Indicators (wail time, service time, crowding) FIGURE 2-7 Schematic relationship among sevice level, service volume, and maximum throughput. In some European countries, the Standard Busy Rate (SBR) is defined as the anticipated level of demand during a busy hour (1). For example, at Amsterdam's Schipol Airport, demand levels during the 20th busiest hour are used, and in France, Aeroports de Paris uses the 40th busiest hour. (There are 8,760 hr in a 365-day year.) The British Airports Authority defines the Busy Hour Rate demand as the level such that 5 percent of annual passenger volume occurs at higher hourly rates (see Figure 2-8). The International Air Transport Association (IATA) recommends that high-traffic holiday periods be excluded in selecting busy periods (2). Transport Canada now conducts capacity evaluations by using the 90th- percentile hour, defined similarly to the British Busy Hour Rate, but considers the variations in demand within this peak period (3). Passenger loads that are maintained or exceeded for a 15-rn in period during this peak hour are the basis for assessing service level. FAA guidance material uses the peak hour of an average day of the peak month as a basis for planning anddesign (4), which is common practice in the United States. If, for example, August is the month

BASIC DEFINITIONS 31 uJ J 201h Busy Hour Volume - (a) Definition In Terms of -i i -- 40th Busyreol SPeflc Busy HourpecifiC Busy —J II uj II II 2,000 4,000 6,000 8,000 10,000 2040 NUMBER OF HOURS VOLUME EXCEEDED D > 5 percent Busy Rate cn 0. cc X/A - 5 percent Total Annual Volume )b) Definition In Terms of Total Passenger Volumes 0 2,000 4,000 6,000 8,000 10,000 NUMBER OF HOURS VOLUME EXCEEDED FIGURE 2-8 Alternative definitions of busy hours (2). with highest passenger traffic, the average-day volume is approximately 3.2 percent (1/31) of the monthly traffic. In the absence of other information, the peak-month average-day peak-hour approach is roughly equivalent to the 20th to 40th busiest hour of traffic, and should give a demand level likely to be exceeded on less than 4 percent of the days in a year (14 to 15 days). Changing airline route structure, schedules, and aircraft sizes may place peak demand on particular components of the landside at times that differ from one component to another and from those for the airport as a whole. The midday hour when many flights converge at an airline hub-and-spoke center may represent the peak demand for gates and gate lounge areas, whereas at the

32 MEASURING AIRPORT LANDSIDE CAPACITY same airport, demand at ticket counters and the terminal curb may peak during the 11 2 to 2 hr before the scheduled evening departures of several widebody aircraft. Similarly, different airlines may experience peaks at the same airport at different times of day. Because components are linked together, capacity assessment may have to consider more than one analysis period. NOTES Terms in italics are defined in subsequent paragraphs of this chapter. In Figure 2-4, people begin to arrive faster than they can be served and a queue forms. During the second half of the second hour, the 20th through the 24th passengers are in the queue at the end of the hour. Landside components can operate at maximum throughput for longer periods only if a high demand level is maintained. Under such conditions service levels can always be expected to be poor. In theory the expected passenger delay at the example ticket counter can become infinitely large. As a practical matter, disrup- tions due to weather conditions and the crush of holiday travel illustrate the extremely low service levels that result when landside facilities are forced to operate at a capacity close to maximum throughput for long periods of time. Glossary of Terms. British Airports Authority, London, Nov. 1981. Airport Terminal Reference Manual, 6th ed. International Air Transport Associa- tion, Montreal, Quebec, Canada, Sept. 1978. Air Terminal Processing Capacity Evaluation. Report TP5120E. Airport Services Branch, Transport Canada, Ottawa, Ontario, Canada, Jan. 1986. Planning and Design Consi derations for Airport Terminal Building Development. Advisory Circular 150/5360-7. Federal Aviation Administration, U.S. Department of Transportation, Oct. 5, 1976.

3 Landside Capacity Assessment Process Landside capacity assessment may address a variety of facilities management, planning, and design problems, as shown by the following examples: An airport is experiencing passenger crowding and delay in a terminal building. The airport operator could use landside capacity assessment to identify problems and to judge whether these problems are a result of airline or airport management practices, passenger behavior and demand characteris- tics, or inadequate facilities. An airline wishes to introduce new service at an airport. The airport operator is concerned that this new service cannot be accommodated by existing facilities without severely reducing landside service levels. Landside capacity assessment could be performed to estimate the effect of increased passenger loads on existing facilities and services. The community served by an airport is concerned that traffic at the airport will soon grow to exceed the intended capacity of the facility and that service levels will sharply deteriorate. Landside capacity assessment results could be used to provide an objective basis for discussions among government authorities and the community to resolve the conflict. An airport intends to add airside facilities. Landside capacity assessment could determine whether additional landside facilities would be required to match the forecast increases in traffic served by the added airside facilities. The airlines and the federal government are contemplating new air traffic control procedures that would raise the maximum number of flights that can be handled during one hour by a given airport. Landside capacity assessment 33

34 MEASURING AIRPORT LANDSIDE CAPACITY could be used by the airport operator to estimate the implications for landside operations of such expanded airside capacity. The federal government must allocate limited funds to achieve the maximum benefit for the air transport system as a whole. Landside capacity assessment could identify airports where federal funds could improve system operations by easing landside constraints. In each case, landside capacity assessment, a technical procedure, would yield information useful for decision makers. Although landside capacity assessment is only one element among a variety of factors that influence decisions regarding funding and construction of new airport facilities, nev- ertheless landside capacity assessment may help decision makers understand what the problems are at a particular airport and what potential solutions are available. Figure 3-1 shows the recommended landside capacity assessment process. Analysts and decision makers can use this process to determine the relation- ship between service levels and passenger service volumes at a particular airport subject to the demand and operating policies of that airport. This analytical process is the core of landside capacity assessment. Interaction and feedback among particular steps in the process are critical. Service levels and capacity measures can be assessed only with reference to one another. Performance of any single landside component depends on performance of other components with which that component interacts. Ser- vice-level targets set for current conditions may have long-term implications for the airport's future development. Only by successively considering how individual components perform and how they interact with one another and with demand to influence service levels can a meaningful estimate of landside capacity be derived. In this chapter the 11 steps in the landside capacity assessment process are described. Examples of application of the process to the critical landside components are given in Part II. STEP 1: IDENTIFY GOALS AND OBJECTIVES OF THE ASSESSMENT OR PROBLEM TO BE SOLVED The reasons for conducting a capacity assessment of all or part of an airport's landside will determine the appropriate scope and level of detail of the analysis. Management and planning problems such as those described earlier or a more general desire to plan for future facilities utilization may lead to application of the landside capacity assessment process. Explicitly identifying

LAND SIDE CAPACITY ASSESSMENT PROCFSS 35 1. Identify Goals and Objectives of the Assessment or Problem To Be Solved I 2. SpecIfy 1_andside Components for Assessment 3. DescrIbe Each Component 4. DescrIbe How Components Relate (functionally and physically) S. Collect Data on Demand 6. Collect Data on Community Characteristics and Operating Factors Factors 7. EstImate Component and 8. Estimate Current and Total Service Levels Maximum Service Volumes INTERMEDIATE DECISION: IS LANDSIDE CAPACITY ADEOUATE? 1 9. ExamIne Trade-Offs in Service Among Components I 10. identify Short-Term Measures To improve Capacity I 11. Review Long-Term Planning and Management ____________ implications Long-Range Planning FIGURE 3-1 Landside capacity assessment, management, and planning process. the problem or the planning and design goals that motivate the capacity assessment helps to focus the thinking of those who will use the results for making decisions. As a part of the identification of goals for which an assessment is being made, proposed service-level targets may be set if the various operating characteristics of the components involved are understood. These targets, to be used in determining capacities of landside components in Steps 7 and 8, may be revised or set later when landside capacity estimates are compared

36 MEASURING AIRPORT LANDSIDE CAPACiTY with the projected costs and related consequences of actions to relieve prob- lems. In each case the interests of all parties concerned with airport capacity should be considered by airport operators. Setting service-level targets is a key part of the assessment process and one in which all parties have an important stake. STEP 2: SPECIFY LANDSIDE COMPONENTS FOR ASSESSMENT Assessment should generally focus on components that are experiencing service problems. Although it may often be unnecessary to involve all parts of the airport's landside in an assessment, components linked with the problem components should be included. Normally an assessment is made only for components critical to landside capacity, although in some cases general passenger amenities may be involved. However, before major capital invest- ments are made to solve the capacity problem of one component or group of components, other components should be reviewed to determine what may be the next limiting component. STEP 3: DESCRIBE EACH COMPONENT The analyst should describe the facilities and usual operating practices for each component specified in Step 2. In Part II of this report the specific information typically required to assess key landside components is indicated. The characterization should focus on aspects of each component that will remain constant as demand varies, which are addressed in Step 5. If data are unavailable for the airport being studied, analysts may adapt data such as average service rates from comparable airports. Adapting data may expedite the assessment and reduce its cost but will reduce reliability of the results as well. STEP 4: DESCRIBE HOW COMPONENTS RELATE The analyst should identify the routes passengers can take through the land- side components under study. Is there only one way to pass through the system or are there alternative paths? Are components linked in series or in parallel? One useful way to show linkage is with a simple diagram of the type shown in Figure 3-2. Components are shown as nodes and passenger paths as the

LAND SIDE CAPACITY ASSESSMENT PROCESS 37 From Curb ( Terminal t\ 15 percent Passenger Security percent reenin 100 Entrance ) / Express Check-in \ \Counter °? "- Regular Ticket Counter Note: Percentages represent traction of passengers Choosing each path dating anatysis period. FIGURE 3-2 Example passenger flow link-and-node diagram for ticket counters. links between nodes. A detailed analysis of a large airport may include hundreds of nodes and links. Linkages may be in series or in parallel. In series linkage passengers move from one component to another in sequence, and in parallel linkage they choose one of several paths through the landside. For example, in Figure 3-2 the terminal entrance and passenger security screening are linked in series and the express check-in and regular ticket counters are linked in parallel. The number or percentage of passengers using each link in the diagram can be determined by direct observation or estimated from demand characteristics. Figure 3-3 shows another example of how passenger and baggage flows may be charted (1). The route of any single passenger through the landside is a set of functional components all linked in series. Different passengers with different demand characteristics may use different parallel routes. Airlines often establish com- ponents in parallel to tailor their services to major passenger demand groups, such as frequent business travelers and vacationers. STEP 5: COLLECT DATA ON DEMAND CHARACTERISTICS AND OPERATING FACTORS The analyst should assemble specific data to describe demand and related operating characteristics for each component of interest. Direct observations at the airport, statistics gathered by the airlines, and analogies from other airports are principal sources of data.

Not Ticketed Terminal Entrance Seat Assignment Baggage Check-in Ticketing Security I I Gate I 1 No Airslde Preticketed Curbsldel t 1 I Baggage Check-In Baggage Seat Assignment Check-In I Seat I Assignm I

S S -, —' Baggage Air ------------------------- , Transport \ Network Ren System tal / Highway Car ;:rs - Regional Aiofl E~> and Baggage Passenger Lounges s:n --ini and e"f"( thusInes Cars Taxicabs Buses Apron Parking Area Remote Shuffle SeMcing, Rail or Rapid Aircraft Transit (if provided> FIGURE 3-3 Typical component linkage diagram (1).

40 MEASURING AIRPORT LANDSIDE CAPACITY The following types of data are typically required: Number and fraction of passengers using each component under review by time of day, with weekly and monthly variations; Staffing levels, with employee or agent efficiency (at ticket counters, customs and immigration, etc.) related to per-passenger service times; Indicators of overall demand loads, such as number of checked bags per passenger, number of pieces of carry-on luggage per passenger, and number of visitors accompanying each passenger; Flight destinations and origins, daily flight schedule, aircraft and likely load factors, gate occupancy times; and Fraction of passengers originating or terminating their trips at this air- port, distribution of passengers among airlines (transfer and origin or destina- tion), typical times of arrival at the airport versus scheduled flight departure time, and fraction of passengers choosing alternative modes for ground transport. Architectural or engineering drawings of the terminal building areas of interest are useful to illustrate typical paths of passengers using landside facilities and where capacity problems are occurring. Ramp charts showing representations of daily gate use are also useful to show the daily flight schedule that serves passenger demand on all landside components. Not all of these data are required for all capacity assessments. Analysts normally gather only data relevant to the purpose for which the assessment is being conducted. If the assessment deals with future conditions at the airport, the analyst must forecast demand and component operating characteristics. STEP 6: COLLECT DATA ON COMMUNITY FACTORS The, analyst should assemble data describing any factors in the local com- munity that influence the demand patterns or operating conditions of the landside components being assessed. These factors typically reflect the current or forthcoming policies, allocations of resources, or sensitivities of local governments and the communities they represent. Thus the data describing the factors may include both specific quantitative and broader qualitative informa- tion. Although many of these factors and their influence are difficult to assess, particularly in quantitative terms, their impact can be significant and they should not be neglected in conducting landside capacity assessments. Community factors influencing capacity at U.S. airports include the follow- ing examples:

LANDSIDE CAPACITY ASSESSMEWT PROCESS 41 Restrictions on aircraft operations, often associated with noise abatement programs, that may influence the type of aircraft using an airport or the hours of their operation. Although data describing existing demand patterns gathered as part of Step 5 will reflect any current restrictions, forecasts of future demand should consider present or possible future policies that may be implemented on behalf of the community. The degree of community support an airport can expect to receive as it competes with other public agencies for funds or other public resources needed to permit improvements or expansion. These resources could include funding (such as tax revenues), utilities (such as water or electricity), or access facilities (such as highways). Land availability, or an airport operator's expected ability to acquire additional property to permit expansion of apron areas, terminal buildings, access roads, parking, or other landside components. Restrictions on vehicular circulation or access routes intended to mini- mize adverse impacts on adjacent neighborhoods or government-sponsored efforts to promote more efficient methods of travel to airports such as buses, coaches, shared-ride taxis, or shuttles. Marketing or promotions, often undertaken by local community business and industry groups, designed to attract new or increased air services or airline passengers, or both (local and distant markets). Absolute limits on number of aircraft operations or airline passenger volumes at an airport established by a local government or by an airport operator in response to community desires. In contrast to flight restrictions for noise abatement, these limits may be designed to divert demand to an under- utilized facility (e.g., an alternative airport in the region) through landing fees or to suppress demand and thereby reduce the airport's perceived adverse impact on the community. Further discussion of this step appears in Chapter 4. This data collection and analysis need not duplicate those conducted as part of Step 5, studies con- ducted for the FAA noise control program ["Part 150" planning (14 CFR 150, 1981)], or other local planning activities. STEP 7: ESTIMATE COMPONENT AND TOTAL SERVICE LEVELS Step 7 must proceed in parallel with Step 8. With information gathered in Steps 3-6, direct observation or mathematical modeling may be used to assess service levels encountered by passengers during the analysis period within each component being assessed. To deal with future conditions, projections of demand and operating characteristics, described in Step 5, must be available.

42 MEASURING AIRPORT LANDSIDE CAPACiTY Mathematical modeling is not necessarily complex. Analogies from other airports, statistical correlations, and simple queueing calculations may be adequate. A variety of models that may be useful are surveyed in Appendix B. The reasons for which capacity assessment is being performed determine the level of. detail and complexity of analysis, and thus whether mathematical modeling is warranted. If service-level targets have been set and are being met or exceeded for all components being assessed, the system service level is adequate. However, if service levels for one or more components are below target, the overall system service-level target may still be met or exceeded. For example, delays at the entry to one parking lot may affect a relatively small fraction of passengers and may add a relatively small amount of time to the total landside journey of a departing passenger. The average service level for all originating passengers may be acceptable, even though the parking lot's service level is not. If some components do not meet targets and no system target has been set, the system service level cannot be defined without consideration of passenger service volumes in Step 8. Feedback and iterative calculations of service volumes and service levels are typically required. If service levels are lower than desired, higher targets might be set. However, setting service-level targets higher than current service levels im- plies a willingness to redirect or restrict passenger flows, to change operating procedures, or to invest in new facilities. STEP 8: ESTIMATE CURRENT AND MAXIMUM SERVICE VOLUMES Passenger service volumes associated with service levels estimated in Step 7 should be determined. If all service levels of components of interest are exactly on target, these volumes represent the achievable service volume of the, system operating as described and subject to the assumed patterns of demand. If some components are operating at better than target service levels, mathematical models may be used to estimate the additional passenger de- mand that could be accommodated if service levels were allowed to decline to targets. If service levels are found in Step 7 to be below target, mathematical models may be used to estimate the reductions in passenger volumes that would be needed for service levels to improve to meet targets. INTERMEDIATE DECISION: IS LANDSIDE CAPACITY ADEQUATE? In many cases, completion of Steps 2-8 will produce enough information about current and target service levels and associated current and achievable

LAND SIDE CAPACITY ASSESSMENT PROCESS 43 d passenger service volumes to answer the questions raised in Step 1. A judg- ment whether landside capacity is adequate to meet demand can be made for the purposes that motivated the assessment. If it appears that landside capacity is insufficient—if service levels are below target—additional analyses should be undertaken to explore ways of making better use of the landside components in question and to identify what changes in system configuration, operations, or demand may be required to increase achievable service volumes or improve service levels. These addi- tional analyses are not an immediate part of the capacity assessment, but rather are part of the management and planning context within which landside capacity assessment occurs. STEP 9: EXAMINE TRADE-OFFS IN SERVICE AMONG COMPONENTS If some components in the assessment exhibit service levels above target and others are below target, the performance of the poorer-performing compo- nents might be improved by shifting patterns of demand or operating charac- teristics of other components, recognizing that some declines in service levels in the higher-performing compbnents may be acceptable to achieve smoother operations overall. Such trade-offs among components obtain more generally satisfactory service levels through a closer match between demand and capac- ity, which means higher efficiency and better overall. system performance. Overall system demand, a result of flight schedules and passengers' choices about time of arrival at and departure from the airport, is at this step of the assessment considered fixed. STEP 10: IDENTIFY SHORT-TERM MEASURES TO IMPROVE CAPACITY Short-term actions to improve capacity generally preclude any major con- struction or substantial restructuring of leases and financial arrangements. Such actions might include encouraging airlines to adjust flight schedules or staffing policies or changing gate operation strategy if airlines involved in the assessment process agree on the need. Otherwise, airport operators may be limited to actions that involve only those areas of the landside fully under their control. Sometimes short-term actions may offer temporary relief for a problem while more permanent solutions are being devised. For example, if existing

44 MEASURING AIRPORT LANDSIDE CAPACiTY baggage-claim and gate lounge areas are adequate to serve the growth in passenger loads, remote parking of aircraft and use of transporter vehicles may increase effective gate capacity while additional gates are being con- structed to accommodate proposed new airline services. STEP 11: REVIEW LONG-TERM PLANNING AND MANAGEMENT IMPLICATIONS Landside capacity assessment is not a substitute for long-range facilities planning, although it is an important element in developing or revising an airport master plan. The long-term implications of the analyses conducted in Steps 1-10 may include development of new facilities, institution of major changes in operating practices at existing facilities, and shifts in policy regarding growth of airport activity. These implications must be considered within the context of the goals and objectives stated in Step 1. If immediate landside capacity problems motivate the assessment, the airport operator and airlines using the airport may be able, working together, to devise solutions to the problems. If future changes in traffic threaten to raise problems where they do not now exist, early recognition of the financial and community impacts of solutions to these problems will aid those making difficult decisions about airline fleet management and investments in airport and community facilities. In both cases, feedback within the landside capacity assessment process may lead to reconsideration of basic goals and expecta- tions about landside service levels and eventual agreement among all parties that problems are well understood and that the proposed solutions are reasonable. REFERENCE 1. Airport Ground Transportation: Problems and Solutions. U.S. Department of Trans- portation. Feb. 1981.

4 Community Factors Airports, often among the largest (in terms of total land area) and most important single public facilities located in a metropolitan area, contribute to the well being of the communities they serve in a variety of ways. They provide access to national and international air transportation systems and employ substantial numbers of people directly and many others indirectly in industries that depend on good air transport service. Land values may rise because of the airport's presence, and businesses may be attracted to invest in regions with good air service. But at the same time, airports can be of concern to their neighbors because of aircraft noise and road traffic congestion. In addition, an airport may occasionally compete for municipal services that are in short supply within the community, such as water and power supplies and police and fire protection. In broad terms, the airport-related community has three principal segments: airport users (travelers and shippers and other businesses that depend on air transport and their employees), residential and commercial neighbors of the airport, and local and state government. This airport-related community to- gether with airport operators, airlines, and the FAA determine how an airport will develop and operate. An airport's long-term ability to match growth in air travel demand with adequate capacity to serve that demand often depends on the perceived balance between the benefits and the problems associated with the airport. When this balance is very favorable, the airport operator typically has relative freedom in directing the airport's operation and growth. However, if a com- munity or key parts of the community perceive that the problems raised by an airport outweigh the benefits, the community may seek to restrict the airport's ability to expand or to operate efficiently. The restrictions imposed may curtail construction of new airport facilities, control aircraft operations, or limit land 45

46 MEASURING AIRPORT LANDSIDE CAPACiTY and municipal services available to the airport. Through such restrictions the community may influence airport capacity. When an airport's contribution to the community is viewed favorably, landside capacity may be expanded to encourage future growth of demand. San Francisco International, for example, has built new terminal buildings and access roads, and the Port Authority of New York and New Jersey is preparing to renovate and expand terminal buildings at John F. Kennedy International to regain service levels consistent with that airport's role as the principal gate- way for foreign visitors to the United States. In such cases, inconvenience during construction and longer-term trends of traffic growth are tolerated by the community. In an assessment of any airport's landside capacity, the potential impact of community factors must be recognized, particularly when these factors may constrain the airport's ability to meet demand. How restrictions may come about, the types of restrictions that may be imposed, and how to assess the impact of such restrictions on capacity are reviewed here. SOURCES OF CONCERN Most of the nation's airports have been in operation for many years at locations established when the metropolitan areas served were considerably smaller than they are today. Few of these airports were expected to meet the needs of today's jet aircraft or the wide range of passenger and cargo services now offered by the industry. Some of these airports—Logan Airport in Boston; National Airport in Washington, D.C.; and Lindburgh Field in San Diego—are close to downtown city centers. New high-rise buildings and changing population characteristics have led to increasing conflicts between these airports and their respective communities. Even newer airports such as Dallas-Ft. Worth International and Dulles International near Washington, D.C., built on large sites in seemingly remote locations, now experience the problems of increasing suburban growth in areas exposed to aircraft noise and increasing highway congestion. Al- though the benefits of improved air transportation access and efficiency are distributed generally throughout the region an airport serves, the associated problems are largely concentrated in the immediate vicinity of the airport (1). Noise Aircraft noise is the most frequent cause of organized community efforts to influence airport operations. As a consequence, aircraft-related noise issues

COMMUNITY FACTORS 47 provide good illusirations of how community factors may affect airport capacity. For example, the Airport Access Task Force reported to Congress in 1983 that at least 44 of the nation's 50 busiest airports (in terms of annual number of passenger enpianements) had imposed operational restrictions to control community noise exposure (2). Although cases involving damage claims based on aircraft noise appeared in the courts much earlier, national policy toward aircraft noise began to take shape in 1968 when Congress amended the Federal Aviation Act of 1958, requiring that national regulations for aircraft noise control be established and giving FAA the responsibility for developing these regulations (3). With passage of the National Environmental Policy Act of 1969 (NEPA), Congress further required that assessments be made of significant environmental im- pacts likely to result from major federal government actions such as participa- tion in airport development. The 1968 aviation act amendment addressed control of noise at its source, that is, at the aircraft, but NEPA emphasized a need to try to ensure compatibility of airports and their surroundings. Subse- quent laws, regulations, and guidelines at federal, state, and local levels have built on these two bases—control at the source and compatibility with the surroundings. Since 1969 FAA standards have set allowable noise emissions for new aircraft operated in the United States. As newer aircraft meeting the most recent and stringent of these standards [termed "Stage 3" aircraft under the regulations (4)] enter the commercial airline fleet, and as older aircraft are retired or retrofitted to meet the more stringent standards, aircraft noise around airports is expected to decline significantly. However, retirement of all noisier aircraft may not be complete until after the year 2005 (2, 5). In the meantime, some communities have established regulations restricting operations by aircraft not meeting Stage 3 standards. In Boston and at Washington's National Airport, for example, these restrictions prohibit night- time operations by other than Stage 3 aircraft.' Because airlines and cargo shippers serving such airports do not yet operate aircraft that are both appro- priate to the types of service desired and compatible with these noise stan- dards, their ability to schedule flights may be severely curtailed by this type of restriction. Other types of restrictions, such as establishing preferential run- way use and flight paths, may have a less drastic impact on airline operations but still affect capacity by reducing the number of operations that can be handled at the airport. Some communities have tried to restrict all types of jet aircraft (2). Federally sponsored programs of local land use planning encourage com- patibility between airports and their surroundings. [The current program is termed "Part 150" planning because of its basis in this part of the Federal Aviation Regulations (6).] The thrust of these programs is toward controlling

48 MEASURING AIRPORT LANDSIDE CAPACITY types and intensity of land use in areas likely to be exposed to aircraft noise, but enforcement of controls is a local concern and strictly voluntary. Even when local government officials involved in the planning are committed to land use control, later administrations may find the advantages of new de- velopment more attractive than the possibility of avoiding yet-to-be-realized noise conflicts. Residents in such newly developed areas then may become leading proponents of restrictions on aircraft operations and airport development. Other Sources Aircraft noise is not the only source of community concern. Others include road congestion and associated environmental degradation, competing de- mands on community resources, and limitations on land availability. Many highways serving airports also serve the general travel needs of the surrounding community. Growth of traffic on these highways leads to conges- tion, and both groups of users suffer. Congestion on major highways may also spill over into local streets, with significant impact on residential neighbor- hoods. This congestion is typically accompanied by automobile air pollution, noise, and perceived safety problems. Restricting air passenger growth is one way to curtail further highway traffic growth. For example, persistent conges- tion on the San Diego Freeway, a principal entry to Los Angeles International, is cited as a reason for imposing a limit on the annual number of passengers at the airport (7, pp. 83-88). In Boston management of Logan Airport is work- ing with highway agencies to find ways to relieve the serious congestion in the two tunnels that are the main link between downtown and the airport. Con- struction of a third tunnel is a high priority for solving the problem. In each case future growth of the airport may be restricted, and although present demands are met, service levels on the highways are very low during busy periods. In communities where basic resources are scarce or where funds to develop infrastructure are limited, an airport may find itself competing with other parts of the community for municipal services. Apparent lack of adequate water supplies threatened to stall plans for Denver's proposed new airport. Lack of power supplies delayed facilities development at Oakland International. Sewage and solid waste disposal are recurring problems at some airports. Cooperation of several political jurisdictions is often required to solve these problems, which may be hampered when voters choose not to endorse bond issues or authorizing referenda. Again, current demands are served, but future expansion may be limited. If demands grow, service levels are likely to decline.

COMMUNITY FACTORS 49 Airports with relatively small land areas typically have limited capacity. For example, National Airport in Washington, D.C., and La Guardia in New York City are able to serve large volumes of traffic because a large fraction of their aircraft operations can occur over water. Although airports such as Hartford's Brainard Field, Detroit's older City Airport, and Cincinnati's Lunken Field are no longer in commercial service, they serve other aviation needs and there is not enough space for new facilities—the airport is surrounded by developed land or physical barriers. At other airports the cost of land or the community's attitude toward airport development has a similar effect. In Carlsbad, Califor- nia, airport expansion is prohibited by local ordinance (8). Such restrictions could influence the ability of these airports to meet future demands for new commercial services. Development of new airports within practical distance of downtown dis- tricts has become virtually impossible in most major metropolitan areas because of suburban growth. The search for a site for a fourth jetport to serve the New York City metropolitan area has been ongoing for three decades (9). This inability to develop new airports may tend to increase the pressures of growth in passenger demand at existing airports. TYPES OF RESTRICTIONS Potential community-imposed restrictions on airport facilities and services fall into two principal categories: Policies that prevent expansion of existing airport facilities or develop- ment of new airports. Such policies may include delays in financing or approving airport expansion plans or related highway projects and local zoning that encourage incompatible development of land near the airport. Regulations that curtail or alter permissible aircraft operations. Such regulations may include limits on number of daily aircraft operations, time of day when operations are permitted, operations by particular types of aircraft, and flight paths. Restrictions in the first category can apply to both airside and landside, affecting taxiways, runways, parking structures, rental-car terminals, cargo handling facilities, or fuel storage tanks. Permit and license requirements may make airport expansions prohibitively expensive, and as shown by the history of the plans for New York City's John F. Kennedy International, environmen- tal impact review procedures can be used to halt particular projects (9). The threat of protracted litigation may also discourage planning for expansion to meet air transport demand.

50 MEASURiNG AIRPORT LANDSIDE CAPACiTY Refusals to provide facilities and implement policies supporting airport growth may permit general urban growth to encroach on the airport and lead to loss of potential capacity. The city of Alexandria, Virginia, surrounding much of the Washington, D.C., National Airport, views the airport as an inappropriate use of valuable—and taxable—urban land and continues to support residential and commercial development within noise-exposed areas. The constituency supporting more stringent airport restrictions thus continues to grow. At Dallas-Ft. Worth International and at Washington's Dulles Interna- tional, new land development around the airports raises the prospect that planned future runways will never be operated at maximum capacity because the airport's neighbors, who were not there when the airports were planned and built, will successfully impose restrictions. Restrictions in the second category, aircraft operations, sometimes take the form of nighttime curfews, perhaps on aircraft not qualifying under the FAA's noise emission standards, or requirements that pilots follow special flight paths and noise abatement procedures on landing and takeoff. At New York City's John F. Kennedy International and Boston's Logan, flight operations are directed to particular runways, which are changed from time to time to avoid exposure of any one area to an excessive share of aircraft noise. In Los Angeles restrictions are established in navigational easements that the airport has had to purchase from individual owners of nearby property. In some cases the threat of litigation may motivate the airport operation to avoid serving certain types of aircraft. Some communities have tried to restrict operations strictly on the basis of inadequacy of landside facilities. In Westchester County, New York, and at Southern California's John Wayne Airport in Orange County, airlines, airport operators, and the community reached out-of-court settlements to limit the number of flights serving these airports. ASSESSING COMMUNITY FACTORS Likely community effects on airport capacity may be assessed in two steps: Estimate the possibility that the community will impose additional re- strictiôns on airport operations and growth, and Estimate the effect of restrictions, both those now in place and those possibly to be imposed in the future, on service volumes. Both steps depend on judgment of social, economic, and political conditions within the community rather than on well-defined analytical models and

COMMUNI7Y FACTORS 51 analysis procedures. Assessment is more a matter of recognizing significant factors than collecting and analyzing detailed data on facilities and operations. Possibility of Restrictions Committee members' experience suggests that several characteristics of the community can increase the possibility that airport restrictions may be imposed: The land area of the airport is small compared with the number of operations or enplanements served. In such cases, airport growth is more likely to create a need for new land and infrastructure. The land area and number of people exposed to aircraft noise are large compared with the number of operations or enplanements served. Data gathered from selected planning studies prepared under FAA noise control programs, summarized in Table 4-1 and Figures 4-1 and 4-2, demonstrate the TABLE 4-1 ACTIVITY LEVELS AND ESTIMATED NOISE EXPOSURE AT SELECTED AIRPORTS Airport Activity Level Annual Airport Operations (thousands) Annual Passenger Enplanerornts (thousands) Airport Grounds (acres) Noise Exposure Land Area Residential (acres) Population Monterey, Calif. 94 212 496 1,024 972 Providence, R.I. 180 496 974 1,920 8984 Islip, N.Y. 231 414 1.350 2,749 NAb Baton Rouge. La. 151 327 1,094 3,430 8,900 Sarasota-Bradenton, Fla. 155 696 1,194 6,010 12,900 Cleveland, Ohio 208 2,932 1,600 9.638 28,730 Portland, Oreg. 204 2,396 3.205 13.696 NA Denver, Cob. 468 13,494 4,651 20,200 34,000' Pittsborgh, Pa. 296 6,699 10,200 26,240 NA Atlanta, Ga. 566 19.610 3,753 31,104 81,188 Seattle-Tacoma, Wash. 212 5,643 2,000 31,232 NA Chicago (O'Hare), 111. 592 20,761 6,795 51,277 285,430 Casper, Wyo. 88 111 4,500 900 NA Jackson. Wyo. 37 62 764 430 25 Scottsdale, Ariz. 162 3 NA 403 NA Groton, Conn. 20 71 434 263 24 Lebanon, N.H. 5 35 390 16 NA 'Area and population where projected noise is above Ldn = 65 dBA. bNA = not available in plan or study reviewed. 'Estimated by staff of Stapleton Airport. Souirca: Pan ISO plans prepared by airport management. Airport Operators Council International survey FAA, National Plan of Integrated Airport Systems. Data are for 1984.

52 MEASURING AIRPORT LANDSIDE CAPACITY 10,000,000 1,000,000 z w w z 100,000 0. z w —J 4 z 10000 z 4 Chicago (OHare) Atlanta • s Denver U 1,000 10 100 1,000 10,000 100,000 LDN 65 EXPOSURE AREA (acres) FIGURE 4-1 Enpianements versus noise-exposed area in 1984 at selected airports (logarithmic scales). past relationship between number of air passengers and noise-exposed neigh- bors at selected airports.2 These indirect relationships may change as new aircraft are introduced and as population shifts within metropolitan areas, but may be useful in judging whether an airport is within the range of typical experience. Airport neighbors include substantial numbers of those likely to be both particularly sensitive to an airport's noise and other adverse impacts and vocal in their opposition to these impacts. In general, wealthy, elderly, and highly educated populations may be more likely to complain and to take political or legal action against the airport. There is a history of adverse community response and resistance to airport activities. Such a history may be evidence of a relative probability of restrictions. The airport operator has failed to encourage cooperation and reasonable recognition in the community of the need for a balance of airport benefits and costs. Such diverse airports as those in Burbank, California, and in Tampa, Florida, have demonstrated how good communications and cooperation with the community can help to resolve community concerns. The presence of any of these characteristics in the community may indicate a relatively greater possibility that service volumes achievable at the airport

COMMUNITI FACTORS 53 10,000,000 Chicago Atlanta • • Denver • 1,000,000 z 1— 4'teveianct Ui Ui z 0. 100,000 F- Sarasota z Ui Ptovidence 8abon Rouge ZO 10,000 F- 4 Jackson G,olov I' 1,000 10 100 1,000 10,000 100,000 LOW 65 EXPOSED POPULATION FIGURE 4-2 Enpianements versus noise-exposed population in 1984 at selected airports (logarithmic scales). may be reduced by community-imposed restrictions. The analyst should then proceed to estimate how large the reduction will be. Effect on Capacity To estimate how much community-imposed restrictions may reduce service volume, the analyst must both assume what form restrictions will take and project what service volumes and service levels will be with and without these restrictions. Assumptions about the form of restrictions may be based on comparisons with other airports in similar community situations. If aircraft noise complaints from neighbors are the most likely source of constraints, experience suggests that restrictions on flight operations are possible. If lack of land use control in the areas around the airport is a problem, restrictions on facilities expansion as well as flight operations may occur. The analyst should state explicitly the restrictions assumed and the rationale for the assumption. Community-imposed restrictions may affect service volumes and service levels in two ways. First, by preventing expansion of facilities to match passenger growth, they can lead to lower service levels as service volumes increase. Second, by curtailing operations they shift demand to different time periods and perhaps restrict total demand, thereby changing service volumes

54 MEASURThIG AIRPORT LANDSIDE CAPACiTY and service levels.3 In either case, the appropriate analysis period will be a full operating day, month, or year. Estimates of decreases in potential service volumes may be based on calculated average unit service volumes achieved at the airport (e.g., passengers per unit time per gate, per unit floor area, per parking space, or per acre of airport land) with and without the restrictions. Comparisons with unit service volume indicators at similar airports may also be useful. The assessment should consider how airlines and the airport may respond to constraints. Airlines may introduce new aircraft rather than cancel scheduled flights or use ticket pricing to shift demand from constrained times to other periods. Faced with declining service levels on access highways, the airport may encourage increased passenger use of transit services, thereby changing both service volumes and service levels. NOTES Restrictions may have no direct impact on landside operations during the busy periods of a day, but nevertheless influence landside capacity by limiting when flights can operate or how many flights can be accommodated at particular times. Ldn, the "day-night" weighted sound pressure level, is a composite measure of cumulative aircraft noise exposure that takes into account the intensity, frequency, and time of day of individual aircraft noise events. Ldn is expressed in decibels (dB). An increase of 3 dB represents a doubling of the perceived severity of noise exposure. A penalty of 10 dB is applied to the noise generated by flights operating during nighttime hours. The FAA has adopted Ldn as the best indicator of noise exposure and uses it as a basis for recommending what types of land use are compatible with anticipated levels of aircraft noise exposure. Areas where the forecast Ldn level is less than 65 dB are presumed to be suitable for any type of land use. Restrictions may have no direct impact on landside operations during the busy periods of a day but nevertheless influence landside capacity. For example, prohibi- tion of nighttime operations in principle will shift passengers into the daytime peak period, raising service volumes and reducing service levels. However, some frac- tion of the demand that might be served by nighttime operations will simply not develop. This effect on capacity is most important at airports serving air cargo and long-distance flight operations for which daytime arrivals are difficult to schedule. REFERENCES Airport Land Use Compatibility Planning. Advisory Circular 150/5050-6. Federal Aviation Administration, U.S. Department of Transportation, Dec. 30, 1977. U.S. Congress. House. Conunittee on Public Works and Transportation. Report of the Airport Access Task Force. H. Report 98-63, 98th Congress, First Session, 1983. Guidance Notebooks for Environ,nenzal Assessment of Airport Development Proj- ects. Report DOT P5600.5. U.S. Department of Transportation, 1978.

COMMUNI7Y FACTORS 55 FAA Noise Control and Compatibility Planning for Airports. Advisory Circular 150/5020-1. Federal Aviation Administration, U.S. Department of Transportation, Aug. 5, 1983. Report to Congress—Alternatives Available to Accelerate Commercial Aircraft Fleet Modernization. Federal Aviation Administration, U.S. Department of Trans- portation, April 11, 1986. FAA Airport Noise Compatibility Planning, Federal Aviation Regulations, Part 150, Federal Aviation Administration, Jan. 1981 [14 CFR 150 (1981)]. M. Kaplan. Planning for Airport Access in Southern California. In Airport Ground Transportation: Problems and Solutions, AGTA/CalTrans Conference Proceedings (R. A. Mundy, ed.), U.S. Department of Transportation, Feb. 1981. Airport System Development. Report OTA-ST1-231. U.S. Congress, Office of Tech- nology Assessment, Aug. 1984. Jamaica Bay and Kennedy Airport; A Multidisciplinary Environmental Study. 2 vols. National Academy of Sciences, Washington, D.C., 1971.

Research Needs The process and guidelines recommended in this study are an important first step toward developing consistent nationwide procedures for assessing airport landside capacity. Substantial research will be needed before this first step will lead to definitive airport landside capacity assessment guidelines and stan- dards. The Transportation Research Board's widely used Highway Capacity Manual, an evolutionary product of more than 30 years and millions of dollars of research, is evidence that no single study can achieve this goal.1 Current quantitative knowledge about landside operations and service lev- els is poorly developed. Reliable comparative statistics on operating condi- tions in airports are scarce. Mathematical models useful in forecasting land- side service conditions are frequentiy proprietary and not available for public use. Airline staffing and space utilization statistics are often not available. What airline passengers may want and be willing to pay for is frequently a matter of conjecture, except perhaps where airline and airport market research has been conducted to shed light on specific problems. The gaps in knowledge can be filled only by a purposeful and coordinated research program. In particular, four major topics warrant immediate attention by the FAA, airlines, and airport operators: 1. Collection of comparable and detailed data on passenger behavior and facilities utilization at a broad cross section of commercial service airports should be undertaken. Such data will give a sound basis for defining service- level measures. The Canadian Airport System Evaluation (CASE), a method and program for such data collection in Canadian airports, is a potentially useful model of what might be done in the United States. The CASE program

RESEARCH NEEDS 57 is one of the data collection activities that has supported Canada's substantial progress in assessing airport landside capacity and maintaining balanced use of airport facilities. Data on aircraft delay because of landside problems should be collected in a format compatible with that for airside delay. Major investment decisions are considered for airside facilities without adequate information on landside consequences. For example, new air traffic control instruments and pro- cedures may permit a significantly greater number of aircraft operations to occur within a peak hour but may substantially increase loads on landside facilities. However, current information provides a very limited basis for weighing the relative costs and benefits of such a change to the air transport system as a whole. There should be continuing refinement and documentation of measures and procedures for landside capacity assessment. The work presented here is necessarily limited by the complexity of the airport landside and by the resources and time available. More meaningful guidelines for landside capac- ity assessment can be developed by building on this base, which will yield substantial benefits of improved resource utilization throughout the aviation industry. Testing and validation of the assessment process presented here are required. Precedents for this process exist, most notably in the work of several consultant organizations and the CASE program. Nevertheless, there is lim- ited experience with landside capacity assessment within the comprehensive context described here. Data collection and setting of service-level targets for an assessment may prove challenging. Test cases at several airports would be useful to demonstrate the value of and problems with this landside capacity assessment process. These pilot assessments would sharpen understanding of community influence as well as airline and passenger demand characteristics in landside capacity measurement. Major barriers to establishing a research program directed toward landside capacity are institutional and financial. No single agency or organization is responsible. Although the FAA maintains a nationwide scope of interest, it has very limited responsibilities for landside development and management. Na- tional professional and trade organizations depend on contributions from their membership for their support and can sponsor few research activities. Few individual airports can afford their own major research programs. In the absence of a single agency with clear responsibility in this area, it is recom- mended that FAA take the research lead, working with airports to develop a specific institutional, management, and financial plan for airport landside research.

58 MEASURING AIRPORT LANDSIDE CAPACiTY NOTE 1. The Highway Capacity Manual may be a useful model of what might be achieved for airport landside capacity assessment, and several of the principles asserted in this report are relate'd to those applied in highway assessment and planning. However, the Manual is directly relevant to only a few elements of the airport landside.

Part II Assessing Capacity and Service Levels of Functional Components Terminal landside service levels and capacity are controlled by the behavior of individual functional components and interactions among components. Most capacity assessments begin with consideration of only a portion of an airport's landside, that is, a few problem components. In such cases the analyst need not apply the process described in Part I to the entire airport. In Chapters 6-16, guidance is presented for applying the landside capacity assessment process to the following potentially critical components: Chapter 6, aircraft parking position and gate; Chapter 7, passenger waiting area; Chapter 8, passenger security screening; Chapter 9, terminal circulation (cor- ridors, stairs, etc.); Chapter 10, ticket counter and baggage check; Chapter 11, terminal curb; Chapter 12, parking area; Chapter 13, ground access; Chapter 14, baggage claim; Chapter 15, customs and immigration, and Chapter 16, connecting passenger transfer. In the concluding chapter, interactions among groups of such components and the landside system as a whole are considered. Each chapter is structured as follows: Description of the component, where its boundaries may normally be drawn for assessment, and the demand and operating factors generally influ- encing that component's service level and capacity. Discussion of the demand patterns that the component must typically accommodate, partiularly the peaking conditions likely to give rise to ser- vice-level and capacity problems. Where relevant data are available, typical examples of demand variations are discussed.

Description of the operating characteristics typical of the component, such as airline staffing practices and processing rate variations, that influence component utilization and effectiveness. Where relevant data are available, typical examples of service conditions and component operating performance are discussed. Review of analysis tools and assessment stdndards found in the literature and current practice to assist in assessing component capacity and levels of service. A selection of available analysis tools is reviewed in Appendix B. Information from the literature that may help in setting service-level targets is cited. An example of the assessment process to demonstrate how data gathered for a particular airport may be used to estimate achievable service volume based on a specific pattern of demand and service-level target. A brief discussion of specific research needs related to capacity assess- ment for the component. These research needs supplement the more general recommendations in Chapter 5.

6 Aircraft Parking Position and Gate When a commercial service aircraft arrives at the airport, it maneuvers from the runway system to the taxiway system and to the ramp area adjacent to the terminal building. This ramp or apron area contains the aircraft parking positions—the designated locations where these aircraft unload and load passengers and baggage and are serviced—and the gates through which passengers pass to board or leave an aircraft. In being routed to its assigned parking position and gate, the aircraft may encounter other taxiing aircraft and ground traffic, may have to wait for its assigned gate to be vacated by another departing flight, and in congested or geometrically constrained apron areas may have to be towed into the parking position. If the parking position is remote from the terminal building, passengers may have to walk some dis- tance on the ramp to reach the terminal or may be carried by transporter vehicles. After an aircraft has been serviced and loaded and has departed, ground crews may need time to prepare the position before the next arrival can be handled. The various activities of arrival and departure combine with facilities characteristics to determine the number of flights that the gate complex can accommodate in a period of time and the delays to which passengers and aircraft may be exposed. In addition, gate operations influence passenger demand characteristics and thus service levels throughout the airport landside. DESCRIPTION Aircraft parking positions are designed to accommodate the particular dimen- sions of specific types of aircraft and may thus be unavailable to other aircraft with significantly different dimensions. If the apron area is not large enough to 61

62 MEASURING AIRPORT LANDSIDE CAPACITY allow safe maneuvering of aircraft under established FAA, airline, and airport standards, capacity may be constrained. If a parking position is not available at the terminal building, the aircraft may be accommodated at a hardstand, an apron parking position made relatively permanent by installation of ground power and sometimes fueling facilities. During periods of very high demand, commercial service aircraft may have to be parked and serviced at remote parking positions. Although airlines typically lease gates, they may own the passenger loading bridge and aircraft service equipment installed at the gates. These aircraft gates may be operated on an exclusive-use basis, under which a single airline has complete use and control of a gate. Lease agreements may give the airport operator the right to renegotiate for underutilized gates, but day-to-day assign- ment of aircraft to gates under the exclusive-use arrangement is usually an airline decision. Because of differences in schedules, a flight may arrive to find all of its company's gates occupied and have to wait for gate access, even though the nearby gates of another airline with a different pattern of arrivals are empty. Some airports provide preferential and joint-use gate strategies. Under the preferential gate use strategy, a gate is leased to a particular airline but the airport operator retains the right to assign it to other airlines when it is not in use by the leasing airline. Under the joint-use gate strategy, gates are usually rented to more than one airline. This strategy is similar to exclusive use in that the airport operator is not typically involved in day-to-day gate assignment decisions. Except where very large numbers of daily flight operations occur, gates operated under preferential and joint-use strategies normally serve more flights than gates operated under an exclusive gate use strategy. Some airports normally operate gates on a common-use basis, in which the assignment of aircraft to gates is entirely an airport operator's decision. This type of operation is common at small commercial service airports. Beyond the required basic physical compatibility between each airline's aircraft fleet and an airport's gates, hardstands, and remote temporary apron parking locations, the principal measure of service level for aircraft parking positions and gates is the time an aircraft and its passengers may be delayed by gate area congestion. Although data on aircraft delays due to airside problems are available, information on delay due to landside problems such as gate access or departure-gate holds is not. Such data are needed to define service levels. Some airport professionals believe that apron configuration is one of the principal characteristics influencing airport landside capacity. Demand and operating factors influencing service level and capacity of aircraft parking positions and gates are given in Table 6-1. Because of typical gate service or turnaround time, capacity over the short term, normally a period of 0.5 to 2 hr, is typically one aircraft per parking position and gate.

AIRCRAFT PARKING POSITION AND GATE 63 TABLE 6-1 DEMAND AND OPERATING FACTORS INFLUENCING SERVICE LEVEL AND CAPACiTY OF AIRCRAFT PARKING POSiTIONS AND GATES Factor Description Number of parking positions and Controls the total number of aircraft at gate at physical layout one time, should include hardstands and apron parking Utilization Ratio of time gate is effectively occupied (service, layover, and recovery) to total service time available (hours of operation), depends on flight turnaround time, including time for recycling between successive flight operations (a function of aircraft type and airline scheduling practices) Hours of operation (especially noise Limits number of operations that can be handled restrictions) per gate in a given day Flight schedule and aircraft mix Determines whether gates are likely to be available when needed, taking into account uncertainty in actual operation times compared with schedule; gates must be physically compatible with type of aircraft scheduled (see Utilization) Airline leases and operating Gate use strategy (see text) controls gate practices, airport management availability and utilization practice However, 100 percent gate utilization may not be achievable because of incompatibility between parking and ramp configuration or gate equipment and types of aircraft seeking access. Over the course of a full operating day, the patterns of arrivals and departures as well as airline ground operations, community factors, and weather determine the average number of operations per gate that can be served over the course of a year and whether a group of gates can accommodate additional flights. DEMAND PATfERNS The demand for aircraft gates or other aircraft parking positions is determined primarily by the flight schedule for the airport, including what type of aircraft is used for each flight. This demand may be influenced by external factors such as weather and airfield and airspace conditions. A ramp chart such as that in Figure 6-1 is often used to show the schedule. The scheduled gate service or turnaround time, which is the time between the arrival of a flight at the gate and availability of the gate for another arrival, is given for each aircraft. Airline operating practices affect turnaround times

o 2 4 6 8 10 12 14 16 18 20 22 0 GATE NUM OF CLASS AIRLINE GATE NUM I I I I I I I I I I I I ARRIVALS NARROW PA 114 A-.... A- A— A A-- A— P 7 NARROW PA 116 fr . 6 A-#— 6 NARROW PA 118 /- P ,4__... fr_. 7 NARROW AA 120 7 NARROW COPI 122 A-- / .' ,' A— #- A- A— A— A— A— A— 8 NARROW NARROW P1 OZ 124 126 • A- A- A- A-- fr A— NARROW OZ 128 A- I A— //- A-- A— A-- 8 NARROW EA 136 - 1 / I I A--.' NARROW EA 138 I / / A- NARROW 06 140 A- I A— 46-6- 9 NARROW EA 142 I A— NARROW EA 144 A— A 6-P 7 NARROW NARROW EA EA 146 148 , I A— A-- A-- A— A- P 8 NARROW EA 150 - / I / I A— A- A-P 8 NARROW GA 156 A- I I A—.— A— A— 8 NARROW GA 158 0 / / fr— A—,'— 7 NARROW CII 160 1-/ I A-- P 5 NARROW TW 162 NARROW TWCH 164 A-- A / Afl— 1— 6 NARROW CII 166 1— A— A-,'--- A— A— A— 7 NARROW AL 168 fr— I I I •t ,' 7 NARROW AL 170 A— / / A-- A-- A-j- NARROW AL 172 - J / A— A— A— 6—P 6 NARROW UA 213 A- / / A / A— A— 7 NARROW UA 215 A- A- A- A- / I I A- 8 NARROW UA 217 A-- A— A— /— •I I A- 7 NARROW NW 219 A- A- A- A- A- NARROW NW 221 A- 1- / A- A- A--- 8 NARROW NW 223 A— A-- A-- A-- 6 NARROW BN NY 225 A--' A- A-- 9 NARROW NY 227 ,i / / f fr A- A- P 7 NARROW DL 235 NARROW DL 237 1- 1 1 . I A-' A- NARROW DL 239 A--' / • / A-- / / 1 10 NARROW DL 241 A- A— / / A— A- A-- 7 NARROW DL 243 ,ø / 1 • A-- 7 NARROW DL 245 A— A— A— A— A— A— A- A-- 9 NARROW DL 297 A--- I I ft- / I I / / P 9 NARROW RC 255 A A- fr_—. A— A— 8 NARROW RC 257 IA-. A— A— A— 7 NARROW RC 258 A-- A— A—"-- A- A'— A— / 8 NARROW OH 261 / A--' A- A- 5 NARROW OH 263 1- / / . A-- 8 6-. 9 NARROW RC 265 NARROW RC 267 NARROW RC 269 A- A- A— A— I 0 I I I I I I I I I I I I 2 4 6 8 10 12 14 16 18 20 22 0 '---- Aircraft on Gate P Aircraft on Gate After 12 MIdnight FIGURE 6-1 Typical ramp chart showing gate utilization (1).

AIRCRAFT PARKING POSITION AND GATE 65 TABLE 6-2 TYPICAL GATE TURNAROUND TIMES FOR COMMERCIAL SERVICE AIRCRAFT (2,3) Flight Type Typical Aircraft Turnaround Timea(min) Long range, particularly Jumbo jet (B-747, DC-b, L-101 1) international 60-150 Medium to long range Long-range jet (B-767, DC-9) 45-90 Short to medium range Short-range, high-payload jet, turboprop (A-300, B-727, DASH 7) 25-60 Short-range, commuter Smaller prop, turboprop jet (Shorts 330-200, F-27, Gulfstream II) 20-45 aincludes gate occupancy and recycle time. Times for continuing flights on medium- to long-haul routes may be shorter. within the broad range associated with each type of aircraft. One of the more important factors in the determination of gate service time at a particular airport is whether the aircraft is making a stop en route or the flight is terminating at that airport. Gate service time is usually longer if the flight is terminating. Typical gate turnaround times for a range of aikraft and flight types (Table 6-2) become longer as flight distance and passenger load increase because more time is required for refueling, cargo and baggage loading, and aircraft service. Typical times are published by aircraft manufacturers for the various aircraft, but in practice these times vary considerably, depending on the nature of a particular airline operation at a particular airport. An alternative way of specifying gate service time is the average time that the aircraft actually occupies a gate and the average utilization of that gate over the period of analysis. Average gate utilization is normally no more than about 70 percent at most U.S. airports even during peak periods (2). At the simplest level, the demand for gates can be expressed as the total time that gate space is required, which is the sum of the service or turnaround times for all flights. This total time cannot exceed the product of the number of gates and the number of available operating hours in a day. The theoretical maximum throughput capacity of a series of gates at the airport is the total gate time supplied divided by the scheduled average gate occupancy time at the airport. In practice, the various factors influencing gate capacity may cause the actual number of flights handled to be less than this theoretical capacity. Actual schedules of individual flight operations are very significant in deter- mining actual gate capacity. A ramp chart (Figure 6-1) shows this actual schedule, including scheduled gate occupancy times. Many airports exhibit a pattern of flight operations that has two or three demand peaks—typically morning and late afternoon "rush hours" and per- haps a midday peak. During these periods all gates may be occupied for busier

66 MEASURING AIRPORT LANDSIDE CAPACITY airlines. At other times many gates may be vacant. Even during off-peak times for the airport as a whole, an individual airline may experience "gates full" at its exclusive-use gates. An airline hub-and-spoke operation presents a special case. Airports with hub-and-spoke operations experience several peaks of demand each day.1 OPERATING CHARACTERISTICS During a daily 1- or 2-hr peak period, capacity of the gates (i.e., the number of flights that can be accommodated) is simply the number of gates or other aircraft parking positions in the complex. If the average gate service time is much shorter than the period of interest, for example because of a high percentage of commuter or through flights, then capacity may be somewhat higher. As gate utilization increases, the risk of delay due to problems with operations increases. Frequent occurrence of such delays may indicate that the capacity of the gate system is being approached. Longer time gaps between departing and arriving flights at aircraft parking positions and gates generally imply greater flexibility to accommodate disruptions and variations in flight operations, and thus may represent higher service levels. Generally, data are not recorded for gate area aircraft delay except when caused by airside problems. Useful definitions of service levels for aircraft parking positions and gates require such data. The analyst must be explicit about the type of aircraft parking positions at an airport, because hardstands and remote parking positions may be included. Variation in how parking positions are counted and reported at different airports combined with the effects of diverse airline operating practices and airport lease arrangements account for the wide range in historic data on gate utilization. At Denver's Stapleton Airport, for example, average daily utiliza- tion at loading bridge-equipped gates currently varies from 1.5 to 12.0 flight turnarounds per gate per day. Similar statistics from the three major airports in the New York metropolitan region show a range from 2.5 to 17.6 flight turnarounds per gate per day.2 ANALYSIS TOOLS AND ASSESSMENT STANDARDS Relatively simple techniques, including a graphic technique published by the FAA (4), are often used to estimate gate capacity on the basis of average achievable number of operations per hour given the percentage of gates able to accommodate a given type of aircraft, the fraction of the daily flight

AIRCRAFT PARKING POSITION AND GATE 67 schedule allotted to each aircraft type, and the average service time for each aircraft type. Airline use restrictions can be incorporated into such procedures by specifying different types of aircraft or by analyzing separately each airline's group of gates. These techniques do not take into account possible delays to aircraft operations if flights are too tightly scheduled and a disrup- tion occurs. A more sophisticated approach to assessing capacity involves examination of the ramp chart for a period such as the peak-month average day. When specific gates are not occupied, there may be time slots available to accommo- date additional aircraft, depending on the gate use strategy at the airport. These slots must be at least long enough to equal the average service time of the type of aircraft being considered and must be available at the scheduled arrival and departure times of the aircraft. Computerized models have been developed (see Appendix B) to perform these slot calculations for the typical operating day. Adequate calculations can be made by hand for relatively small numbers of gates. EXAMPLE OF ASSESSMENT PROCESS3 Suppose that a unit terminal at an airport has nine gates. Airport management would like to determine whether current flight schedules will permit addi- tional service to be accommodated at these gates. Describe Component Each gate is equipped with a passenger loading bridge. Geometry of the apron area allows accommodation of only narrowbody aircraft. Current lease ar- rangements grant exclusive use of each gate to the lease holder. Six airlines hold leases on one to three gates each. Several other airlines use gates under subleases, operating one to two flights per day. Describe Demand and Operating Factors A ramp chart is drawn to show the current daily demand at the gate complex (Figure 6-2), which consists of 120 flight operations (e.g., 60 flights).4 Several airlines use this airport as an overnight layover, leaving late-arriving aircraft parked for early-morning departure. Passengers are primarily vacation and leisure travelers during the airport's peak-month average day. This ramp chart shows that aircraft service times vary substantially, from 30 min to more than 75 mm, during the period when the airport is active (6:00

. :....& UA 727 72S DC9 72S 08$ 08$ UA RC N1 NI UA 2 — 095 jW1 ~ PA DOS D9S 08$ PA PA NW PA1 PA I PA — 72S 72S 72S D1O 72$ — 72$ 72S AA NW TW OZ 1W AA OZNW AA 1W 72S 72S 72S D9S 72$ M80 72S 09$ 72S 72$ LLI AL AL PA AL AL AL 725 72S 73S 72$ 73$ 73S — 735 735 6 . I &.. . ..... 095 095 727 72S A83 72S 757 72S 72S EA1 EA EAI I EAI — D9S 757 72$ 72$ 095 — 09$ 8_EA J EA A83 A83 A53 A83 PE PEPE PE PE PE 72$ 72S 737 72S 737 737 725 72$ 72S 0600 0900 1200 1500 1800 2100 2400 TIME OF DAY FIGURE 6-2 Ramp chart for example airport [adapted from McKelvey (5)].

AIR CRAFI' PARKING POSITION AND GATE 69 0 w 0. 0 0 0 U, w C, U. 0 cc UJ z 0900 1500 2100 TIME OF DAY FIGURE 6-3 Aircraft on ground during average day. a.m. to 11:00 p.m.). If overnight layovers are counted as 75 min of active gate use, the average turnaround time for 60 flights in a 24-hr period is slightly more than 30 min and the median time is approximately 45 mm. Individual gates are empty for extended periods during the day, and at no time are all gates occupied (Figure 6-3). Estimate Service Levels and Service Volumes An indicator of current service level is given by the observation that during the active 17 hr of an operating day, the current ramp chart contains in principle 23 to 34 slots of 30 to 45 min each. A simple measure of gate utilization is the ratio of these slots to slots available: Utilization = flights/[(gates) x (slots/day)] = 60/[(9) x (23-34)] = 0.20-0.30 This analysis suggests that additional capacity may be realized above the 60 flights per day now served if some of these slots could be filled with new flights. However, the formula does not allow for time required between departure of one flight and arrival of the next at a single gate position nor for inevitable day-to-day variations in flight operations. Attempts to achieve service volumes of 200 to 300 flights per day, based on estimated 100 percent utilization of the nine-gate complex, would unquestionably result in serious declines in service level. However, this simple method provides a useful initial indication that an increase in service volume may be achievable.

70 MEASURING AIRPORT LANDSIDE CAPACiTY A somewhat more realistic analysis of gate utilization may be made by constructing a graph like that in Figure 6-3 from the ramp chart, which shows how many gates are available at any hour throughout the day. The ratio of the area under the plot of gate occupancy to the total area under the line of maximum gate occupancy is an indicator of gate utilization, and is computed as follows: Utilization = total aircraft at gates/[(gates) x (hours)] = 611(9 x 17) = 0.39 Using this measure, the airport operator might estimate that with 100 percent utilization a total schedule of about 150 flights per day would be achievable, although such a schedule would allow little flexibility for accommodating flight delays due to weather, air traffic, or ground service conditions. Also, the analysis fails to recognize that different types of flights and aircraft may have different gate service time requirements. A still more exact estimate of gate capacity could be made by direct inspection of the ramp chart to identify open periods adequate to accommo- date new flights. For example, if that duration is 60 mm, approximately 72 such slots are available. If the required duration is 75 mm, the number of slots declines to 46. At 90 min duration, the number is 34. Total flights at the nine- gate complex might in principle be increased to 94, 106, or 132 flights per day, depending on the time allotted for each slot. However, these increasing service volumes reflect a probable decline in service level indicated by the reduced time between scheduled flight operations. Selection of a realistic slot time duration depends on characteristics of expected flights, such as whether they are through flights or terminating ones, and the likely passenger loads. To realize this capacity would require either that each airline schedule new flights to fully occupy its leased gates or that the gate use strategy be changed. Inspection of Figure 6-2 shows, for example, that the nine flights using Gate 9 could be served, with the exception of one arriving at approximately 8:30 a.m., at other gates. However, the airline operating these flights has exclusive use of Gate 9, which limits the current service volume to 120 flight operations. Fewer gates would be required to serve current demand if all gates were operated under a common use strategy and thus a considerably greater number of flights might then be served by the nine-gate complex. RESEARCH NEEDS Few data are now collected on delays due to landside problems such as gate access and holds. There is a pressing need to fill this gap in the comprehensive

AIRCRAFT PARKING POSITION AND GATE 71 statistical data base on air carrier delay and airport system operating capacity maintained by the FAA. Data on airline gate holds, delayed gate access for arriving flights, and similar measures of landside-based delay should be collected on a routine basis in a form compatible with data on airside delay statistics. Such data, reviewed within a context of other sources of delay, would be a basis for developing operationally justified service-level targets. NOTES Peaking in this case depends on airline scheduling as well as on underlying passenger demand for travel. These daily statistics are taken from monthly reports by airport operators. Statistical analysis of annual operations at 153 U.S. airports indicates that at airports with 10 or more gates, daily service volumes as high as 18 flights per gate may be sustainable, but only under ideal conditions of aircraft mix, flight routes, airline operating practices, and weather conditions. Subsections correspond to Steps 3, 5, 7, and 8 in the assessment process shown in Figure 3-1. Attention may also be given to relevant relationships among compo- nents (Step 4) and community factors (Step 6). Arrival and departure are considered separate operations. Each flight serving an airport thus results in two flight operations. REFERENCES Peat Marwick and Mitchell & Co.—Airport Consulting Services. LaGuardia Air- port Capacity Study Phase]; Preliminary Analysis Final Report. Port Authority of, New York and New Jersey, Jan. 1986. R. Horonjeff and F. X. McKelvey. Planning and Design of Airports, 3rd ed. McGraw-Hill, New York, 1983. Ralph M. Parsons Co. The Apron and Terminal Building Planning Report. Report FAA-RD-75-191. Federal Aviation Administration, U.S. Department of Transporta- tion, July 1975. Techniques for Determining Airport Airside Capacity and Delay. Report FAA- RD-74-124. Federal Aviation Administration, U.S. Department of Transportation, June 1974. F. X. McKelvey. Palm Beach International Airport: Interim Airport Operating and Use Plan. Aviation Planning Associates, Cincinnati, Ohio, July 1984.

7 Passenger Waiting Area Passengers waiting in areas' serving aircraft gates and terminal lobbies may be subjected to crowding and congestion if facilities are inadequate. Avail- ability of seating, general quality of the surroundings, and length of time the passenger waits have a substantial influence on perceived service levels in these areas. DESCRIPTION The number of passengers waiting for flight departures and arrivals depends primarily on the number of aircraft served by the waiting area, aircraft seating capacity, aircraft passenger load factors, degree to which passengers are accompanied by family or friends, passenger arrival time at the airport, and the length of time between commencement of boarding of a flight and its departure. Interdependence among components may have a substantial impact on number of passengers waiting in various areas of the terminal. For exam- ple, delays at a passenger security screening device may delay passengers' arrival at waiting areas, and thereby reduce the number waiting to board. Variations in aircraft departure times may increase the number of passengers waiting. For example, almost half of the departing passengers at New York City's John F. Kennedy International arrive two or more hours in advance of their scheduled departure times (Port Authority of New York and New Jersey, 1982 survey). Such behavior, related at least in part to the predominantly long- haul flight schedule at this airport, substantially increases the number of 72

PASSENGER WAITING AREA 73 passengers waiting in the terminal. Final steps in processing enplaning pas- sengers, including seat assignment and ticketing, may impose delays and generally increase the length of time during which passengers occupy waiting areas as well as the number of people waiting. Airlines normally seek to avoid crowding in their exclusive-use areas. However, during the 15 to 20 min before departure when about 70 to 90 percent of the passengers are in the vicinity of the gate, crowding is sometimes unavoidable. Design of a common waiting area for several gates is used at some airports to avoid severe crowding. Demand and operating factors influencing service level and capacity in waiting areas are given in Table 7-1. Service levels and service volumes over TABLE 7-1 DEMAND AND OPERATING FACTORS INFLUENCING SERVICE LEVEL AND CAPACITY OF PASSENGER WAITING AREAS Factor Description Waiting and circulation area Space available for people to move around (lounge and accessible corridor) and wait for departing flights; depends on terminal configuration, for example, waiting areas may be shared by passengers on several departing flights or restricted to single gate Seating and waiting-area geometiy Seated people may occupy more space but are accommodated at higher service levels Flight schedule, aircraft type, Larger aircraft typically mean higher passenger load, and gate passenger loads; areas used jointly to serve utilization simultaneous departures Boarding method Availability and type of jetways, stairs, and doors from terminal to aircraft affect rates at which passengers board as well as airline passenger handling procedures Passenger behavioral How soon before scheduled departure people characteristics and airline service arrive at gate areas, amount of carry-on characteristics baggage, knowledge of system, and percentage of special needs passengers (families with small children, elderly, handicapped, first class and business travelers); airline passenger service policy, seat assignment and boarding pass practices the short run—typically a period of one-half to three-quarters of an hour—are determined primarily by comparing areas available to passengers, the number of passengers waiting and the amount of baggage they have with them, and targets for available space per person.

74 MEASURING AIRPORT LANDSIDE CAPACITY DEMAND PATFERNS The number of passengers waiting is determined by flight schedules and passenger behavior, including the length of time it takes for passengers to pass through the other components of the landside. The number of waiting pas- sengers in an area generally is greater when passengers arrive at the airport early for their flights and decreases when more time is required for check-in or transfer. Waiting areas such as gate lounges serve originating and transfer passengers, whereas terminal lobbies accommodate primarily originating pas- sengers and their nontraveling companions. When weather or air traffic conditions disrupt flight operations schedules, corridors and airline passenger service counters may become important passenger waiting areas and help to maintain higher service levels throughout the terminal as the number of waiting passengers increases. Demand is frequently expressed as the percentage of departing passengers arriving at the waiting area in discrete intervals of time before scheduled flight departure. Rules of thumb have been proposed to generalize this relationship, but direct observation of conditions at a, particular airport remains the only reliable way of describing demand patterns. Peak demand levels are likely to occur during periods when all gates in a terminal concourse are occupied by flights with closely scheduled departure times. Airline hub-and-spoke operations, with their limited scheduling win- dows, are likely to exhibit sharp peaking of waiting-area occupancy levels. OPERATING CHARACTERISTICS Effective design is the primary means for assuring adequate service levels in waiting areas. Some of the more frequently used design space standards for gate lounges and other terminal waiting areas are given in Table 7-2 (1-5). These standards appear to be generally upheld in current practice, except perhaps at those airports where introduction of larger aircraft or a new airline hub-and-spoke operation has produced larger passenger loads per gate than was anticipated in design. Service-level targets for lounges may differ among airlines and among airports but are typically based on market conditions. For example, the management of New York City's La Guardia Airport tries to maintain 10 ft2 per person in gate lounges, whereas planning for the expan- sion of John F. Kennedy International with its different passenger mix is based on a target of 15 ft2 per plan-year passenger (design memorandum, Port Authority of New York and New Jersey, 1985-1986).

PASSENGER WAITING AREA 75 TABLE 7-2 TYPICAL SPACE STANDARDS USED IN PLANNING AND DESIGN Design Situation Space Standard (ft2/person) IATA design standard for departure lounges (1) 8.5 per aircraft seat IATA suggested breakdown level of service in > 6.5 for more than 15 mm hoidrooms (2) IATA suggested breakdown level of service in > 10.8 for more than 15 mm waiting and circulation areas (2) Unofficial FAA minimum-space guidelines for 6.7-10.0 per aircraft seat; 15 departure lounge design (3) per seated waiting passenger Architectural reference standard for adequate 13 waiting and circulation space with baggage (4) Design loading of urban transit vehicles (5) 3-4 Nom: IATA = Intemational Air Transport Association. Transporter vehicles used in some airports to connect the terminal to remote apron parking areas are a special case. The transporter vehicle serves as an extension of the gate lounge during the time between commencement of boarding and transporter departure. Standards within the transporter vehicle are generally lower than those for gate lounges and may approximate those used in design of urban transit vehicles (Table 7-2). Capacity is generally determined by manufacturers' standards. ANALYSIS TOOLS AND ASSESSMENT STANDARDS In analyzing the capacity of passenger waiting areas, service levels are usually indicated by the ratio of the number of people in the area and the size of that area. Targets for this ratio may vary with the time passengers wait for boarding, but in many cases, only a space standard is stated. Airlines may also employ standards for the number of seats that should be available for given numbers of passengers. In most applications, estimation of passenger demand over time is neces- sary. Observation and sample passenger counts are often used to make these estimates. Mathematical queueing and simulation models to predict the arrival of enplaning passengers at waiting areas before scheduled departure may also be used (see Appendix B). Simulations for the terminal building as a whole may be used to develop curves of the percentage of a flight's passenger load likely to be in the gate lounges of that terminal during discrete time intervals before scheduled flight departure. Often these curves indicate simply the number of people expected to be in the waiting area versus the time before

76 MEASURING AIRPORT LANDSIDE CAPACITY departure. These curves may then be used to determine whether introduction of new aircraft or scheduled flights would create overload conditions. Such models and field observations may be used to develop standard distributions of passenger arrivals over time for particular types of flights at a particular airport. These standard distributions may then be used to approximate ex- pected conditions if new services are introduced or new facilities are being planned. EXAMPLE OF ASSESSMENT PROCESS2 Assume that there is a departure lounge serving a single gate. The lounge is fully enclosed except for the entry where airline personnel check tickets and assign seats and is thus relatively isolated from other waiting areas. Figure 7-1 shows such a departure lounge. Seating in the lounge has been arranged to permit free circulation of passengers with carry-on baggage in the area near the aircraft loading bridge entry. Additional seating might be provided at the airline's discretion. Describe Component This gate is currently leased by a major commercial service carrier that normally schedules Boeing 727 aircraft for this gate but may soon schedule the Boeing 767 also. The departure lounge is large, with a gross floor area of approximately 2,900 ft2. Net space for passengers, deducting the corridor for deplaning passengers and the check-in counter area, is approximately 2,600 W. There is currently seating for 96. Describe Demand and Operating Factors Experience at the airport has shown that 80 percent of the passengers on a flight are likely to be in the gate lounge 20 min before scheduled departure. When the airline begins boarding, normally 15 min before scheduled depar- ture, this fraction may be 87 percent. During the busy month of August, this airline's flights have an average load factor of 85 percent, and during the busiest parts of the day, flights are usually

36 ft I 6ft H — Aircraft Loading Bridge Movable Seating 11 Telephone and Airilne Storage Airline Service Center ILI1LII1 IuILru a1IIi1 111 I!II!1 I- 80ff FIGURE 7-1 Example departure lounge.

78 MEASURING AIRPORT LANDSIDE CAPACiTY full and there are standby passengers waiting. For flights served by Boeing 727 aircraft, which have a seating capacity of approximately 130 to 160 passengers, there are typically 110 to 140 passengers in the lounge just before boarding. When boarding begins, the number of passengers waiting in the lounge area begins to decline. One agent is assigned for boarding, able to pull tickets and check boarding passes at a rate of approximately 10 passengers per minute. Flight schedules and aircraft service procedures keep this gate busy, with approximately one aircraft arriving and departing every hour. Estimate Service Levels and Service Volumes Because only one flight uses the gate per hour, performance of the gate agent determines maximum throughput. The agent can board 10 passengers per minute or a total of 150 passengers in the 15 min used for boarding. That is, Throughput = boarding rate x boarding time = [(10 passengers per minute per agent) x (1 agent)] x (15 mm) = 150 passengers per flight However, passenger demand reflected by aircraft loads is not that high. The airline operates a premium service and attempts to ensure that its passengers are comfortable. The company would like to maintain a target service level of at least 15 ft2 per passenger when the departure lounge is most heavily occupied, which is usually immediately before flight boarding begins, and would like most passengers to have a seat available in the lounge. The airline is concerned that starting service with new aircraft will cause what this airline believes to be excessive crowding, that is, less than 15 ft2 per person. This is an unacceptably low service level for this airline in this market, although it is a relatively high target when compared with the 8-10 ft2 per passenger used in many planning and design studies. The service-level target establishes the maximum service volume of the lounge: Maximum service volume = lounge area/space standard = 2,600 ft2/(15 ft2/passenger) = 173 Given the pattern of passenger arrivals at the lounge, this maximum service volume would be 87 percent of the flight's passenger load. The gate lounge

PASSENGER WAITING AREA 79 can then accommodate an aircraft with as many as 199 passengers and still meet the service-level target. The Boeing 727 aircraft is clearly in an acceptable range, with typical load levels of 110 to 140 passengers. With 96 seats in the lounge, approximately half of the capacity load can be seated and over two-thirds of the typical B-727 passenger load. In the absence of a specific seating stafldard, this figure would be considered acceptable. If Boeing 767 aircraft are introduced, with 210 to 230 seats per aircraft, the airline's service-level target may be violated if load factors and passenger behavior do not change. RESEARCH NEEDS Architectural and planning standards for waiting areas appear to be adequate for decision making. Although questions may arise in particular cases of facility, design, general research on space standards is not a high-priority need. Similarly, airlines operating gates on an exclusive-use basis will set their own standards for boarding times. However, airports considering joint-use operations may benefit from having better data on reasonable lengths of time to board for different types of aircraft to serve as a basis for setting schedules. NOTES Waiting areas adjacent to gates, which may be typically termed gate lounges, departure lounges, or holdrooms, are the principal focus of this discussion. However, the design of some airports includes waiting areas outside the immediate vicinity of the gate lounge or holdroom for a particular gate but close enough to be used by passengers for that gate. Therefore, the more general term "waiting area" is used here to cover all available waiting space. Subsections correspond to Steps 3, 5, 7, and 8 in the assessment process shown in Figure 3-1. Attention may also be given to relevant relationships among compo- nents (Step 4) and community factors (Step 6). REFERENCES Airline Aircraft Gates and Passenger Terminal Space Approximations. AD/SC Report 4. Air Transport Association of America, Washington, D.C., July 1977. Airport Terminal Reference Manual, 6th ed. International Air Transport Associa- tion, Montreal, Quebec, Canada, Sept. 1978.

80 MEASURING AIRPORT LANDSIDE CAPACiTY Ralph M. Parsons Co. The Apron and Terminal Building Planning Report. Report FAA-RD-75-191. Federal Aviation Administration, U.S. Department of Transporta- tion, July 1975. J. Panero and M. Zelnick. Human Dimensions and Interior Space: A Source Book of Design Reference Standards. Whitney Libraiy of Design. New York, 1979. M. Jacobs, R. E. Skinner, Jr., and A. Lemer. Transit Project Planning Guidance: Estimation of Transit Supply Parameters. Report UMTA-MA-09-9015 -85-01. Transportation Systems Center, Cambridge, Mass., Oct. 1984.

8 Passenger Security Screening All originating passengers must pass through a security screening. In addition, interline transfer passengers at some airports may be required to clear a security screening on their way to a connecting flight. These areas are often points of queueing and delay for passengers. DESCRIPTION Passenger security screening occurs in concourse corridors at entrances to terminal gate areas or at the entry to gate lounges. Equipment configuration and staffing are the primary factors influencing capacity. Corridor width often controls screening capacity, determining the number of inspection channels that can be set up. Each channel consists of a metal-detecting device (magnetometer) through which a passenger walks and one or sometimes two x-ray devices for inspec- tion of carry-on luggage. Inspectors may undertake additional screening at their discretion, including hand searching of carry-on items and close screen- ing (using a hand-held magnetometer) of individual passengers. Security is often supervised by the airport operator. Sometimes airlines, in cooperation with the airport operator, hire security personnel themselves and establish their own procedures to supplement or replace those of the airport operator. Heightened concern over possible terrorist threats has led a number of airlines, especially overseas carriers, to augment their security screening procedures. Demand and operating factors influencing service level and capacity for passenger security screening are given in Table 8-1. Because of the steady flow of passengers served at most security screening areas and because the 81

82 MEASURING AIRPORT LANDSIDE CAPACiTY TABLE 8-1 DEMAND AND OPERATING FACTORS INFLUENCING SERVICE LEVEL AND CAPACiTY OF PASSENGER SECURiTY SCREENING AREAS Factor Description Number of channels, space, and personnel Type, equipment sensitivity, and airport/airline/agent policy and practice Passenger characteristics Building layout and passenger circulation patterns Flight schedule and load Influences number of passengers processed per unit time (magnetometer and x-ray considered separately) Determines average service time per passenger and like- lihood of close inspection Amount of hand luggage, mobility, and patterns of ar- rival influence average service time as well as number of passengers Interference among pedestrian flows can influence flow rates and create congestion Basic determinant of number and direction of people on concourse security screening is among the last of several possible barriers facing depart- ing passengers, short-term capacity of this component is the primary con- cern—queues typically build and clear over short periods. Assessment of delays related to security screening is usually made with respect to a peak 1-hr period, but attention should also be given to the shorter time over which service quality can decline. DEMAND PA11ERNS The timing of passengers' arrival at the gate for flight departure determines the basic demand on security screening facilities. Patterns of this timing have been documented at specific airports. When such data are not available, arrival patterns can be estimated from passenger arrival times at the airport if allowance is made for the several parallel paths by which passengers reach the security screening area. In addition to passenger demand, other people accom- panying passengers to the gate area—nontravelers—typically must pass through security screening as well. Processing rates at the security screening area are affected by the number and size of pieces of hand luggage carried on. Holiday travelers, tourists, and business travelers seeking to avoid checking baggage may have a larger number of parcels to be inspected. High percentages of passengers in wheel- chairs or children in strollers may also lead to slower processing. To reduce the total number using the security screening area, many airports have posted signs discouraging nontravelers from entering the secure area. This restriction is enforced in some airports by requiring passengers to show their tickets or boarding passes before entering the security screening area.

PASSENGER SECURI7Y SCREENING 83 OPERATING CHARACTERISTICS The sensitivity of magnetometers can be varied to pick up smaller amounts of metal on the passenger's person. Less sensitive settings will tend to decrease average service time by reducing the frequency of intensive inspections. When passengers arrive at rates exceeding the service rate of the security screening area, queues form. The delays associated with waiting in this queue are the principal basis for judging the service level of the passenger security screening area. Observed processing times for security screening are pre- sented for several airports in Table 8-2. TABLE 8-2 TYPICAL PROCESSING TIMES FOR SECURiTY SCREENING (1,2) Average Processing Airport Time (mm/passenger) Miami (1) 0.47-0.51 Denver (1) 0.18-0.56 La Guardia (1) 0.15-0.77 La Guardia (five concourses)a 0.62-0.90 Kennedy (eight unit terminals)a 0.07-0.16 Hand-checked baggage (2) 0.50-1.00 Automated check (2) 0.50-0.67 °Survey by Port Authority of New York and New Jersey. In some situations space for passengers to queue may be limited, and crowding may become another important clement of service level. Data on passenger delays and queue lengths for security screening are not generally available. A British survey of passenger attitudes found that passengers were willing to tolerate delays of 6.5 to 10.5 min per passenger before they judged the level of service to be bad (3). However, U.S. practice suggests that such delays are encountered only rarely. A delay of 5 mm, for example, at a facility with service and demand characteristics producing an average service time of 0.50 min per passenger would indicate an average queue length of 9 to 10 passengers. At typical peak-period passenger volumes, small differences be- tween service and arrival rates can rapidly cause large queues to build. Persistence of such queues during a peak hour is often evidence of a capacity problem at the security screening area.

84 MEASURING AIRPORT LANDSIDE CAPACiTY ANALYSIS TOOLS AND ASSESSMENT STANDARDS Among the various components of the airport landside, security screening most closely fits the assumptions of a simple queueing model. The average time required for clearance of a passenger, the variability of that time, and the rate of passenger arrival at the security screening area are key variables for capacity assessment. EXAMPLE OF ASSESSMENT PROCESS1 Assume that there is a single corridor leading to a concourse of 10 gates, all used by a single airline. After ticketing, departing passengers must pass through a security screening at the beginning of this corridor (Figure 8-1). Describe Component The equipment consists of one magnetometer and one x-ray machine with belt drive manned by a single inspector. Two other inspectors are generally on duty to relieve the x-ray operator and to conduct hand searches of luggage and close inspection of passengers who set off the magnetometer alarm. The corridor is intersected by a circulation corridor through which pas- sengers from other airlines pass in their trip from the terminal lobby to gates. This cross flow may become congested when traffic in both directions is heavy or when a long queue builds at the security screening area. Although this problem has occurred with fair regularity, no detailed data have been collected on passenger flows through the security screening. Describe Demand and Operating Factors The passengers using this security screening area are primarily business travelers taking 1- to 3-day trips. They generally have carry-on luggage, seldom check bags, and tend to arrive at the airport with as little time before their scheduled flight departure as they believe prudent. The airline has observed that 40 percent of their passengers at this airport typically

PASSENGER SECURFIT SCREENING 85 To Another Airline's Ticketing To Gates lI C 0 kD l I 0 0 CL • 18 ft To Another Airline's Ticketing To Ticketing 2Oft FIGURE 8-1 Example of passenger security screening. arrive within 112 hr of their scheduled flight. The fraction increases to 60 percent 20 min before flight time and to 95 percent 10 mm before flight time. The airline's flight schedule is arranged to serve these travelers, with peak numbers of departures in the early morning, at midday, and in late afternoon. Each of these peaks has a 20-min period when six flights are scheduled to depart, each flight served by aircraft with seating capacity of approximately 110 passengers. Over the course of each peak hour, there are 10 such flights. The airline's load factor is approximately 78 percent, except during the hectic Christmas and Thanksgiving seasons, when it is higher, and during the relatively quiet August vacation period. The airline's observations of passengers' arrival characteristics indicate that 35 percent arrive in the first 10 min of the 20-min period before flight time. Under the worst case, with all six peak-period flights scheduled for the same departure time, the peak passenger arrival rate may be estimated as follows: Peak arrival rate = (no. of flights x aircraft capacity) x (load factor) x (percent arriving in peak time) = (6 x 110) x (0.78) x (0.35) = 180 passengers during the peak 10 mm

86 MEASURING AIRPORT LANDSIDE CAPACiTY The security screening system achieves a normal average inspection rate of approximately 15 passengers per minute. During peak periods passenger behavior and inspection procedures may cause the rate to increase to 18 passengers per minute. Estimate Service Level and Service Volumes On the basis of the inspection rates achieved, the security screening normal throughput is estimated as follows: Normal throughput = (normal inspection rate) x (time) = (15 passengers/mm) x (20-mm peak) = 300 passengers in peak 20 min = (15) x (60) = 900 passengers per hour During peak times the maximum throughput is Peak maximum throughput = (peak inspection rate) x (time) = (18 passengers/mm) x (20-rn in peak) = 360 passengers in peak 20 rnin = (18) x (60) = 1,080 passengers per hour The capacity of the security screening would then be in the range of 900 to 1,080 passengers per hour. The actual hourly service volumes are much lower: Demand service volume = (flights) x (seats) x (load factor) = (10) x (110) x (0.78) = 858 passengers The security screening appears to be adequate on this hourly basis. However, the peak arrival rate is estimated to be 180 passengers during a 10-min period, or about 18 passengers per minute. Because this rate results in a higher than normal throughput, this is a worst-case situation in which queue- ing may occur until the processing rate catches up with passenger arrivals. The total estimated length of the queue is the difference between the number of arrivals-180 passengers—and the number served—at 15 passengers per min- ute, 150 during this peak 10 mm—or approximately 30 passengers. As a service-level target, the airline has suggested that passengers never encounter delays of more than 3 min at the security screening, although management recognizes that occasional lapses may occur during holiday

PASSENGER SECURITY SCREENING 87 seasons. The last person entering the queue would require approximately 2 min to pass through security (at a rate of 15 passengers per minute for those ahead of him to clear the facility), which is well within the airline's standards. RESEARCH NEEDS Substantial research efforts are being directed toward developing new devices for screening both passengers and baggage. To the extent that new devices are introduced and influence average service times, data on these new times will be needed to ensure that security screening does not become a constraint on landside capacity. NOTE 1. Subsections correspond to Steps 3, 5, 7, and 8 in the assessment process shown in Figure 3-1. Attention may also be given to relevant relationships among components (Step 4) and community factors (Step 6). REFERENCES P. Mandle, F. LaMagna, and E. Whitlock. Collection of Calibration and Validation Data for an Airport Landside Dynamic Simulation Model. Report TSC-FAA-80-3. Federal Aviation Administration, U.S. Department of Transportation, April 1980. R. Horonjeff and F. X. McKelvey. Planning and Design of Airports, 3rd ed. McGraw-Hill, New York, 1983. S. Mumayiz and N. Ashford. Methodology for Planning and Operations Manage- ment of Airport Terminal Facilities. In Transportation Research Record 1094, TRB, National Research Council, Washington, D.C., 1986, pp. 24-35.

9 Terminal Circulation The terminal circulation component is used by all air passengers, but the focus of this chapter is primarily on the beginning or the end of their trip at the airport of interest. Conditions facing passengers transferring between connect- ing flights are addressed in Chapter 16. DESCRIPTION Generally speaking, the total time it takes for a passenger to move through the airport's landside is the sum of the time waiting for service and being served at each of the functional components used along the way, such as check-in or baggage claim, plus the time required to travel between components. If only the travel time is added, the sum represents an estimate of the time it takes to travel through the landside without stopping. A business traveler with all tickets and boarding passes in hand and with no luggage to check or retrieve might allow just this much time plus time for brief delays at the security screening, at the gate awaiting departure, and at ground transportation for the terminal portion of his trip. Except at airports where individual airlines exercise complete control over unit terminals, virtually all aspects of this terminal circulation component are the responsibility of the airport operator. The principal demand and operating factors influencing service level and capacity for the terminal circulation component are given in Table 9-1. These same factors may be cited for their influence on general circulation. For elevators and people movers, details may also be required on the dimensions and operating characteristics of the specific mechanical system. 88

TERMINAL CIRCULATION 89 TABLE 9-1 DEMAND AND OPERATING FACTORS INFLUENCING SERVICE LEVEL AND CAPACiTY OF TERMINAL CIRCULATION Factor Description Terminal configuration Space available for people to move fiely without conflict of flows; availability of alternative paths; placement of seat- ing, commercial activity, stairs, escalators Passenger characteristics Amount of hand luggage, mobility, and rate of arrival before scheduled departure influence demand loads and service time Flight schedule and load Basic determinant of number and direction of people on concourse Nom: These same factors affect circulation on elevators and people movers; specific mechanical systems, however, may differ. DEMAND PATTERNS Passenger demand within this component is determined primarily by patterns of passenger arrival at the airport before scheduled flight departures, by the paths passengers take going between gates and the terminal curb, and by speeds at which both arriving and departing passengers make this trip. The rate at which passengers move through the landside depends on such charac- teristics as age, purpose for the travel, and time available before the flight or following the arrival; on the degree of crowding encountered along the way; and on the geometry of the path traveled. Typical free-flow distributions of walking speeds in areas similar to an airport terminal (Figure 9-1) show considerable variation. As the average circulation space per person falls below approximately 10 to 15 ft2, walking speed may begin to decrease substantially below these free-flow levels (1, p. 13-7). OPERATING CHARACTERISTICS Although terminal building design determines what routes may be available to passengers, airport operators and airlines personnel can use partitions and signs to direct passenger traffic and improve overall terminal circulation in an attempt to optimize the path a passenger must take through the terminal building. Escalators and elevators may become bottlenecks but generally improve service levels. Reduction of the degree of physical obstruction and likelihood of intersection of pedestrian paths going in different directions tends to improve pedestrian travel speeds and reduce congestion. Passengers normally do not take the shortest route through the terminal. In a study of Vancouver International Airport, for example, Transport Canada

40 P..t A.,th...It., fl,. r.....I..., iIJ v fl >. 30 90 MEASURING AIRPORT LANDSIDE CAPACiTY 0 1 1 I - r I I I 0 100 200 300 400 WALKING SPEED (feet per minute) FIGURE 9-1 Typical free-flow pedestrian walking speeds (1). found that actual walking distances of travelers were 1.3 to 2.1 times longer than the ideal shortest-path route. In two sections of the terminal, the ratio exceeded a factor of 5 (2). Concessions, rest rooms, and pay telephones located along corridors typically create some congestion and slow general travel speeds as well as increase the path lengths of the passengers who use these facilities. ANALYSIS TOOLS AND ASSESSMENT STANDARDS In general, the terminal circulation component may be considered a pedestrian circulation problem and analyzed by using procedures and 'standards such as those suggested in TRB's Highway Capacity Manual (3). The length of the passenger's pathway, the passenger's walking speed, number of level changes, and the degree of interference the passenger encounters along the way are key variables in the assessment. The time spent traveling between curb and gate is the principal measure of service level and a determinant of capacity. Number of level changes and how complicated the pathway appears to the passenger may also affect service levels. Very little generally comparable data have been collected for describ- ing circulation service levels at typical airports. Service-level definitions presented in the Highway Capacity Manual may be useful at airports and should be considered when circulation problems

TERMINAL CIRCULATION 91 develop. Procedures suggested in that report may be useful in assessing the capacity of specific terminal design configurations. EXAMPLE OF ASSESSMENT PROCESS1 Consider the situation described in Chapter 8 for passenger security screening. Passenger flow (i.e., pedestrians) in a circulation corridor crosses flow going to and from the security screening for a concourse of 10 gates. The airport operator wishes to determine what the maximum tolerable queue length at the security screening is such that pedestrian cross traffic is not seriously impeded. Describe Component The circulation corridor is 18 ft wide, unobstructed by columns or furniture (see Figure 8-1). It is intersected by another corridor 20 ft wide leading to the security screening devices. Pedestrian flow in the circulation corridor is typically 75 to 85 percent in one direction, with that direction depending on time of day. Describe Demand and Operating Factors Airport planning staff is assigned to observe the pedestrian flows during busy periods for one week. When there are no queueing problems at the security screening, volumes in the circulation corridor are in the range of 110 to 130 pedestrians per minute in the peak direction and 35 to 45 pedestrians per minute in the opposing direction. Walking speeds were not recorded, and therefore must be assumed. Estimate Service Levels and Service Volumes The airport operator decides that the conditions described in the Highway Capacity Manual as level-of-service D are tolerable, but only for short intervals—iS to 25 min at a time, two to three times a day. As described in the manual, level-of-service D represents restricted pedestrian walking speeds, and the probability of conflict with pedestrians moving across the flow or in

92 MEASURING AIRPORT LANDSIDE CAPACITY the opposite direction is high. Although flow remains "reasonably fluid," considerable friction and interaction are likely (3). The manual suggests that level-of-service 1) is achievable with pedestrian flows of up to 15 pedestrians per minute per linear foot of corridor cross section, and with average space as low as 15 ft2 per person. Using these figures as targets, the following estimates of capacity may be made: Corridor service volume < (flow standard for level of service) x (corridor width) ~ (15 ped/min/ft) x (18 ft) <270 ped/min for both directions This volume is considerably higher than the total of 145 to 175 pedestrians per minute observed by staff. Applying the space standard in the crossover area shared between the two corridors, the following estimate is made: Maximum pedestrian density < (area)/(density standard) ~ (18 ft x 20 ft)/(15 ft2/ped) <_ 360/15 = 24 pedestrians in the crossover area Suppose an average pedestrian walking speed of 180 fVmin is assumed (see Figure 9-1). Each passenger would require approximately 6 sec (0.1 mm) to cross the circulation corridor and enter the security screening channel. If passengers are being processed at the security screening at a peak average rate of 18 per minute (see Chapter 8), or 1 every 3.3 sec, there will be 1 to 2 such passengers in the crossover area at any one time. Similarly, passengers walking at the same average speed in the circulation corridor are in the crossover area for approximately 6.7 sec, implying that at flow rates of 145 to 175 pedestrians per minute, there will be 16 to 20 such pedestrians in the crossover area. The total number of pedestrians in the crossover area then is 17 to 22, still within the range likely to maintain the service level above level-of- service D. Some block'age of the circulation corridor can therefore occur without violating the service target. The ratio of current observed pedestrian flows to achievable maximum flows for the target standards is 35 to 45 percent. This fraction of the 18-ft width of the circulation corridor represents the tolerable blockage by pas- sengers queued at security, a queue of 6.3 to 8.1 ft. Such a queue would constrict the flow of people in the circulation corridor and increase density to the limit allowed by the target. If queueing passengers occupy 2.5 to 3 ft per person, the tolerable queue length is no more than three to four people.

TERMINAL CIRCULATION 93 RESEARCH NEEDS Although research conducted to ensure that terminals are free of barners to the handicapped has contributed to knowledge about special circulation needs, there is still relatively little information on walking speeds for various types of passengers in airports. Research on these speeds is needed to develop service- level standards that can be readily adopted by a range of airports. Further, the frequency with which airports experience circulation problems indicates that this component warrants greater attention than it is now receiving. NOTE 1. Subsections correspond to Steps 3, 5, 7, and 8 in the assessment process shown in Figure 3-1. Attention may also be given to relevant relationships among components (Step 4) and community factors (Step 6). REFERENCES J. Fruin. Pedestrian Planning and Design. Cited in Special Report 209: Highway Capacity Manual. TRB, National Research Council, Washington, D.C., 1985. Canadian Airport System Evaluation: Vancouver Airport. Report AK-14-06-031. Transport Canada, Ottawa, Ontario, 1981. Special Report 209: Highway Capacity Manual. TRB, National Research Council, Washington, D.C., 1985.

10 Ticket Counter and Baggage Check Operation of the ticket counter and baggage check component begins when the passenger enters a queue to obtain a ticket and check his baggage and ends when that passenger leaves the ticket counter area. Curbside baggage check is a part of this component. The focus of this chapter is on activities located primarily within the terminal building. DESCRIPTION Airlines normally rent ticket counter space from the airport operator and manage this space on an exclusive-use basis. This leased area may include office space for administration and baggage handling. Airline personnel staff the ticket counters and operate according to procedures and standards set by the individual airline. The principal demand and operating factors influencing service level and capacity for the ticket counter and baggage check component are given in Table 10-1. In general, each airline establishes its own service standards for this compo- nent on the basis of company policy and local competition. Airlines may operate counter positions to segregate passengers by class of travel (e.g., first class, frequent traveler) and by service required (e.g., checking baggage only, purchasing tickets). Baggage handling facilities are at many airports the single largest airline component and may play a dominant role in facilities planning. In some airports a single queue may feed a bank of check-in positions. Service rates and preexisting queues determine the waiting time encountered by passengers in a peak period. Long queues do not necessarily indicate that 94

TICKET COUNTER AND BAGGAGE CHECK 95 TABLE 10-1 DEMAND AND OPERATING FACTORS INFLUENCING SERVICE LEVEL AND CAPACITY OF TICKET COUNTER AND BAGGAGE CHECK Factor Description Number and type of position Airline procedures and staffing Passenger characteristics Space and configuration Flight type, schedule, and load Airline lease agreement and airport management practices Processing rates are function of position type (baggage check only, ticket purchase, frequent or first class traveler, etc.) Number of positions manned and processing times Number preticketed or with boarding pass, amount of luggage, and distribution of arrival before scheduled departure influence demand loads, fraction of passengers by-passing check-in Available waiting area for queues approaching agent positions; banked or separate queues; conflict with circulation patterns Basic determinant of number of people arriving at ticket Counter use policy, as formalized in lease agreements, similar to gate issues and options the component has a capacity problem. The length of time required to go through the line and the amount of waiting room available for each person in the queue are generally the minimum criteria for judging service levels. DEMAND PATFERNS Passenger demand is determined primarily by scheduled flight departure times, types of aircraft, and load factors. Length of arrival time before a scheduled departure may be expected to vary by type of service offered and by size of airport. Some airports survey departing passengers to develop repre- sentative distributions of such arrival times; an example determined in 1985 by the Port Authority of New York and New Jersey is shown in Figure 10-1. Surges of demand may occur within the peak period for the airport as a whole, particularly during arrivals of groups of passengers from airport buses or other high-occupancy ground transport vehicles. Average demand condi- tions over the course of a peak hour are generally an appropriate basis for making judgments about service and capacity. However, airline or ground access operations may create sharp variations in passenger demand, making it more appropriate at some airports to use shorter periods such as 15 to 30 mm. Increased use of advance ticketing and seat assignment has raised the fraction of passengers bypassing the ticket counter and baggage check compo- nent. At New York City's major airports, the fraction of passengers able to

96 MEASURING AIRPORT LANDSIDE CAPACiTY 100 90 80 70 60 50 40 30 20 10 0 0 40 80 120 160 200 240 280 MINUTES BEFORE SCHEDULED DEPARTURE FIGURE 10-1 Observed departing-passenger arrival times at John F. Kennedy International. avoid stopping at the ticket counter has been observed to vary from almost none on international flights, for which passports must be checked, to more than 40 percent on airlines serving primarily business travelers (1). OPERATING CHARACTERISTICS Data from several U.S. large hub airports show that average processing or contact time per passenger at ticket counters varies widely. A sample of experience at several larger airports showed times in a range of 1.4 to 5.6 mm (Table 10-2). Typical processing time at express check-in counters in this sample is 50 to 70 percent of that for full-service counters. Processing times at any particular airport will depend on airline staff experience, flight market, and passenger characteristics, as well as on airline operating policies. Surveys are typically required to determine these times. Passenger processing times may vary substantially around the average. In Vancouver, Canada, for example, the average processing rate for a bank of ticket counters was observed to be 23 passengers per hour, or approximately 2.6 min per passenger. However, detailed observation showed that 60 percent of the passengers were served in less than 2.7 min each, and for 20 percent of the passengers more than 4 min was required (3).

TICKET COUNTER AND BAGGAGE CHECK 97 TABLE 10-2 TYPICAL PROCESSING TIMES OBSERVED AT TICKET COUNTER AND BAGGAGE CHECK (1,2) Typical Service Time Airport (mm) Miami (1) Full service 1.9-5.6 Express 2.3 Denver (1) Full service 1.8-3.9 Express 1.5-2.7 La Guardia (1) Full service 2.8-5.5 Express 1.2-3.1 La Guardia (five concourses)a 1.4-2.6 Kennedy (eight unit terminals)a 1.4-4.0 Manual ticketing (3) With baggage 3.0-4.0 Without baggage 1.7-3.3 Baggage only (3) 0.5-0.8 Automated ticketing (3) With baggage 2.7-3.7 Without baggage 1.5-3.0 °Suey by Port Authority of New York and New Jersey. TABLE 10-3 SPACE STANDARDS FOR TERMINAL CHECK-IN AREAS (4,5) Space Standard Source (ft2/person) LATA level of service (4) Level A (excellent) > 17.2 Level E (inadequate) < 10.8 for> 15 mm System breakdown <8.6 for> 15 mm FAA implied guidelines (5) Multipurpose check-in 15-23 Baggage check only 12-18 Ticketing only 4.3-7.6 Nom: IATA = Intemational Air Transport Association. Data are not generally available on actual crowding conditions in check-in areas of U.S. airports. Frequently applied architectural and planning space standards are summarized in Table 10-3. A guideline of 8 ft2 per person, for example, allows approximately a 3-ft separation between passengers in a queue. Typical airport design standards call for a queueing space 15 ft deep in front of ticket counters (5) and specify different spacings between service

98 MEASURING AIRPORT LANDSIDE CAPACITY positions to allow for different functions. However, because airports differ with regard to geometry and services, there are many different passenger queueing patterns. ANALYSIS TOOLS AND ASSESSMENT STANDARDS Capacity of the ticket counter and baggage check component is judged by considering the average time required for processing passengers and by comparing number of passengers in the terminal lobby queueing area with the size of that area. Direct observation or simple queueing analysis is often used to estimate both the average wait time and the number of people waiting. Observations in several large East Coast airports indicate that queue lengths of 8 to 10 persons may trigger the opening of additional counter positions, when available, by short-haul domestic service airlines. Queues for interna- tional departures are typically permitted to grow longer. Such queues repre- sent delay and possibly crowding for the passenger. A survey at Birmingham International Airport in England (Table 104) confirms that the length of wait time that passengers find tolerable varies substantially with the market being served. Passengers for scheduled Euro- pean flights were satisfied with times of 7.5 min or less and found times of 14 min or greater "intolerable." Passengers for scheduled long-haul flights— mainly vacationers—were willing to tolerate wait times almost twice those of the short-haul market. Charter passengers were willing to tolerate waits in the range of 11 to 21 mm (4). EXAMPLE OF ASSESSMENT PROCESS' Assume that there is a single ticket counter, manned by one agent, as shown in Figure 10-2. TABLE 104 VARIATION OF TOLERABLE WAIT TIME AT CHECK-IN FACILITIES FOR BIRMINGHAM INTER- NATIONAL AIRPORT (6) Wait Time (mm) by Passenger Rating Market Segment Good to Tolerable Good to Bad Scheduled European 7.5 14 Scheduled long haul 15 25 Charter 11 21

TICKET COUWTER AND BAGGAGE CHECK 99 Describe Component The physical dimensions of the terminal are such that the counter position has a queue wait- ing area of 60 ft2 and is equipped to handle baggage and full ticketing of passengers. dff Describe Demand and Operating I Factors --- Suppose that passengers arrive at the counter 6 ft -area unaccompanied, so only passengers with few pieces of luggage are waiting in the queue. FIGURE 10-2 Example of ticket counter and baggage Passengers arrive at the counter during a peak check. hour according to the pattern shown in Figure 10-3. During the first 15 mm, six passengers arrive at the counter at fairly uniform intervals of 2.5 min apart. The arrival rate then increases, so that by the end of the first half-hour, 10 more people have arrived, a total of 16. All peak-hour pas- sengers, a total of 20, have arrived by the end of 55 mm. No passenger arrivals are expected during the final 5 min of the peak hour. An experienced agent staffs the counter. In spite of the diverse demands placed on the agent, an average service time of approximately 3 min per passenger is maintained during the peak period. Estimate Service Levels and Service Volumes Maximum average throughput of the counter is estimated as follows: Throughput = (peak-period time)/(expected service time) = (60 min)/(3 mm/passenger) = 20 passengers/hr This estimate suggests that all passengers can be served during the hour. However, during the first 15 min of the peak period, six passengers arrive at a steady average rate. Only five are served and leave the counter, so there is, on average, someone at the counter throughout the period. Thus queues begin to grow as passengers arrive more quickly during the next 15 mm. At the end of the first 30 mm, when 16 people have arrived and only 10 have been served, the queue reaches its maximum length of 6 in the queue area, and the waiting time increases to 18 min for the sixth passenger (five passengers

100 MEASURING AIRPORT LANDSIDE CAPACiTY 20 ahead at 3 min per passenger plus 3 min to serve the sixth). The six passengers share the 15 rDIstribution ~rmu queue area in front of the counter-60 ft2- 4 10 with 10 ft2 per person. Queue length declines during the rest of the hour. 5 If the competitive stance of the airline using this counter requires relatively high levels of 1 hour service—for example, that passenger time at FIGURE 10-3 Distribution the counter (including waiting in queue) not of passenger arrivals at exam- exceed 12 mm—then a second counter posi- ple counter. tion may be required. Further, suppose that the airline believes that passengers waiting for ser- vice should not be so crowded that they have less than 10 ft2 per person. This space standard is violated when more than six people are in the queue. At a service time of 3 min per passenger, only four people can be in the queue if a maximum wait time of 12 min is to be maintained. Some passengers will have to be served at another counter if an acceptable service level is to be maintained. The time standard becomes effective with the twelfth passenger and remains a con- straint until the arrival rate declines and allows the agent at the counter to catch up. Two more passengers arrive during this period and cannot be served. The achievable service volume is then only 18 passengers.in the peak hour at the target service level and subject to the given demand pattern. RESEARCH NEEDS Because airline practices vary so much, limited data are available on service times and on how counter and queue configuration can improve service times. More information in this area would give airlines and airport management a basis for assessing space needs and for working together to ensure efficient use of available terminal lobby areas. NOTE 1. Subsections correspond to Steps 3, 4, 7, and 8 in the assessment process shown in Figure 3-1. Attention may also be given to relevant relationships among components (Step 4) and community factors (Step 6).

TICKET COUNTER AND BAGGAGE CHECK 101 REFERENCES P. Mandle, F. LaMagna, and F. Whitlock. Collection of Calibration and Validation Data for an Airport Landside Dynamic Simulation Model. Report TSC-FAA-80-3. Federal Aviation Administration, U.S. Department of Transportation, April 1980. R. Horonjeff, and F. McKelvey. Planning and Design of Airports, 3rd ed. McGraw- Hill, New York, 1983. Air Terminal Systems Capacity/Demand Study—Vancouver International Airport. Transport Canada, Ottawa, Ontario, Aug. 1986. Guidelines for Airport Capacity/Demand Management. Airport Associations Coor- dinating Council, Geneva, Switzerland, and Jnternational Air Transport Associa- tion, Montreal, Canada, Nov. 1981. Ralph M. Parsons Co. The Apron and Terminal Building Planning Report. Report FAA-RD-75-191. Federal Aviation Administration, U.S. Department of Transporta- tion, July 1975. S. Mumayiz and N. Ashford. Methodology for Planning and Operations Manage- ment of Airport Terminal Facilities. In Transportation Research Record 1094, TRB, National Research Council, Washington, D.C., 1986, pp. 24-35.

11 Terminal Curb Most passengers, their baggage, and sometimes accompanying visitors are dropped off or picked up at the terminal building curb frontage. In this area passengers leave ground transportation (automobile, taxi, bus, limousine, or courtesy van) and become pedestrians on their way to or from the aircraft gate. DESCRIPTION A variety of pedestrians, private automobiles, taxis, buses, commercial deliv- ery trucks, and hotel and rental car courtesy vans use the terminal curb area. Passengers may be carrying luggage to or from the terminal building, check- ing luggage at curbside facilities, and waiting for access to taxis or other vehicles. At some airports passengers must cross frontage roads to reach parking areas from the terminal curb, slowing vehicular traffic circulation. The principal demand and operating factors influencing service level and capacity for the terminal curb are summarized in Table 11-1. The primary determinant in the amount of curb frontage space required at a terminal is the length of time that vehicles stop for loading and unloading, referred to as the dwell time. For example, terminal designers estimate that reducing the aver- age dwell time of those automobiles and taxis stopping at the terminal curb from 120 to 90 sec can increase the capacity of a curb area by 15 to 20 percent (1). Dwell time is influenced by whether drivers stop and leave their vehicles to accompany passengers into the terminal building or to meet or find an arriving passenger. Airports may seek to limit dwell times and overall conges- tion in the curb area through enforcement of regulations on access and by using signing and traffic management to separate user groups having substan- tially different demand characteristics. As an airport grows, parking and leaving vehicles along terminal curbs may be prohibited and efforts to channel 102

TERMINAL CURB 103 TABLE 11-1 DEMAND AND OPERATINC FACTORS INFLUENCING SERVICE LEVEL AND CAPACITY OF TERMINAL CURB Factor Description Available frontage Length of curb frontage modified by presence of obstructions and assigned uses (e.g., airport limousines only, taxi only), separation of departures and arrivals Frontage roads and pedestrian Number of traffic lanes feeding to and from paths frontage area; pedestrians crossing vehicle traffic lanes Management policy Stopping and dwell regulations, enforcement practices, commercial access control, public transport dispatching Passenger characteristics and Passenger choice of ground transport mode, motor vehicle fleet mix average occupancy of vehicles, dwell times at curb, passenger patterns of arrival before scheduled departure, baggage loads Flight schedule Basic determinant of number of people arriving and departing at given time in given area automobile traffic directly to parking areas may be made. At large airports already practicing strict enforcement of traffic regulations, physical design changes often are the principal means for dealing with curb congestion. However, curb access is in general a major policy issue, requiring a balance among concerns of commercial operators and private automobile users, access service levels, and safety. The policy issue at many airports extends well beyond the technical questions of curb frontage design and vehicle demand characteristics. DEMAND PATTERNS Because private automobiles are the dominant ground access mode at most airports, they are the principal source of terminal curb frontage demand. Such demand can be reduced at some airports by increasing availability of conven- ient parking, which typically raises the proportion of motorists who enter or exit parking areas directly without stopping at the curb frontage, or by encouraging passengers to use off-airport check-in facilities if these are available. Demand for curb frontage is also determined by flight schedules and particularly by the arrival pattern of originating passengers (how far in advance of the scheduled departure time they arrive at the airport) and the route through the terminal of terminating passengers (how long it takes them

104 MEASURING AIRPORT LANDSIDE CAPACITY to travel from an arriving flight to the curb). Type of flight and trip purpose also influence terminal curb demand. For example, originating passengers on international flights are requested to arrive at the airport earlier than those aboard domestic flights. Terminating international passengers also are typ- ically slower than domestic passengers to reach the curb frontage because of required customs and immigration procedures. Passengers on business trips arrive at the airport closer to their departure time than those traveling for recreation or vacation. Deplaning business passengers, who may carry all their baggage aboard an aircraft and thus not need to stop at the baggage claim, reach the curb frontage sooner than those deplaning passengers who have checked bags. Transfer passengers at some airports use buses operating on frontage roads and thus also contribute to the demand on terminal curb facilities. The curb frontage demand resulting from shuttle buses and courtesy vans may be related to the number of trips per hour they make to the terminal and not directly to number of passengers. The operators of these vehicles, seeking to ensure that all passengers are picked up promptly and reliably, may provide frequent service operated on specific headways and allow some vehicles to be underutilized in order to reduce waiting time for their patrons. Vehicle dwell time varies with type of vehicle, number in the vehicle, and baggage loads of passengers. Dwell times for originating (enplaning) pas- sengers, as shown in Table 11-2, are typically shorter than those for terminat- ing (deplaning) ones. OPERATING CHARACTERISTICS An airport operator may influence curb frontage operations primarily through traffic management, development and enforcement of regulations on access to and use of the terminal curb, and minor modifications of the basic design of the terminal curb frontage. For example, passenger arrivals and departures may be physically separated. Some terminals are designed initially or are retrofitted with two levels, an enplaning area on one level and deplaning on the other level. Some airports use signing on frontage roads to segregate arrivals and departures; vehicles meeting arriving passengers are concentrated in the vicinity of building exits and close to baggage claim areas. Different types of vehicles may also be segregated: commercial vehicles (buses, taxis, courtesy vans, and scheduled limousines) may be separated from private automobiles. Enforcement of regulations limiting vehicle dwell times in curb frontage areas influences traffic congestion, curb service levels, and capacity.

TERMINAL CURB 105 TABLE 11-2 OBSERVED CURB DWELL Tl14ES AT SELECTED AIRPORTS (2-4) Airport Average Dwell Time (mm) Miami International (2) Enplaning 1.6-4.5 Deplaning 2.3-4.5 Denver Stapleton (2) Enplaning 1.2-2.8 Deplaning 4.8-6.9 La Guardia, main terminal (2) Enplaning 1.0-1.6 Deplaning 2.1-4.8 Suggested representative Statistics (3) Automobile Enplaning 1-3 Deplaning 2-4 Taxi Enplaning 1-2 Deplaning 1-3 Limousine Enplaning 2-4 Deplaning 2-5 Bus Enplaning 2-5 Deplaning s—io John F. Kennedy International (deplaning only, two unit terminals) (4) Automobile i .2-1.9 Taxi 0.4-1.3 Greater Pittsburgh International (deplaning or enplaning) (4) Automobile 1.0-2.4 Taxi 0.8-1.3 ANALYSIS TOOLS AND ASSESSMENT STANDARDS Analyses of the terminal curb sometimes use procedures adapted from traffic engineering. Simplified planning models such as that developed by Mandle et al. (5) relate traffic volumes and service levels to available curb length appropriately corrected for specific operational and physical characteristics of the airport. Service levels are analogous to those defined in TRB's Highway Capacity Manual (6). The Canadian government uses similar analysis pro- cedures to calculate an effective curb utilization rate reflecting the estimated fraction of time that the available curb is occupied (7). A primary element of terminal curb service level is the motorist's ability to find a space for loading or unloading. The probability of a motorist's finding

106 MEASURING AIRPORT LANDSIDE CAPACITY an empty curb space or having to double park is typically used to describe service level, although other parameters such as general traffic congestion may be used as indicators of this probability. During the busiest periods some degree of double parking is considered acceptable at many airports. It is important to note that the capacity of the terminal curb lane is distinct from the capacity of the travel lanes adjacent to it. These travel lanes are part of the ground access component (Chapter 13). EXAMPLE OF ASSESSMENT PROCESS' Suppose that frequently recurring congestion in front of the terminal building at a medium hub airport (served by four airlines) has caused consideration of a new roadway design to increase available curb frontage. Before hiring a consultant, management would like to know whether service levels can be improved and new construction avoided by better utilization of existing facilities. Describe Component The single-level terminal building is 400 ft long with a sidewalk and a loading and unloading lane adjacent to a two through lanes (Figure 11-1). Extensions of sidewalk beyond the ends of the terminal building bring the total curb frontage to approximately 460 ft. The terminal building has three entrances spaced at intervals of approximately 100 ft, and serving both arriving and departing passengers. Describe Demand and Operating Factors There are 16 scheduled aircraft operations (arrivals and departures) during the peak periods with total seat capacity of approximately 1,700 passengers, and average load factors are 62 to 78 percent. Approximately 85 percent of the passengers originate or terminate their journeys during these periods. Depar- tures represent approximately 55 to 60 percent of the passengers during the morning early in the week and during the afternoon in the latter part of the week. Arrivals represent a similar majority during other daily peak hours. The airport operator has never surveyed passenger ground transportation characteristics, but a staff analyst estimates that 90 to 95 percent travel by private automobile, and 3 to 6 percent by rental car. There is taxi service but virtually no airport van or bus service. The curb frontage is unregulated with respect to passenger loading and unloading. Passengers try to stop as close as possible to the door serving their

400 ft Terminal Building lOon - I 100ft---j-I_.___100ft -I-_ lOOft — — ------ Traffic Flow Approximately 460 ft FIGURE 11-1 Example of terminal curb component (drawing not to scale).

108 MEASURING AIRPORT LANDSIDE CAPACiTY airline. Drivers awaiting arriving passengers typically park along the curb and even leave their vehicles, although there is a traffic regulation prohibiting parking. Eight to 12 cars may be parked in this way on a typical Thursday or Friday afternoon. A few commercial delivery vehicles may also be seen in the traffic stream and at the curb. Estimate Service Levels and Service Volumes Using the figures available, total peak-hour passenger load is estimated: Total passenger load = (seats) x (load factor) x (1 - percentage of transfers = (1,700) x (62 to 78 percent) x (85 percent) = 900 to 1,130 passengers in the peak hour With a directional split of 55 to 60 percent in the peak hour, the estimate is carried further: Enpianing passengers (or deplanements) = (55 to 60 percent) x (900 to 1,130) = 500 to 680 Deplaning passengers (or enpianements) = (40 percent to 45 percent) x (900 to 1,130) = 360 to 510 Because this is a medium hub airport and so heavily automobile-oriented, the average number of passengers per vehicle is assumed to be low, about 1.05. Noting that the parking lots at the airport are seldom full and observing that many passengers are dropped off or picked up by family members, the analyst estimates that approximately 70 percent of the vehicles going in the peak direction use the curb. Using definitions of service level as adopted by Mandle et al. (5), the analyst observes that level-of-service D or E appears to occur during the morning and afternoon peak hours 3 to 5 days each week and operational problems are frequent. Management would like to ensure that conditions are stable for all but the busiest times at level-of-service C or better.2 Using the procedure developed by Mandel et al. (5), curb requirements may be estimated for the situation just described. This procedure indicates that with best-case base estimates of service volume of 500 to 680 enplaning pas- sengers, approximately 240 to 300 ft of curb frontage is needed to maintain level-of-service C (Figure 11-2).

REQUIRED DEPLANING CURB LENGTH (feet) 0 240 480 720 960 1200 14401680 1920 2160 2400 2640 2880 3500- •/: 3000 - / 2500 - 2000 1500 LEVEL A - NO TRAFFIC QUEUES, NO DOUBLE PARKING LEVEL B - EFFECTIVE CURB UTILIZATION EQUAL TO 1.1 TIMES ACTUAL CURB FRONTAGE LEVEL C - EFFECTIVE CURB UTILIZATION EQUAL TO 1.3 TIMES ACTUAL CURB FRONTAGE LEVEL 0 - EFFECTIVE CURB UTILIZATION EQUAL TO 1.7 TIMES ACTUAL CURB FRONTAGE 1000 - 2' LEVEL E - OPERATIONAL BREAKDOWNS. EFFECTIVE CURB UTILIZATION EQUAL TO 2.0 TIMES 500 - ACTUAL CURB FRONTAGE NOTE: Values presented reflect typical characteristics Of airport activities. 0 1 1 1 I 1 1 I 1 I I I 0 200 400 600 800 1000 12001400 1600 1800 200022002400 REQUIRED EMPLANING CURB LENGTH (feet) Excludlng Connecting Passengers FIGURE 11-2 Suggested method for estimating curb frontage needs (5).

110 MEASURING AIRPORT LANDSIDE CAPACiTY Although the deplaning passenger level is less than the lower limit of the graph, a rough estimate might be made that for 360 to 510 deplaning pas- sengers 200 to 240 ft of curb frontage is needed. The total base estimate in Figure 11-2, 440 to 540 ft, might be increased 25 to 35 percent to account for low average vehicle occupancy. On balance, the analyst might conclude that the airport should have 550 to 700 ft of curb to maintain level-of-service C. Reversing the same procedure, the model confirms that level-of-service E or worse can be expected given the combination of estimated demand patterns and current curb frontage. However, no consideration was given in the analysis to the impact of parked vehicles, which at times fill 40 percent of the curb frontage. Given the uncertainty and level of detail of the analysis, the estimated terminal curb requirements are not so much greater than what is available (at- worst, 50 percent greater need than availability), and because conditions are observed to be in the level-of-service D to E range, management might justifiably expect that enforcement of the existing traffic regulations might bring the service level up to target. Such a solution might be worthy of a trial while data are gathered to support consideration of more costly construction alternatives. RESEARCH NEEDS The levels of understanding and assessment procedures for the terminal curb are well developed in comparison with those for other components of the landside. Although models of the sort used in the preceding example are helpful and might warrant further expansion, the underlying principles are based on traffic analysis and are less in need of research. Nevertheless, professionals in the field have observed that local idiosyncrasies of driver behavior and roadway geometry can have substantial impact on curb perfor- mance, and that operational changes can relieve many problems. There may then be benefits to increased publication of good solutions to existing prob- lems, which may suggest what might work for other airports with similar curb frontage situations. NOTES Subsections correspond to Steps 3, 5, 7, and 8 in the assessment process shown in Figure 3-1. Attention may also be given to relevant relationships among compo- nents (Step 4) and community factors (Step 6). Other examples of assessment have been published by Tiles (4), Mandle et al. (5), and Hart (8). Service levels are defined generally in the Highway Capacity Manual (6) as follows:

TERMINAL CURB 111 C—Stable flow, but the beginning of the range of traffic flow in which operation of individual users becomes significantly affected by the presence of others and maneuvering within the traffic stream requires substantial vigilance on the part of the user. The general level of comfort and convenience declines noticeably at this level. D—High-density but stable flow. Speed and freedom to maneuver are severely restricted. Small increases in traffic flow will generally cause operational problems. E—Freedom to maneuver within the traffic stream is extremely difficult. Comfort and convenience levels are extremely poor. Operations at this level are usually unstable because small increases in flow or minor perturbations within the traffic stream will cause breakdowns. REFERENCES F. LaMagna, P. Mandle, and E. Whitlock. Guidelines for Evaluating Characteristics of Airport Landside Vehicle and Pedestrian Traffic. In Transportation Research Record 732, TRB, National Research Council, Washington, D.C., 1979, pp. 54-61. P. Mandle, F. LaMagna, and E. Whitlock. Collection of Calibration and Validation Data for an Airport Landside Dynamic Simulation Model. Report TSC-FAA-80-3. Federal Aviation Administration, U.S. Department of Transportation, April 1980. F. X. McKelvey. Access to Commercial Service Airports. The Planning and Design of On-Airport Ground Access System Components. Final Report. College of Engi- neering, Michigan State University, East Lansing, June 1984. R. Tilles. Curb Space at Airport Terminals. Traffic Quarterly, Oct. 1973. P. Mandle, E. Whitlock, and F. LaMagna. Airport Curbside Planning and Design. In Transportation Research Record 840, TRB, National Research Council, Wash- ington, D.C., 1982, pp. 1-6. Special Report 209: Highway Capacity Manual. TRB, National Research Council, Washington, D.C., 1985. A Discussion Paper on Level of Service Definition and Methodology for Calculat- ing Airport Capacity. Transport Canada, Airport Services Branch, Ottawa, Ontario, April 2, 1979. W. Hart. The Airport Passenger Terminal. John Wiley, New York, 1985.

12 Parking Area Parking areas consist of surface lots or multilevel garages used to store the vehicles of air passengers and visitors. Although parking and storage areas are also needed for employee vehicles, rental cars, taxis, and buses, these require- ments have relatively little influence on the capacity or service level of the airport as viewed by a passenger. The focus of this section is therefore public parking. DESCRIPTION For planning purposes, parking is divided into two or three general categories: short-term, long-term, and remote (which is usually long-term parking). Short-term parking is usually located close to terminal buildings and serves motorists dropping off or picking up travelers. These motorists usually remain at the airport for less than 3 hr. Some airports distinguish other periods of time as "short term." The most expensive parking is often found in the short-term lots. Long-term parking serves passengers who leave their vehicles at the airport while they travel. With the low turnover rate and long duration of stay, long- term parking accounts for 70 to 80 percent of the occupied parking spaces at an airport, even though only 15 to 30 percent of all vehicles parked over the course of a year is represented (1). Remote parking consists of long-term parking lots located away from the airport terminal buildings. Often buses or vans may be available to transport passengers to the terminal. At some airports these parking facilities are called 112

PARKING AREA 113 shuttle lots or, when used only during peak periods, holiday lots. Because parking rates at remote lots are less expensive than those for other airport parking facilities, these lots are often termed "reduced-rate" parking. Entry to airport parking areas is usually controlled by automatic gates, and parking fees are collected by a cashier located in a booth at the exit. At some airports motorists leaving the parking facilities are delayed because there are insufficient cashiers. In addition to delays encountered entering and leaving parking areas, passenger service levels will also be affected by the distance to be walked between parked vehicles and the terminal and by the environment in which this walk occurs. At some airports weather-protected walkways improve passenger assessment of the service level. Other airports provide escalators, moving sidewalks, buses, people movers, or other mechanical assistance to reduce passenger walking distances and number of level changes. The principal demand and operating factors influencing service level and capacity for parking areas are given in Table 12-1. Although the number of parking spaces provided is a snapshot measure of capacity, the total time required for access and egress influences the assessment of service level provided. Such factors as the allocation of parking between short- and long- term uses and parking prices influence service levels. DEMAND PAFERNS Demand at parking areas is characterized by accumulation of parked vehicles, which is measured by both length of time as well as number of parking spaces occupied. This accumulation is influenced by passenger arrival times at the airport and trip purposes. Bisiness travelers have been found to be less sensitive to parking costs than those on pleasure trips and thus more likely to use more expensive, close-in, short-term parking. Passengers on vacation are more cost sensitive but value their time less. They are more likely to seek reduced-rate, remote parking areas. Passengers who expect to be away for long trips have greater amounts of baggage and are more likely to be dropped off or greeted by motorists who use short-term parking. Parking demand is in general very sensitive to the cost of parking. Effective parking capacity can be increased by altering parking fees to increase the cost difference between short-term and long-term parking and thus encourage price-sensitive motorists to divert to less expensive parking areas. General increases in parking fees may also encourage passengers to choose other means than driving their own automobiles for their trip to the airport. Parking space needs are determined primarily by the amount of long-term parking, because it generates the most space-hours. Because of their higher

114 MEASURING AIRPORT LANDSIDE CAPACiTY TABLE 12-1 DEMAND AND OPERATING FACFORS INFLUENCING SERVICE LEVEL AND CAPACiTY OF PARKING AREAS Factor Description Access (enpianing) Available space Access times Passenger characteristics Pricing Flight schedule Egress (deplaning) Access time Exit position and employee efficiency Passenger characteristics Flight schedule and load As a function of distance from terminal area, systems for reaching terminal, prices for parking, and availability of weather-protected waiting and walking areas Total, including search for space, wait, and travel from remote locations Percentage of people driving, automobile occupancy, visitor ratios, length of stay Higher fees may suppress demand or divert some to lower-cost lots Basic determinant of number of people arriving at parking areas Total, including wait and travel to remote locations, with consideration for availability of weather-protected wait and walk areas Number and direction to exits, service times to exit lots Fraction driving, automobile occupancy, length of stay Basic determinant of number of people arriving at parking areas turnover rate, short-term spaces generate more entry and exit movements than do long-term spaces. Demand for automatic gates at the entry and cashier booths at the exit of the parking area is a function of the number of entries and exits, which, although not indicative of space needs or parking demand, may be significant for capacity estimation. The number of parking spaces required to provide adequate service levels is normally greater than total parking demand. This is because at a large parking facility in which many areas cannot be seen simultaneously, for example, in a multilevel garage or extensive open lot, it is more difficult to find the last empty spaces. Thus a large parking facility may be considered full when 85 to 95 perëent of the spaces are occupied, depending on its use by long- or short- term parking, size, and configuration. OPERATING CHARACTERISTICS The balance between long- and short-term parking is critical to capacity estimation. As already noted, distribution of demand may be significantly

PARKING AREA 115 influenced by parking fees. For example, when New York City's LaGuardia was faced with a serious shortage of parking space, structured parking facili- ties were constructed, and all parking rates were raised significantly to sub- stantially reduce long-term parking at the airport and make space available to satisfy short-term demand. Such a strategy generally requires that adequate alternative transport services be available to meet the demand for airport ground access. Besides determining number of parking spaces available and the effective distance of spaces from terminal buildings, parking operators may be able to adjust the physical design of the parking lot entries and exits to reduce or avoid congestion and use informational signing to direct motorists to under- utilized facilities. However, lengthening the pedestrian path between the parking lot and the terminal may reduce overall perceived service levels. Some airports are experimenting with "pre-cashiering," in which passengers pay their parking fees before entering their vehicles, as a means to reduce delay and congestion at exit plazas. ANALYSIS TOOLS AND ASSESSMENT STANDARDS A variety of rules of thumb and computational procedures have been de- veloped to estimate parking space requirements at an airport. Indexes of spaces per annual or average monthly enplaned passenger are frequently used in planning: Source Index Roads and Transport Association 1.5 spaces/peak-hour passenger; of Canada (2) (smaller airports) 900 to 1,200 spaces/million annual enpianed passengers FAA (3) (non-hub airports) Approximately 1 space/500 to 700 annual enpianed passengers Because transfer passengers do not create parking demand, indexes based on originating passengers only (Figure 12-1) are likely to be most meaningful. Used together with local experience or rules of thumb relating annual pas- senger loads to total parking demand, such indexes may be used to judge the likelihood that the number of parking spaces available at an airport is appro- priate to current or anticipated levels of demand. However, such calculations do not give true service-level indicators.

116 MEASURING AIRPORT LANDSIDE CAPACITY ,, 6,000 w 0 CL 4 5,000 C, z 4,000 4 0. 3,000 0 o 2,000 4 1,000 CL 0 .4 .8 1.2 1.6 2.0 2.4 2.8 3.2 3.6 4.0 ORIGINATING PASSENGERS (millions) FIGURE 12-1 Estimated requirements for public parking at U.S. airports (4). - Mathematical simulation has been used to model parking accumulation (1). Detailed information on time of arrival and departure of automobiles relative to scheduled flight times and duration of stay relative to parking cost and distance from the terminal are generally required. Trip purpose and traveler income data may also be useful for estimating sensitivity of passenger ground transport mode and parking location choice to parking fees. Useful data and analysis procedures may be available through the local transportation agency. Peak-period conditions at entry and exit of the parking lot may be observed in field surveys or estimated as queueing problems. Likely levels of demand at cashier exits may be forecast from flight arrival schedules, with assumptions about the time it takes passengers to reach their vehicles after a flight has landed. Time required for passengers to locate a parking space and walk or ride to or from the terminal building and the conditions under which passengers must make this trip are important elements of service level. Typical parking area travel and service time parameters at a number of airports are summarized in Table 12-2. However, actual field data are not available to compare these elements from a variety of airports. Some airports try to maintain a maximum walking distance of 800 ft between parking lots and airport terminal buildings and provide bus service for parking beyond that distance. Other airports might find this distance inappropriate.

PARKING AREA 117 TABLE 12-2 PARKING-AREA TIME PARAMETERS AT SELECTED AIRPORTS Parameter Time (mm) Walk from terminal to parking, BWIa 6.4 Walk, wait, and ride shuttle from terminal to parking, BWIa 8.6 Search for parking space following entry to lot, BWIa 1.4 Automatic ticket dispenser service (1) 0.15-0.16 Parking lot exit booth service Miami International (5) 0.5 Denver Stapleton (5) 0.5 New York La Guardia (5) 0.2-1.1 Typical (1) Single fee policy —0.24 Variable fee policy 0.5-0.6 Drive from space and exit, Tampa International° 4.5 Nora: BWI = Baltimore-Washington International. aReped by airport management. EXAMPLE OF ASSESSMENT PROCESS 1 Suppose that an airport with annual traffic of approximately 5 million pas- sengers expects a new airline to begin operations in the near future. Manage- ment is concerned that parking capacity may be insufficient to meet this demand and wants to be prepared to take corrective or preventive action. Describe Component The airport currently has a total of 2,200 parking spaces available and finds that lots close to the terminal building—for both short-term and long-term parking—are 75 to 90 percent filled between Tuesday afternoon and Thursday afternoon almost every week. Remote lots, which contain 600 spaces and are used primarily for long-term parking, are seldom more than 50 percent full. Describe Demand and Operating Factors Approximately 70 percent of annual enplaned passengers are originations. There have been no recent surveys of passenger access modes, but manage- ment estimates that perhaps 55 percent of the business-traveler market served by the new airline will choose, if possible, to drive and park at the airport and that the average duration of trips will be approximately 2 days.

118 MEASURING AIRPORT LANDSIDE CAPACiTY The new airline is expected to offer six flights a day, serve primarily originating passengers, and use aircraft with a seating capacity of 100 to 130 passengers. Long-term parking rates are currently set at $2.00 per hour or fraction with a daily maximum of $6.00. Short-term rates are $1.00 per 1/2 hr or fraction with a daily maximum of $24.00. Long-term remote lots charge $1.00 per hour or fraction with a daily maximum of $4.00 and are linked by bus to the main terminal. Estimate Service Levels and Service Volumes Because close-in long-term lots are nearly full midweek, there may indeed be a need for more parking. However, there are at least 300 spaces potentially available in the remote lots. In an effort to maintain the overall service level, the airport management would like to add capacity to accommodate demand at least above current levels and to maintain this condition as demand grows. Current midweek space needs are estimated as follows: Current space needed = (filled spaces) x (115 percent allowance for service level) = (2,200 - 300) x (115 percent) = approximately 2,200 spaces That is, current parking capacity is apparently just equal to demand given the desired service level. The parking supply of 2,200 spaces represents an average rate of 0.63 space per 1,000 enpianing passengers at the airport. If all new passengers are assumed to be originating or terminating their trips, 150,000 to 195,000 new enpianing passengers are expected annually. Maintaining the level of parking supply at this rate would then require 94 to 123 new parking spaces. Assuming that the new airline can achieve perhaps a 70 percent load factor, six flights a day with 100 to 130 seats per flight may represent new traffic of 300,000 to 390,000 passengers annually, most of whom are expected to be originating or terminating their trips. Because spaces are available only at remote locations and the new airline is serving business travelers, who are normally willing to pay premium prices to park, consideration might be given to raising parking fees at nearby lots. The increased fees might shift some travelers to other access modes (perhaps putting some pressure on terminal curb facilities), but probably would im- prove utilization of remote lots and generate additional revenue.

PARKING AREA 119 RESEARCH NEEDS Additional data on parking access, search, and time and on times and dis- tances passengers must travel from parking areas to terminal buildings would be useful to support service level assessments. Research on the degree to which parking service levels and cost can mu uence choice of modes for travel to the airport would also be useful for capacity management. NOTE 1. Subsections correspond to Steps 3, 5, 7, and 8 in the assessment process shown in Figure 3-1. Attention may also be given to relevant relationships among compo- nents (Step 4) and community factors (Step 6). REFERENCES F. X. McKelvey. Access to Commercial Service Airports. Final Report. Federal Aviation Administration, U.S. Department of Transportation, June 1984. Guide for the Planning of Small Airports. Roads and Transport Association of Canada, Ottawa, Ontario, 1980. Planning and Design of Airport Terminal Facilities at Non-hub Locations. Ad- visory Circular 150/5390-9. Federal Aviation Administration, U.S. Department of Transportation, April 1980. Ralph M. Parsons Co. The Apron and Terminal Building Planning Report. Report FAA-RD-75-191. Federal Aviation Administration, U.S. Department of Transporta- tion, July 1975. P. Mandle, F. Lalvlagna, and E. Whitlock. Collection of Calibration and Validation Data for an Airport Landside Dynamic Simulation Model. Report TSC-FAA-80-3. Federal Aviation Administration, U.S. Department of Transportation, April 1980.

13 Ground Access The roadways for access and circulation and public and private transit—both on and off the airport—make up the ground access system. When available, off-airport passenger check-in facilities or downtown tenmnals form part of this system. Typically only those off-airport elements of ground access that serve significant volumes of airport traffic are considered in planning and analysis. For example, a particular intersection far from the airport may be a constraint for one passenger but will have no material impact on airport operations. For many airports the access system extends only to the nearest interchange or intersection, whereas at those airports where most passengers arrive via a single primary route (such as the Sumner-Callahan tunnels in Boston or the expressway leading to Dulles International in Washington, D.C.) the access system may include many miles of a particular roadway or transit line. DESCRIPTION Ground access is provided by an assortment of private and public transport modes. Except in those few cases where a rail transit system serves the airport, these ground access modes all use the metropolitan highway and street network and share the same roadways for circulation at the airport. For the three airports shown in Table 13-1, as at most U.S. airports, private auto- mobiles and taxis are the principal access modes used. Those accompanying or meeting passengers influence the demand on ground access systems. Such individuals overwhelmingly travel by private 120

GROUND ACCESS 121 TABLE 13-1 GROUND ACCESS MODES OF PASSENGERS AT SELECTED AIRPORTS (1) Percent Choosing Mode by Type of Passenger Originating Terminating Access Mode Miami Denver La Guardia Miami Denver La Guardia Private automobile 42 56 25 47 70 31 Rental car (van) 11 14 9 20 8 4 Taxi 22 14 46 18 10 35 Limousine 10 5 13 10 5 20 Bus 15 3 5 5 5 5 Other 0 9 2 0 3 5 NoTE: Percentages may not add to 100 because of rounding. TABLE 13-2 GROUND ACCESS MODES OF THOSE ACCOMPANYING PASSENGERS AT SELECTED AIRPORTS (1) Percent Choosing Mode by Type of Passenger Originating Terminating Access Mode Miami Denver La Guardia Miami Denver La Guardia Private automobile 99 80 82 84 97 90 Rental car (van) 1 - - 6 2 1 Taxi - 7 9 5 1 5 Limousine - - 9 3 - I Bus - 7 - 2 1 1 Other - 7 - - - 3 Nom: Percentages may not add to 100 because of rounding. automobile, as do airport employees (Table 13-2). Additional vehicle trips result from the delivery of cargo, priority packages, mail and terminal build- ing and concession supplies and the numerous service and maintenance requirements of an airport. The peak hours for employee and other highway travel not related to the airport and the conditions on the on-airport roadways most heavily ued only by employees and airport operations differ from those for passenger traffic and are not addressed here. The principal demand and operating factors influencing capacity and ser- vice conditions for ground access are summarized in Table 13-3. Although it is often necessary to view many of these factors on a metropolitan scale, the focus of capacity assessment is on the service provided between the terminal curb or parking area and the interchange linking the airport with the regional transportation system.

122 MEASURING AIRPORT LANDSIDE CAPACITY TABLE 13-3 DEMAND AND OPERATING FACTORS INFLUENCING SERVICE LEVEL AND CAPACITY OF AIRPORT GROUND ACCESS Factor Description Available modes and prices Connections from various parts of the metropolitan area served, considering pnces, comfort, and conve- mence, particularly with respect to baggage and re- quired vehicle changes Access times Total, including wait for vehicles or access and travel from representative locations Passenger characteristics Fraction choosing each mode, vehicle occupancy, num- ber of people accompanying passenger, other visitors, baggage loads, origination/destination share Vehicle operator behavior Fraction going directly to curb or to parking, weaving, curb dwell time, knowledge of traffic patterns Flight schedule and load Basic determinant of number of people using ground facilities Facilities and background traffic Highway and transit routes, interchanges; levels of conditions traffic on facility for other than airport purposes; availability of remote check-in facilities DEMAND PATFERNS Access demand is primarily determined by the travel modes selected by passengers and visitors, the number of persons per vehicle (Table 13-4), the circulation patterns of these vehicles, and how long before or after a flight a person arrives at or leaves the airport. Demand patterns of courtesy vehicles and scheduled limousines and buses may not be directly related to air pas- senger activity patterns. Access demand is influenced by the extent of the public transportation system available, passenger trip purpose, the availability of parking, type of flight, and availability of alternative check-in areas. Cost of parking can have a particularly significant impact on access mode choice at large airports. Ground access demand is generated by both originating and terminating passengers. At airports with multiple terminal buildings the volume of con- necting passenger transfers may influence the need for terminal-to-terminal shuttle bus service. At other airports passenger transfers do not leave the terminal building and do not affect demand for either access or circulation. Circulation patterns are influenced by the location of the entrance and exit to parking and rental car areas, the availability and location of a recirculation road, the cost of short-term parking, and configuration of the terminals (single or multiple).

GROUND ACCESS 123 TABLE 13-4 TYPICAL AVERAGE VEHICLE OCCUPANCY RATES FOR AIRPORT GROUND ACCESS (2) No. of Passengers Access Mode per Vehicle Private automobile 1.9 Rental car 1.2 Taxi 2.5 Limousine 5.6 Other 4.2 OPERATING CHARACTERISTICS Driver familiarity with the roadway system and the complexity of the system significantly influence ground access operations. Complex road systems, such as those often found at large airports, require quick decision making by motorists unfamiliar with the airport and often involve frequent merging and weaving. Traffic control devices at at-grade intersections also influence sys- tem performance. The management of taxi, limousine, and courtesy bus operations may also influence ground access operations. Control of taxi entry to the terminal area, issuance of taxi permits for airport service, and encouragement of limousine services are among the actions taken by management at some airports to improve ground access conditions. Control of cargo vehicles and employee access are also important at some airports. ANALYSIS TOOLS AND ASSESSMENT STANDARDS The airport's road access facilities may be analyzed on the basis of standard traffic analysis procedures such as those presented in the TRB Highway Capacity Manual (3). The service-level descriptions and capacity analysis procedures described in the manual may be used to address airport ground access capacity. More specialized advice, including worksheets to guide analysis, is given in the FAA's Airport Ground Access Planning Guide (4). For many purposes, approximations of capacity may be adequate. Average hourly volume of service roadways of typical facilities at Highway Capacity Manual level-of-service C and D is summarized in Table 13-5. The breadth of these ranges makes such approximations useful primarily for initial testing for problems.

124 MEASURING AIRPORT LANDSIDE CAPACITY TABLE 13-5 GROUND ACCESS FACILITY VOLUME (3) Facility Type Average Hourly Volumea (vehicles/hr/lane1') Main-access and feeder fzeeways (controlled access, no signalization) 1,000-1,600 Ramp to and from main-access freeways, single lane 900-1,200 Principal arterial (some cross streets, two-way traffic) 900-1,600 Main-access road (signalized intersections) 700-1,000 Service road 600-1,200 aHiJ.,way level-of-service C and D (see Chapter 11). 1'Passenger-car equivalents. Analysis of the demand on a typical airport access system can be accom- plished by estimating separately the vehicle trips (on each road link) associ- ated with arriving passengers, departing passengers, visitors, employees, ser- vice vehicles, empty taxicabs, and courtesy vehicles and then combining these volumes. Peak periods for regional traffic using airport access facilities should be considered as well as peak periods for passenger traffic. EXAMPLE OF ASSESSMENT PROCESS1 Suppose that a small hub airport is competing to attract an airline hub operation. There is some concern that traffic growth may overburden the four- lane arterial serving the airport. Describe Component The access road is a major arterial within the urban region and serves some local traffic. Existing traffic volumes are estimated by the city traffic engi- neer's office to be approximately 1,400 vehicles per hour in the direction of peak flow, which coincides with that of airport passengers. Describe Demand and Operating Factors Current demand is the same as that described for the analysis of the terminal curb in Chapter 11. Estimated peak-hour passenger volumes are 500 to 680 passengers in the peak direction, and peak airport road demand is estimated to

GROUND ACCESS 125 be currently 480 to 650 vehicles per hour, or approximately 35 to 45 percent of the access road's volume. These estimates may be considered somewhat low because they were developed without detailed consideration of buses, commercial vehicles, and other traffic not generated by air passengers. Such traffic typically is a significant part of access demand. General community growth is projected to continue at approximately 2.5 percent annually for the next 5 years, so that background traffic (e.g., traffic using the highway but not related to the airport) on the road may increase 13 percent over that period, regardless of airport growth: Future background traffic = (growth rate) x [(total traffic) - (airport traffic)] = (113 percent) x [(1,400) - (airport traffic)] = 850 to 1,040 peak-hour passenger cars in peak direction The airport management hopes to achieve a growth of 30 percent in passenger traffic during that same period. Because airport employees report at different times due to passenger peaks, employee ground traffic will not make a significant contribution to access problems that may occur. The access road is a typical urban arterial with signalized intersections and a moderate level of truck traffic. Levels of service are generally in the B to C range during the peak hour. Estimate Service Levels and Service Volumes Assuming no changes in the modes chosen for travel to the airport, growth of 30 percent would mean an increase to 620 to 850 vehicles per hour in the peak direction. Combined with projected background traffic, total volumes of 1,470 to 1,890 vehicles per hour are forecast. On the basis of Highway Capacity Manual information (Table 13-5), the capacity of the two inbound lanes of the main access road would be 1,400 to 2,000 vehicles per hour at service levels in the C to D range. The capacity of the road therefore appears to be adequate to serve forecast traffic. However, because the analysis was based on assumptions with little supporting information, the city traffic engineer is called in to explore whether signal timing, a reversible lane scheme, or some other transportation system management action might be taken to relieve congestion problems if they develop as traffic grows.

126 MEASURING AIRPORT LANDSIDE CAPACiTY RESEARCH NEEDS Methods and standards for characterizing ground access conditions are rela- tively well developed in comparison with those for other components of the terminal landside. However, data are still needed on how driver unfamiliarity with an airport and ability to understand complex roadway systems affect safety and capacity of an airport access system. Low-speed weaving patterns, which are peculiar to airports and similar facilities, are an aspect of traffic operations that merits particular attention. There is also a need for airport-specific evaluations of the relative signifi- cance of ground access delays to air travel. Data from such evaluations would be useful to decision makers responsible for assuring an airport's effectiveness within the context of a regional economy. NOTE 1. Subsections correspond to Steps 3, 5, 7, and 8 in the assessment process shown in Figure 3-1. Attention may also be given to relevant relationships among compo- nents (Step 4) and community factors (Step 6). REFERENCES P. Mandle, F. LaMagna, and E. Whitlock. Collection of Calibration and Validation Data for an Airport Landside Dynamic Simulation Model. Report TSC-FAA-80-3. Federal Aviation Administration, U.S. Department of Transportation. April 1980. F. X. McKelvey. Access to Commercial Service Airports: The Planning and Design of On-Airport Ground Access System Components. Final Report. College of Engi- neering, Michigan State University, East Lansing, June 1984. Special Report 209: Highway Capacity Manual. TRB, National Research Council, Washington, D.C., 1985. M. Gorstein. Airport Ground Access Planning Guide. Report FAA-EM-80-9. Federal Aviation Administration, U.S. Department of Transportation, 1980.

14 Baggage Claim Terminating passengers with checked luggage frequently judge their deplan- ing experience largely in terms of the service provided at the baggage claim. Delays at this area have encouraged many business travelers to carry all of their luggage on board, a practice that affects operations and capacity of other airport components such as security screening and passenger waiting areas. DESCRIPTION Baggage claim areas are typically located adjacent to the direct route of deplaning passenger circulation to provide an area suitable for an activity that involves waiting and heavy circulation. Often a physical barrier is used to separate the claim area from the rest of the terminal building. The area may be leased exclusively to one or more airlines or may be operated by the airport. Although baggage handling equipment may be in- stalled by the airport operator, baggage claim personnel are typically airline employees, and operations within the baggage rooms are normally managed by the airline. Airlines try to avoid significant crowding and delay primarily by staffing to meet demand levels. The principal demand and operating factors influencing service level and capacity in the baggage claim are summarized in Table 14-1. Capacity over the short run—typically a period of 20 min to 45 mm—is determined pri- marily by how many passengers can wait in the same area and the speed with which their luggage arrives and is displayed. Baggage handling equipment capacity is generally measured in terms of numbers of pieces of luggage that can be displayed in a given period of time. 127

128 MEASURING AIRPORT LANDSIDE CAPACiTY TABLE 14-1 DEMAND AND OPERATING FACTORS INFLUENCING SERVICE LEVEL AND CAPACiTY OF BAGGAGE CLAIM Factor Description Equipment configuration and Type, layout, feed mechanism, and rate of baggage claim area display; space available for waiting passengers; rela- tion of wait area to display frontage; access to and amount of feed belt available Staffing practices Availability of porters (sometimes called "sky caps") and inspection of baggage at exit from claim area influence rates of exit; rate of baggage loading/un- loading from cart to feed belt Baggage load Numbers of bags per passenger, fraction of passengers with baggage, time of baggage arrival from aircraft Passenger characteristics Rate of arrival from gate, ability to handle luggage, use of carts, number of visitors DEMAND PATTERNS The number of passengers waiting in the baggage claim depends on the rates at which passengers arrive from the gate and luggage is processed following arrival of a flight. Maximum demand levels are likely to occur during periods when arrivals of larger aircraft cause surges of passengers at the baggage claim. The impact of 300 passengers arriving on a single widebody aircraft is greater than that caused by the same number of passengers arriving on three separate smaller aircraft, because in the latter case, the arrival of both pas- sengers and luggage at the baggage claim is typically spread over a longer period of time. Airline operations serving long-distance, tourist, and vacation traffic will usually carry greater baggage loads than those serving primarily short-haul, business, and commuter markets. Introduction of new types of service can bring radical shifts in the demand on baggage claim facilities. OPERATING CHARACTERISTICS Baggage claim devices may serve two or more flights on one or more airlines. Devices are allocated to flights and airlines according to lease arrangements and demand. Passengers typically form layers (very wide queues) around the baggage claim device, which tend to be deepest around the upstream end of the device and near the primary access point to the claim area. A row of passengers one to two deep has direct access to the claim device and will be able to see and reach their bags. Other passengers wait to gain access to this queue.

BAGGAGE CLAIM 129 Bags are brought on carts from the arriving aircraft and placed on feed belts for transfer to the claim device display area. Rates at which bags are trans- ferred from aircraft to carts and from carts to feed belt and times required to transport carts from the aircraft to the baggage claim area vary with airline operating practices, airport configuration, and operating congestion. ANALYSIS TOOLS AND ASSESSMENT STANDARDS Capacity of baggage claim areas is judged by considering the average time passengers must wait to retrieve their checked baggage and by comparing number of people in the claim area with the size of that area. Simple queueing analysis is often used to estimate the average time passengers will have to wait for bags and the number waiting. Planning and design standards are then selected, generally on the basis of square feet of floor area and linear feet of baggage display device per person. There are no generally accepted industry standards for waiting times in baggage claim areas.1 The guidelines and methods presented in the FAA- sponsored Parsons manual (4) are widely used. In Table 14-2 commonly used guidelines for judging crowding are given. Short-term surges of crowding are typical of any baggage claim facility, and estimated (computed) average levels of crowding will typically decline as progressively longer periods of time are used in the analysis. For example, average waiting area available per person during a peak 20-min period might be only 40 to 60 percent of that available during the peak hour as a whole. TABLE 14-2 PLANNING AND DESIGN GUIDELINES FOR BAGGAGE CLAIM Typical Design Situation Space Standard (ft2/person) IATA standards for claim area waiting (3) System breakdown <8.6 for more than 15 mm High to excellent level of service, comfort > 15.1 at all times FAA implied planning guideline (4) based on claim and wait queue 13 ft deep and approx. 2 passenger/ft display in peak 20-min period - 6.5 at all times Nom: IATA = International Air Transport Association. Demands on the baggage claim may be characterized by using a baggage claim device schedule diagram (Figure 14-1), which is similar in nature to a ramp chart (see Chapter 6). This schedule diagram shows the times when

Flights (aircraft) Assigned Device 727 AB3 D951 72$ A53 72$ 72S I AB3 DSS D95 - C 72$ 73$ 72$ 72S --r- D9S I - I 725 757 757 72S 73$ A83 I - - D - - I73S 73$ I 095 I 73$ 73S 0600 1200 1800 2400 TIME OF DAY FIGURE 14-1 Example baggage claim device schedule diagram (5).

BAGGAGE CLAIM 131 various flights are assigned to each claim device in a baggage claim. From such a diagram, it can be determined when or whether additional flights may be accommodated. When this diagram is used with a passenger arrival dis- tribution and baggage delivery rates, congestion in the claim area can be determined. For judgments about possible future conditions, mathematical queueing and simulation models have been developed to predict the arrival of deplaning passengers and baggage at baggage claim areas (2). Such models and field observations may be used to develop standard distributions of passenger accumulation versus time for particular types of flights at a particular airport, taking into account relative ratesof arrival of baggage and passengers. These standard distributions may then be used to approximate expected conditions if new services are introduced or new facilities are being planned. EXAMPLE OF ASSESSMENT PROCESS2 Suppose that an airline has started service at an airport in which the passenger travel distance between gates and baggage claim is shorter than usual. The airlines and the airport operator have been receiving passenger complaints about the length of time they must wait for baggage. Management wants to explore whether new baggage handling equipment may be needed to improve service. Describe Component A single baggage claim serves flights arriving at three gates, the walking distance between the claim area and each gate ranging from 150 to 300 ft. The baggage claim device is circular, 25 ft in diameter, and located within a square enclosure approximately 50 ft on each side (Figure 14-2). Describe Demand and Operating Factors The flights at this airport have passenger loads of 100 to 160 people, 90 percent of whom have checked baggage. The average baggage load for those with checked luggage is 1.3 bags per passenger. During the peak period of activity, three flights land at 10-min intervals. The claim device delivers 15 bags per minute to the carousel. It usually takes 7 min from aircraft arrival until the first bags arrive at the claim area.

132 MEASURING AIRPORT LANDSIDE CAPACiTY SOft Baggage Claim Enclosure J 50 ft k r.;41 M## FIGURE 14-2 Baggage claim example. Passengers normally begin to leave the aircraft within 2 min and walk to the baggage claim area. Estimate Service Levels and Service Volumes The physical capacity of the system to deliver baggage is determined by the claim device. With current operating practice, the minimum processing time for a single flight may be estimated: Mm. process time = (time for first bag to arrive) + [(no. of bags)/(avg. delivery rate)] = 7 + ([(100 to 160 pax.) x (90 percent) x (1.3)11(15 bags/mm)) = 7 + (7.8 to 12.5) = 15 to 19 mm If passengers walk at 180 to 280 ftJmin, then the expected time of arrival of the first passenger at the baggage claim may be estimated as follows: First passenger's arrival = deplane time + walk time = 2 + [(150 to 300 ft)/(180 to 280 ft/mm)] = 2.5 to 4.7 = 3 to 5 mm

BAGGAGE CLAIM 133 Expected maximum passenger wait time is then the difference between the time of the first passenger's arrival and first bag arrival plus one-half of the delivery time, or 6 to 11 minutes. Average passenger wait time is 2 to 7 mm. At a second airport, the only other one in the region, passengers walk an average distance of 850 ft to the baggage claim area in perhaps 3 to 5 mm. With the same deplaning procedures, passengers at the second airport begin to arrive at the baggage claim areas 5 to 7 min after the flight's arrival. Expected maximum wait time is estimated as only 3 to 8 mm (i.e., 0 to 2 min for the first bag plus an average 3 to 6 min for a particular bag to arrive). It thus appears that service levels at the first airport are even lower than they are at the second airport and that passenger complaints may stem largely from their enhanced perception of the wait, which is based on their experiences at the second airport. Further, the range of time when passengers arrive at the baggage claim is greater when walking distances are longer, so there may be fewer people waiting for luggage at any one time at the second airport. Any perceived crowding may further encourage complaints. If all passengers from one of the larger flights are waiting, there may be approximately 145 people in the claim area, with a net floor area available of approximately 14 ft2/person (2,500 ft2 less the 490ft2 area occupied by the device), which appears more than adequate. The 78.5-ft perimeter of the claim device represents 1.8 ft of display per passenger, again a seemingly adequate amount of space. It would then appear that new baggage claim equipment is not required. RESEARCH NEEDS Additional data are needed on characteristics of bags, passengers, and equip- ment, as well as on airline and airport procedures. Because of the importance of the baggage claim to the passenger's overall perception of an airport and an airline, research into what levels of delay passengers may tolerate and under what conditions is also needed. These data would be valuable as well in mathematical modeling of baggage handling operations, a necessary tool for exploring consequences of new larger aircraft and changes in flight schedules. NOTES In a survey conducted at Birmingham International Airport in England, it was found that passengers were willing to accept delays in waiting for luggage of as much as 22.5 min before terming service bad. Wait times in the range of 12.5 to 22.5 mm were generally found to be tolerable (1). Subsections correspond to Steps 3, 5, 7, and 8 in the assessment process shown in Figure 3-1. Attention may also be given to relevant relationships among components (Step 4) and community factors (Step 6).

134 MEASURING AIRPORT LANDSIDE CAPACITY REFERENCES S. Mumayiz and N. Ashford. Methodology for Planning and Operations Manage- ment of Airport Terminal Facilities. In Transportation Research Record 1094, TRB, National Research Council, Washington, D.C., 1986, pp. 24-35. Air Terminal Processing Capacity Evaluation. Report TP5120E. Airport Services Branch, Transport Canada, Ottawa, Ontario, Jan. 1984. Guidelines for Airport Capaciy/Deniand Managemeru. International Air Transport Association, Montreal, Canada, and Airport Associations Coordinating Council, Geneva, Switzerland, Nov. 1981. Ralph M. Parsons Co. The Apron and Terminal Building Planning Report. Report FAA-RD-75-191. Federal Aviation Administration, U.S. Department of Transporta- tion, July 1975. F. X. McKelvey. Palm Beach International Airport Interim Airport Operating and Use Plan. Aviation Planning Associates, Cincinnati, Ohio, 1984.

15 Customs and Immigration Passengers arriving on international flights must generally undergo customs and immigration formalities at the airport of their initial landing in the United States. Federal Inspection Services (FIS) conducts these formalities, which include passport inspection, inspection of baggage and collection of duties on certain imported items, and sometimes inspection for agricultural materials, illegal drugs, or other restricted items. In recent years, introduction of streamlined procedures for returning U.S. citizens, the "red channel, green channel" system for passing through customs', and computerized access to records at inspection stations have substantially speeded the flow of passengers at many airports. Flights from some Canadian and Caribbean airports are precleared at the originating airport, so arrival formalities are substantially reduced or eliminated. Nev- ertheless, the simultaneous arrival of several fully loaded widebody aircraft can bring a surge of demand that causes service levels to drop dramatically in the international arrivals area. DESCRIPTION International passengers generally arrive in an area segregated from other parts of the airport. All passengers must leave the aircraft and proceed through customs and immigration at a flight's first arrival in the United States. There is little layover or transfer activity in international areas of U.S. airports. U.S. citizens currently proceed directly to baggage claim and then to customs, whereas foreign nationals must first clear immigration. 135

136 MEASURING A]RPORT LANDSIDE CAPACITY On arrival at one of the several inspection booths, foreign passengers present their passports and other documents and parallel queues form. In some busy airports, roving immigration officers examine documents of passengers in queues, helping to ensure that all documents are in order and thereby reducing the average time required for each passenger to clear immigration. At most U.S. airports, U.S. citizens' immigration and customs inspections are combined. Following reentry to the United States, U.S. passengers retrieve their baggage and proceed to customs inspection. Conditions at the baggage claim may become a capacity problem when too many people are crowded into the baggage claim area and baggage arrives slowly. On the basis of declarations made by the arriving passenger and the judgment of the customs inspector, passengers may be required to open their luggage for inspection and may have to pay duties on imported goods. Most passengers proceed directly through an inspection station and exit to the arrivals lobby. The time spent waiting for and undergoing immigration and customs inspections and the conditions of crowding in which the passenger waits determine the service level and capacity of the FIS facility. Although some airlines have full inspection facilities within their unit terminal areas, these facilities are always under the management of FIS officers. Airlines work with the FIS to ensure that flight arrival schedules are known and that an adequate number of inspectors is available to handle arriving passenger loads. However, variations in airline arrival schedules, government operating standards, and budget constraints may sometimes cause staffing shortages or excessive demand loads. Consequent passenger delays may affect the passengers' opinions of airline and airport operating efficiencies. The demand and operating factors influencing service level and capacity of customs and immigration are given in Table 15-1. Capacity is generally determined over the short run—typically a peak period of 1 to 2 hr during which several flights may arrive. DEMAND PATTERNS Flight arrival schedules are a major determinant of demand at the FIS facili- ties. Traffic loads are thus frequently influenced by the location of an airport and the consequent predominant points of origin of arriving flights. Airports on the East Coast of the United States often have daily peak load periods in mid-afternoon and late afternoon, when flights from Europe arrive. West Coast airports with significant numbers of Asian flights may have a mid- morning peak.

CUSTOMS AND IMMIGRATION 137 TABLE 15-1 DEMAND AND OPERATING FACTORS INFLUENCING SERVICE LEVEL AND CAPACITY OF CUSTOMS AND IMMIGRATION Factor Description Number of channels, space, and Inspector channels, U.S. citizen pass-through posi- personnel lions in immigration, "red-gacen" channel use in customs Inspector Average processing time per passenger, efficiency rate of selection for close inspection policy Passenger characteristics Fraction U.S. citizens, flight origin, citizenship of for- eign nationals, baggage loads Space and configuration Available queue space, access to and configuration of baggage display devices, use of carts Flight schedule load Basic determinant of number of people arriving at FIS areas Flight origins may also influence the degree of attention that arriving passengers receive from FIS inspectors. Flights from some parts of the world may receive careful examination because of concern for drug smuggling or may have large numbers of passengers whose visas and other entry papers are carefully examined. Rights from some countries carry large baggage loads, which places an extra burden on customs inspectors. The number, size, and load factor of arriving aircraft can be used to estimate passenger loads at customs and immigration facilities. Walking speeds and distances from arrival gates to the inspection areas determine the distribution of actual passenger arrivals. OPERATING CHARACTERISTICS Operating procedures and planning standards for customs and immigration facilities are specified by the FIS. However, growth in international travel has made it difficult to maintain planning standards at many airports. Very little information on actual performance of FIS facilities has been assembled, and the average processing rate of 50 passengers per hour per agent suggested by FAA guidance material (1) is cited in many publications. However, it has been observed at New York City's John F. Kennedy Interna- tional that average inspection rates can increase when conditions are crowded and passenger characteristics permit FIS officers to maintain inspection standards. Queues may grow very long at some airports. During peak periods at John F. Kennedy International, which is by far the most frequently used point of arrival in the United States, foreign nationals may sometimes have to wait

138 MEASURING AIRPORT LANDSIDE CAPACiTY 20 min or more to clear immigration at the International Arrivals Building.The Port Authority of New York and New Jersey is now planning major expansion of these facilities. ANALYSIS TOOLS AND ASSESSMENT STANDARDS The wait for immigration and customs inspections and the crowding to which passengers may be subjected during these waits are the bases for determining service levels and capacity. A peak period of 1 to 11/2 hr is usually appropriate to observe the impact of multiple flight arrivals at most international airports. The FIS system is reasonably well represented by mathematical m,ultichan- nel queueing models, and general-purpose simulation procedures may be applied with relative ease. However, for most purposes direct observation of cotiditions and simple calculations of average delays and queue sizes are adequate for capacity assessment. Space requirements for waiting passengers are similar to those for other passenger waiting and circulation areas (Chapters 7, 9, 10, and 14). Canadian standards for areas of preclearance to the United States, for example, are similar to those for other departure lounges (2). Those standards define as a breakdown condition space availability less than 0.6 m2/person (6.5 ft2/ person) for more than 15 min during the peak hour. Space availability of more than 1.2 m2/person (12.9 ft2/person) is considered a high level of service and comfort. Allowance should be made for the passenger's load of carry-on luggage at both immigration and customs Stations and for the total baggage load at customs. Many airports provide carts for passengers to use in bringing their luggage to the customs inspection station. At John F. Kennedy International, for example, a planning assumption of 6.25 ft2 per cart—used by a group of two to four passengers—is being used for sizing expansion of the Interna- tional Arrivals Building (Port Authority of New York and New Jersey, 1985). EXAMPLE OF ASSESSMENT PROCESS2 Suppose that a small international airport currently serves four international flight turnarounds per day, two of which use Boeing 747 aircraft. Flight arrivals are spaced so that no two flights arrive within 2 hr of each other. A new airline wishes to enter the market and intends to schedule its L-101 1 aircraft to arrive at almost the same time as one of the Boeing 747 flights

CUSTOMS AND IMMIGRATION 139 already operating. The airport operator is concerned that facilities are not adequate to handle the new passenger load. Describe Component The customs and immigration area is equipped with eight customs inspector positions. There are 12 immigration booths, and the hail in between immigra- tion and customs has approximately 2,800 ft2 for passengers to wait for luggage or queue for customs inspection. Describe Demand and Operating Factors If fully loaded 747 and L-101 1 aircraft land at approximately the same time, 550 to 800 passengers may all enter the immigration area within a short period. Average load factors on the existing flights have been 60 to 75 percent. FIS personnel are available to man all positions if necessary. Queues form at both immigration and customs stations, but there has never been a case in which more than 1 hr was required between flight arrival and clearance of the last passenger. Queues at immigration are seldom longer than six passengers per inspector. Estimate Service Levels and Service Volumes Typical standards for customs inspection indicate that inspection rates of 50 to 70 passengers per hour per inspector can be achieved on a regular basis if both flights arrive from countries unlikely to require special scrutiny. Maximum throughput is estimated as follows: Customs throughput = (inspectors) x (inspection rate) = (8) x (50 to 70 passengers per hour) = 400 to 560 passengers per hour Time required to clear both flights would then be 50 to 65 mm, assuming normal (60 to 75 percent) load factors. Fully loaded aircraft would require 80 to 90 mm. Although these times are a significant increase over the existing situation, they may be acceptable to the airlines. However, it may be expected that at least one-third to one-half of the passenger load may be waiting for baggage or queued at a customs station at

140 MEASURING AIRPORT LANDSIDE CAPACITY one time. In that case, even with typically loaded aircraft, there could be as many as 400 waiting in the 2,800 ft2 of space in the hall. At 7 ft2 per person, conditions could approach serious crowding. RESEARCH NEEDS The FIS generally tries to maintain high service levels in customs and immigration but may be limited by the availability of inspectors. In such cases the airport operator is requested to provide additional space for waiting passengers and to somehow lessen the annoyance of these passengers as waits grow long. Additional research on willingness to tolerate delay and how this willingness might be improved by terminal design and operations measures would prove useful. Continuing efforts to streamline arrival procedures and to improve speed and accuracy in customs screening may relieve problems at airports experiencing rapid growth in international traffic. NOTES This system, long popular in European ports of entry, permits arriving passengers with no goods to declare to proceed without stopping through the "green channel" corridor of customs. The baggage of all those passengers using the "red channel" and of randomly selected passengers in the "green channel" is scrutinized by customs inspectors. Subsections correspond to Steps 3, 5, 7, and 8 in the assessment process shown in Figure 3-1. Attention may also be given to relevant relationships among compo- nents (Step 4) and community factors (Step 6). REFERENCES Planning and Design Considerations for Airport Terminal Building Development. Advisoiy Circular AC 150/5310-7. Federal Aviation Administration, U.S. Depart- ment of Transportation, Oct. 1976. Air Terminal Systems Capacity/Demand Study—Vancouver International Airport (draft). Transport Canada, Ottawa, Ontario, Aug. 1986.

16 Connecting Passenger Transfer An airport's ability to accommodate the quick and efficient transfer of con- necting passengers and their baggage from an arriving aircraft to a subse- quently scheduled aircraft departure is important to passenger safety, comfort, and convenience, as well as to airline operating efficiency. Airports serving significant numbers of connecting passengers increasingly play a key role in the nation's air transportation system. DESCRIPTION Transfer passengers must travel with their carry-on baggage from one gate to another by walking or with the aid of buses or other mechanical devices, sometimes moving between scparate terminal buildings, possibly leaving and reentering secure areas, and sometimes using check-in and other facilities along the way. Arriving international passengers must pass through customs and immigration and claim and recheck their luggage. When an on-line connection is made (between two flights operated by the same airline), the airline will typically try to ensure that the passenger is assisted with the transfer. Airline hub-and-spoke operations depend on the ability of passengers to make the transfer quickly and easily. Transfer pas- sengers arriving and departing on flights operated by different airlines must make an interline transfer. Typical problems encountered by transfer pas- sengers making transfers at some airports include long distances to be tra- versed, obstacles such as changes in elevation and unprotected areas separat- ing terminals, and poor information on where the next flight's gate is located. 141

142 MEASURING AIRPORT LANDSIDE CAPACiTY TABLE 16-1 DEMAND AND OPERATING FACTORS INFLUENCING SERVICE LEVEL AND CAPACITY FOR PASSENGER TRANSFER Factor Description Terminal configuration Distance between gates, information for connect- ing passengers, intervening security screening Ground transport Connecting passenger assistance systems, bag- gage transfer systems Passenger characteristics Fraction needing assistance for ground transport, intergate travel speeds, baggage loads Flight schedule and load factors Basic determinant of number of people making peak-period connections Standard minimum connection times listed in the Official Airline Guide (OAG) and agreed on by airlines and reported to the travel industry are generally based on access time for passengers and 100 percent transfer of baggage. In this chapter the focus is on passenger transfer, and it is assumed that the baggage transfer times agreed on by the airlines are adequate. The principal demand and operating factors influencing service level and capacity for passenger transfers are given in Table 16-1. These factors influ- ence how long it may take for passengers to make the transfer, which is the primary basis for judging service level and estimating capacity. DEMAND PA11ERNS Transfer passenger traffic in general varies with the number of flight arrivals and departures scheduled within a period of 60 to 120 min. On-line connecting passengers usually have a short journey for their transfer. However, rapid growth in activity at airports where airline hub-and-spoke operations are centered has sometimes led to widely separated on-line gates and subse- quently longer connection times. If an airline operates a route hub at a particular airport, the number of interline passengers will be reduced, although total transfer traffic may be high. Large airlines have in recent years formed associations with small commuter carriers in which flight times of the major carrier's long-distance flights are coordinated with those of the commuter carrier, so that the latter serves as the feeder to the hub. Often the two carriers share terminals and gate space, making transfers easier for passengers on those particular airlines. At some large airports, the passengers of interest should also include originating and terminating passengers who may have an automobile parked near the terminal of an airline other than the one on which they are currently

CONNECTING PASSENGER TRANSFER 143 traveling and who then use the airport's interline transfer system to reach their vehicle. Some analysts have observed that the number of these "phantom transfers" may become relatively significant (1). OPERATING CHARACTERISTICS The physical design of the airport's terminal facilities is the principal variable influencing service provided to transfer passengers. However, effective sign- ing and other assistance to aid the transfer passenger may influence their ability and perceptions of service offered and mitigate some difficult aspects of making a transfer. In many airports, interline transfer passengers have no choice but to walk from one airline's area to another's. But in some large airports, systems are available to aid the passenger in this movement, such as moving walkways, people movers, and interterminal buses. Buses, however, are subject to con- gestion on airport roadways and at the terminal curb. Collection of fares for buses and people movers makes these facilities less effective and desirable from the passenger's point of view. The time that passengers require for transfer will depend on the characteris- tics of the passengers as well as the design of the airport. Some airports in Florida, for example, have a high proportion of elderly passengers who may require longer times to traverse a given distance. People movers, moving sidewalks, and buses may improve transfer times in such airports. Large hub airports' typically require a longer time for transfer passengers than do small hub airports. For example, in the OAG, which lists standard minimum connection times for major airports in the United States, the average time required for interline transfers at large hub airports (approximately 45 min for domestic and 75 min for international connections) is nearly twice that for small hub airports (2). These statistics are summarized in Figure 16-1 and Table 16-2. ANALYSIS TOOLS AND ASSESSMENT STANDARDS There are virtually no analytical techniques intended to deal specifically with passenger transfers. Assessment of this component of the landside typically requires a direct estimation of transfer times by using assumptions about passenger walking speeds, measured distances at the airport, and obstacles or aids to movement. At some airports, the assessment may be appropriately.

0.8 LU N 0.7 I 0.6 0.5 0 0.4 4 Li- o 0.3 LU C, 4 I- 0.2 z LU C) cc 0.1 LU 0. 0 40 mIn 30-40 mIn 45-60 mIn >1 hr OAG STANDARD MINIMUM INTERLINE TIMES FIGURE 16-1 Distribution of minimum standard interline connection times at 112 primary U.S. airports by hub size (2, 3). TABLE 16-2 AVERAGE MTh1TMUM ALLOWABLE INTERLINE CONNECTION TIMES AT U.S. AIRPORTS (2, 3) Airport Type and Flight Avg Minimum Interline Connection 'limes (hr:min) Domestic International Hub Large 0:45 1:16 Medium 0:30 1:02 Small 0:23 0:47 Flight Long haul 0:39 1:10 Medium haul 1:35 1:08 Short haul 0:20 0:10

I nterterminal Distance Approximately 1/2 mi Terminal C Pedestrian Connections Terminal A CONNECTING PASSENGER TRANSFER 145 conducted in conjunction with the assessment of access or general circulation conditions. EXAMPLE OF ASSESSMENT PROCESS2 Suppose that a large hub airport has several separate unit terminals linked by a bus traveling on the frontage road (Figure 16-2). The OAG recommends a minimum interline transfer time of 40 mm. Interline transfer passengers may walk between terminals or wait for the bus. The airlines and airport have received numerous complaints about delays and difficulty in transferring between terminals. FIGURE 16-2 Example of assessment for interline transfers (drawing not to scale). Describe Component The three terminals are separated by approximately 112 mi. Buses circulate at 5-mi intervals, with an average time between stops of approximately 2 mm, although frequent traffic congestion may slow circulation substantially.

146 MEASURING AIRPORT LANDSIDE CAPACiTY Describe Demand and Operating Factors Airport management has observed that deplaning transfer passengers with baggage generally reach the curbside bus stops within 14 to 22 min of their arrival. Without baggage, passengers reach this point within 4 to 6 mm. Passengers then choose either to wait for a bus or to walk to the terminal where their departing flight is located. Approximately 30 percent of interline transfer passengers have little hand- carried luggage and could walk to the terminal if necessary. Approximately 50 percent of those choosing the bus must go to the second terminal away from their current position before reaching their destinations. Bus operations are as described in the previous paragraph. Passengers choosing to walk may travel through covered corridors, although there are some stairs or ramps to be negotiated. Walking routes are well signed. Estimate Service Levels and Service Volumes Under uncongested conditions, transfer passengers using the bus to travel to the most distant terminal wait approximately 2 to 3 min for a bus to arrive and then ride for 4 mm. Total time to reach the curb of the second terminal is then approximately 20 to 29 min. Allowing 7 to 10 min for check-in and walking to the gate, the total time required for these passengers is 27 to 39 mm. At the upper end of this range, the service level is close to the OAG standard of 40 min minimum connecting time, but there is little safety margin for time lost to traffic congestion along the terminal curb. A transfer passenger choosing to walk may go directly to either terminal. Such a passenger travels at an average speed of perhaps 180 to 250 ft/mm. The 1/2 mi between terminals is traveled in approximately 21 to 30 mm. This passenger then arrives at the second terminal within 25 to 36 mm. Again allowing 7 to 10 min at check-in, the total time is 32 to 46 mm. At the upper end of this range, the 40-min target is clearly not met. On the basis of this analysis, the airport operator may conclude that there is too little safety margin for slow traffic. A more detailed study of the problem may be warranted to investigate whether improvements in service levels are needed. RESEARCH NEEDS Data on length of time and maximum walking distances for interline transfer that passengers will tolerate would be useful for assessing service levels and determining capacity to serve connecting passengers.

CONNECTING PASSENGER TRANSFER 147 NOTES As defined by FAA; see the glossary. Subsections correspond to Steps 3, 5, 7, and 8 in the assessment process shown in Figure 3-1. Attention may also be given to relevant relationships among compo- nents (Step 4) and community factors (Step 6). REFERENCES R. deNeufville. Airport Systems Planning. Macmillan Press, London, 1976. Official Airline Guide, North American Edition, Vol. 12, May 1, 1986. National Plan of Integrated Airport Systems 1984-1993. Federal Aviation Admin- istration, U.S. Department of Transportation, 1985.

17 Landside System as a Whole The components discussed in the preceding chapters are linked together in an airport terminal into a total system through which passengers move to and from aircraft. Small queues and short delays in each component, although individually well within tolerable ranges of performance, may still combine to produce a landside capacity problem. The capacity of the landside system of a particular airport taken as a whole is more difficult to assess than that of an individual component, but nevertheless it can be done. However, there is no single service level or capacity for the whole system in the same sense as there is for a single component unless demand on all components is perfectly matched to each component's maximum throughput or there is an accepted set of comparable service-level targets for all components and all components are operating at these target service levels. Discussions of the terminal system as a whole consider total processing time for the enplaning and deplaning passenger. This total processing time is a sum of three elements—the service time at each component, the wait time at each component, and the travel time between components that make up the system. There are no generally accepted standards for this total processing time, although an individual airport may set its own maximum acceptable standards and compare actual performance with these. DESCRIPTION Individual components are linked together in parallel and in series, defining the paths passengers may take through the system. Figure 17-1 shows an example of these linkages. 148

Runway/Apron Gates Domestic Hoidrooms International Hoidrooms 1,135 passengers 1,885 passengers Pre-Primary Inspection Line 580 passenaers Queue Area ondary immigr and Health N/A Arrival Level I Bag Claim Area 1,258 persons 1,229 passengers Bag Claim 3,300 I Bag Claim 1,000 passengers/hr J L passengers/hr Queue Area Secondary Customs Concessions I 41 Arrivals Level N/A I 926 Dersons Curb/Parking N/A = not available Groundside Access FIGURE 17-1 Example terminal landside flow schematic (deplaning) (1). Terminology corresponds to Transport Canada practices. Primary inspection line corresponds approximately to FIS activity in U.S. airports.

150 MEASURING A]RPORT LANDSIDE CAPACiTY If all components are operating at their maximum throughput rates, throughput of the terminal as a whole is determined by the most constrained component in each independent parallel path. In theory, although service levels for these most constrained components may severely deteriorate, the system as a whole will still be able to continue processing passengers. In practice, serious crowding and congestion in one component often affect demand in connecting components and service levels decline overall. In general terms, all the factors previously discussed as influencing the levels of service and capacity of individual components influence the landside system as a whole. The number of passengers that can be served by the whole system is limited by the conditions that can be tolerated in individual compo- nents. The achievable service volume of an airport's landside as a whole is the level of aggregate demand that places loads on an individual component that in turn cause service conditions. to drop to a minimum acceptable service level. In an airport where all passengers use the same facilities, this limit may be highly visible and may restrict the airport's operations. Capacity assessment for smaller airports and airports served by a limited range of access facilities may for this reason be particularly meaningful. In more complex landside systems severe problems may develop in one part of the airport; for example, one of several unit terminals may perform at very low levels of service. Greater numbers of people could be served at this airport if these new passengers used the less severely stressed terminals. However, given currant patterns of demand, operating procedures, and facilities configurations, it may be entirely reasonable to say that the airport has reached a limiting service volume. Although total time required to traverse the landside system is a frequently discussed measure of overall service level, other measures may be used. For example, the risk that a passenger will miss his flight's scheduled departure can be estimated if there are adequate data. Whatever the measure, all components may individually perform with acceptable service levels, yet cumulative effects may cause the landside system as a whole to be judged unable to satisfy demand. DEMAND PATFERNS Both timing and distribution of demand among components are important. For an obvious example, two aircraft can occupy the same gate at different times but will need two gates if they are to be served at the same time. Passengers arriving on these aircraft may encounter serious delays in baggage claim if they all arrive at one time and no delay at all if they arrive at different times.

LANDSIDE SYSTEM AS A WHOLE 151 They may also encounter no delays at baggage claim if a large fraction of the passengers are transfening between flights. The passenger demand associated with the full day's flight schedule must generally be considered, both by time of day and by where these aircraft are parked. Passengers entering the landside—deplaning through gates in a sys- tem such as that shown in Figure 17-1 or enplamng via access roads and the terminal curb—make their way from one component to the next and in doing so influence service conditions along the way. The likely rate at which this movement occurs may be estimated from the flight schedule and estimated characteristics such as walking speed and walking distance. Alternatively, periodic observations of service time, occupancy level, queue length, and other indicators of service level may be made and linked by inference to flight schedules and passenger characteristics. Standardized relationships may then be developed for a particular airport or part of an airport (Table 17-1) and used in capacity assessment and planning, although these relationships are neces- sarily only an approximation of actual conditions. OPERATING CHARACTERISTICS It is difficult—if not impractical—to discuss operations of the landside as a whole without reference to individual components. Shifting demand among components to achieve a balance in service conditions is the principal means for achieving optimum passenger service volumes. In an ideally balanced landside system, demand is matched to component capacities so that all component service levels are comparable. (This presumes that comparable definitions of service levels can be developed for many diverse components.) The landside system as a whole can then be said to operate at a service level similar to that of each component. If service-level targets can be established for critical components at a particular airport, system service level may be termed adequate if all components are operating at or above their target and if overall processing time is also considered adequate. ANALYSIS TOOLS AND ASSESSMENT STANDARDS There are at present no generally accepted standards of overall landside system performance. The relative consistency of passenger behavior in arriv- ing at an airport in advance of scheduled departure demonstrates that such standards may be developed but data are at present not adequate. Landside systems of even moderate complexity cannot be effectively as- sessed without some type of simulation, and computer assistance is often

152 MEASURING AIRPORT LANDSIDE CAPACITY TABLE 17-1 TYPICAL PASSENGER DEMAND DISTRIBUTION: TERMINATING FLIGHTS, VANCOUVER INTERNATIONAL (1) Maximum Percent- age of Flight Load Time After Arrival Component in the Area of Aircraft (mm) Customs baggage Transborder 52 15 International 42 30 Domestic baggage Short haul 56 8 Long haul 62 10 Main arrival-level concourse Short-haul domestic 30 15 Long-haul domestic 32 14 Main departure-level concourse Short-haul domestic 8 12 Long-haul domestic 4 10 Transborder 8 25 International 14 100 International arrival-level concourse Transborder 26 22 International 34 55 Terminal total Short-haul domestic 98 4 Long-haul domestic 100 4 Transborder 100 4 International 100 8 needed. A number of attempts have been made, with mixed success, to devise detailed and generally applicable computer-based simulations of the terminal landside as a whole, at least for the terminal building and gate complex (see Appendix B). Analytic procedures include the critical path method (CPM), probabilistic queueing and network modeling, and "brute force" (in which the computer goes through many iterations of applying demand and estimating service levels in the abstract system model of the airport). In each case the analysis results are presumed to represent likely conditions in the real airport. Assessments of individual components yield estimates of service volumes and service levels achievable over some period of time under known or assumed demand characteristics. These individual component service vol- umes converted to equivalent hourly volumes may be used to represent the system. When demand is not balanced to capacity throughout the system (e.g., when service level falls below target in one or more individual components), an approximation of achievable system service volume may be estimated by calculating the ratio of the below-target component's maximum service

LANDSIDE SYSTEM AS A WHOLE 153 volume to that component's share of total system demand served. The compo- nent yielding the lowest computed equivalent system volume then determines the approximate limit of overall landside service volume. EXAMPLE OF ASSESSMENT PROCESS' Suppose that an airport has two identical unit terminals. Assume that the entire assessment process has been completed for each terminal and that the airport operator wants an estimate for planning purposes of total airport service volume achievable with current facilities. Describe Component Assume also, for simplicity, that the peak-hour service volume for each terminal was found in the preceding assessment to be limited by gate occu- pancy, although proposed introduction of widebody antraft could overload existing ticket counter and baggage claim facilities. Describe Demand and Operating Factors Flight schedules have been analyzed and counts have been made of passenger movements through the terminals. Both terminals have 10 gates, all in use during their peak hours. The peak total number of passengers passing through either terminal during the respective peak hours is approximately 1,500, 60 percent of whom are moving in the peak direction (i.e., either enplaning or deplaning). However, peak hours for Terminal A are in the early morning and late afternoon, whereas those for Terminal B occur at midday. Average total daily passenger counts are 9,000 at Terminal A and 7,500 at Terminal B. The airport's annual traffic is approximately 5 million passengers. In preceding analyses, the maximum acceptable future peak-hour demand for each terminal was estimated to be 2,000 total passengers in the peak hour, based on assumptions that widebody aircraft are introduced, no new landside facilities are constructed, and 60 percent of peak-hour passengers move in the peak direction. At this demand level, service levels forecast at ticket counter and baggage claim are below what the airlines and airport operator are likely to find acceptable.

154 MEASURING AIRPORT LANDSIDE CAPACITY Estimate Service Levels and Service Volumes Apparent maximum service volume for the two terminals together, acting independently and in parallel, would be the sum of their individual maximum volumes, or 4,000 total passengers per hout This estimate presumes that ground access and parking facilities, on the one hand, and airside facilities, on the other hand, can accommodate the combined peak-hour operations. Over the course of the 16-hr service day, traffic volumes could be as high as 64,000 total passengers per day. However, if patterns of demand are maintained, Terminal A will continue to serve nearly 55 percent of the airport's daily traffic. As the airport's total traffic grows, Terminal A will achieve its maximum service volume before Terminal B, and can serve daily traffic of approximately 12,000 total pas- sengers. In the absence of changes in patterns of demand, daily traffic at the airport would then be approximately 22,000 passengers, which is probably a more realistic estimate of the terminal area's maximum achievable service volume with existing facilities. RESEARCH NEEDS Measures of total landside system performance are extremely valuable. Al- though data are collected on airside delay due to a variety of causes, compara- ble data are seldom collected for the landside. Such landside data would not only support development of service-level targets at individual airports but would also assist airport operators and the FAA in discussions of appropriate balancing of investments in airside and landside facilities to achieve optimum airport system operations. NOTE 1. The subsections of this example correspond generally to Steps 3-8 in the assessment process described in Part 1. However, analysis of the landside system as a whole generally must be preceded by or include assessments of individual components. System analysis is accomplished by using these previous results and is concentrated in the interactive accomplishment of Steps 7-9 in the general process. REFERENCE 1. Air Terminal Processing Capacity Evaluation. Report TP5120E. Airport Services Branch, Transport Canada, Ottawa, Ontario, Jan. 1984.

Glossary Included here are definitions of selected terms helpful to understanding and discussing airport landside capacity. Terms in italics are defined elsewhere in the glossary. ACCESS. In the context of aviation activities, the airlines' ability to offer service at new airport locations or to increase service at airports already served (see Ground access). AIR TRAFFIC CONTROL (ATC). The federally regulated system of rules, procedures, instruments, and personnel intended primarily to assure flight safety. ATC procedures may have substantial influence on timing of aircraft departures and arrivals at an airport. AIRCRAFT OPERATION. Aircraft departure from or arrival at an airport; takeoffs and landings. Airside facilities must serve all operations (includ- ing those of general aviation), whereas landside activity is typically related only to departures or arrivals of commercial service aircraft. AIRPORI COMMUNITY. In broad terms, those served by an airport. This includes passengers, shippers, and other airport users; employees of the airport and businesses relying on air transportation; neighbors of the airport, especially those exposed to aircraft noise, airport access traffic congestion, noise, and pollution; and local and state government. AIRSIDE. Airport facilities associated with aircraft movement to transport passengers and cargo, used primarily for landing and take-off, for example, runways, taxiways, and ATC facilities. The airside may overlap the air- space at ends of runways. AIRSIDE CAPACITY. See Capacity (airside). AIRSPACE. Designated area beyond the airport where aircraft are permitted to operate, often under ATC regulations; may overlap the airside. ANALYSIS PERIOD. Specified period of time, typically a peak period, used for analysis of landside capacity. Choice of period depends on functional components considered and demand characteristics. 155

156 APRON. Aircraft interface between landside and airside. It includes ramps and aircraft circulation area. AUTOMATED GUIDEWAY TRANSIT (AGT). Fixed-guideway system for transporting passengers between central terminal and remote terminal, among unit terminals, or to other airport facilities. BAGGAGE SERVICES. Processing of passengers' checked baggage. In- cluded are destination tagging, movement to baggage room, sorting, move- ment to and from aircraft, loading and unloading, and delivery to baggage claim display device. Interline transfer, storage, and delivery may be included. CAPACITY (AIRSIDE). As defined by the FAA, the maximum number of aircraft operations that can take place in an hour. This is a maximum throughput rate. CAPACITY (LANDSIDE). As defined in this study, capability of the land- side or its functional components to accommodate passengers, cargo, ground transport vehicles, and aircraft. Service volume is the principal indicator of landside capacity in this report. CHECK-IN. Initial step in passenger processing, involving passenger contact with the airline immediately before flight departure. It may include ticket inspection, issuance of boarding pass and seat assignment, baggage check- ing, ticketing, and preliminary inspection of immigration documents and may occur at ticket counter or gate area. To speed processing, some steps may be completed in advance of passenger arrival at the airport. CLEAR ZONE. Area at ends of runways and other areas surrounding airport in which height and land use limitations are imposed to ensure that no obstructions to safe aircraft operations occur. COMMERCIAL SERVICE AIRPORT. As defined by the FAA, public use airport receiving scheduled passenger air service and enplaning at least 2,500 passengers annually. There are 552 such airports included in the FAA's 1984 National Plan of Integrated Airport Systems (NPIAS). COMMUTER. Airline providing service primarily over short-haul route seg- ments connecting small airports to hub locations. The term also refers to the smaller aircraft used for such services, typically with seating for 10 to 50, and to the general service conditions associated with such airline operations. CONNECTING PASSENGER. A transfer passenger. CROWDING. Density of people in airport waiting areas, or number of people per unit area. Those accompanying departing passengers or greeting arrivals may be included as well as passengers themselves. DELAY. For the airside, added time spent in accomplishing an aircraft opera- tion because of airport congestion, or the difference between time required under constrained conditions caused by simultaneous demands on the

157 facility and time required under unconstrained conditions. Landside delay is the added time required for a passenger to complete processing at a functional component because of limits to capacity. Wait time and process- ing time are included. Acceptable delay depends on type of service being delivered, demand characteristics, and local conditions at an airport. DEMAND CHARACTERISTIC. Number of air passengers and aspect of their behavior that materially affect the ability of a functional component or group of components to accommodate them. Such factors as the timing of passenger arrivals at the airport, age, trip purpose, fare paid, baggage carried or checked, and whether passenger has a ticket and boarding pass are often important. Airlines often try to tailor their services to their passengers' demand characteristics. FEDERAL INSPECTION SERVICES (FIS). Federal government processing of international passengers and baggage, primarily on arrival in the United States. Immigration, customs, agricultural, public health, and narcotics control functions are included. Terminal areas for international arrivals include FIS facilities staffed and operated by federal employees. FUNCTIONAL COMPONENT. Element of the landside such as a gate or ticket counter that provides specific service to air passengers or cargo. Functional components unable to meet demand characteristics and main- tain adequate service levels may become limits to capacity. GATE. Terminal portal for passengers to enter and exit aircraft. The term is commonly used to mean a loading bridge-equipped entry adjacent to a holdroom, but may include entry to a transporter or directly onto an apron. It sometimes includes the hardstand. GENERAL AVIATION. Activities associated with private and business air- craft as opposed to common-carrier passenger aircraft and airports with less than 2,500 annual enplaned passengers or used exclusively for such activities. In addition to reliever airports, 2,440 general aviation airports are included in the NPIAS. GROUND ACCESS. Highways, local streets, fixed guideway systems, and public and privately operated transit services linking an airport to the area that it serves. GROUND HANDLING. Unloading and loading of catering supplies, bag- gage, and cargo; fueling; and minor maintenance on the apron associated with servicing an arriving aircraft and preparing it for departure. HARDSTAND. Aircraft apron parking position equipped with fixed facili- ties for ground handling but not directly linked to a terminal. It may be considered as a gate. HOLDROOM. Passenger waiting area adjacent to gate. The term is also used to refer to other passenger waiting areas such as that for immigration or baggage claim devices.

158 HUB-AND-SPOKE OPERATION. A pattern of airline routes that brings direct flights from many points (the spokes) to a centrally located airport (the hub). Flight schedules allow passengers to transfer quickly between flights during periods when many aircraft are simultaneously at the hub location. Such a route structure is intended to maintain high levels of aircraft utilization and loading. Airline hubs increase proportions of pas- senger transfer traffic. These transfer centers do not necessarily qualify as hub airports as defined by FAA. HUB AIRPORT. A standard metropolitan statistical area (SMSA) and the commercial service airports serving that area that account for at least 0.05 percent of all passengers enpianed annually in the United States. Because some SMSAs are served by more than one airport, there are fewer hubs than there are hub airports. Hub airports are classified by the percent of total domestic enpianements as large (1 percent and more), medium (0.25 to 0.99 percent), and small (0.05 to 0.25 percent). The NPIAS includes 140 hub airports. iNTERLINE TRANSFER. A passenger and his checked baggage changing from one air carrier to another while in transit at an airport. Such transfers, in contrast to transfers within the same airline hub-and-spoke operation, pose particular problems of baggage-handling logistics and may require passengers and baggage to move between terminal buildings at larger airports. LANDSIDE. Facilities and services associated with air passengers or cargo movement between aircraft and trip origin or destination. The landside includes aprons, gates, terminals, cargo storage areas, parking, and ground access. LANDSIDE CAPACITY. See Capacity (landside). LOADING BRIDGE. Mechanical device and passenger pedestrian pathway to link terminal to aircraft. Sometimes called a "jetway," although this term is a registered trademark. LONG HAUL. Flights longer than 1,500 mi. Such flights normally require more preparatory ground time before departure than short-haul flights (those less than 500 mi long) and are often flown by larger aircraft. MAXIMUM THROUGHPUT. Maximum rate at which passengers (or air- craft, ground transport vehicles, pieces of baggage, tons of cargo, etc.) can be processed by a functional component or group of components. In practice this rate is observed only when demand equals or exceeds a component's processing capability, and is typically sustained only for brief penods, because excess demand usually produces significant delays and crowding. OFF-LINE TRANSFER. Passenger changing planes between flights operated by different airline companies. Also termed interline transfer.

159 ON-LiNE TRANSFER. Passenger changing planes between flights operated by the same airline company. Times required for on-line transfer may be shorter than those for off-line transfers because gates are located closer together and flight schedules are coordinated. PART 150. Portion of Federal Aviation Regulations (FAR) implementing aircraft noise measurement and compatible land use planning to limit areas and population exposed to aircraft noise. PASSENGER CIRCULATION AREA. Corridor, stairway, escalator, or mov- ing walkway connecting processing components, generally only in a terminal. PASSENGER SCREENING. Security inspection of passengers and hand- carried baggage in preparation for enpianement. Such screening typically includes x-ray and occasional hand search of baggage and metal-detector (magnetometer) examination of passengers. PEAK LOAD FACTOR. The ratio of demand during the peak period (for example, a peak hour) to average demand during a reference period (for example, the daily average hour). Generally expressed as a number or percentage, for example, 1.2 or 120 percent. PEAK PERIOD. Time period, which may be one hour, several hours, or one day, representative of busy conditions within a functional component. It is typically defined from historical records by frequency of occurrence. PEOPLE MOVER. A type of automated guideway transit (AGT). PRIMARY AIRPORT. Commercial service airport at which at least 0.01 percent of all U.S. passengers are enplaned annually (as reported in the NPIAS, equal in 1982 to about 31,000 enpianements). The NPIAS lists 280 such airports. RAMP. Aircraft parking position, often used to refer to gate parking positions. RAMP CHART. A graphical presentation of the daily schedule of flight operations for a group of gates. The chart shows scheduled arrival and departure times, and thus when a gate is occupied. RELIEVER AIRPORT. General aviation airport that has the designated function of relieving congestion at primary airports. Such airports increase access for general aviation to the community and may be candidate loca- tions for airlines wishing to expand service or enter new markets. The NPIAS lists 227 reliever airports. REMOTE PARKING. Automobile parking areas located at some distance from the terminal and connected to it by shuttle bus service or a people mover. REMOTE TERMINAL. Facility located at some distance from an airport where passengers may undergo some part of the processing associated with

IM the landside portion of the trip. Remote parking and check-in may be included. SERVICE LE\'EL. The quality and conditions of service of a functional component or group of components as experienced by passengers. Such factors as delay, crowding, and availability of passenger amenities for comfort and convenience measure service level. SERVICE-LEVEL TARGET. Minimum or maximum tolerable service level during a particular analysis period established by airport operator, airlines, the FAA, and the community to guide decision making. SERVICE TIME. Time required, excluding waiting time, to process a pas- senger at a functional component such as a ticket counter or passenger security screening facility. SERVICE VOLUME. Number of passengers (or aircraft, ground transport vehicles, etc.) with particular demand characteristics that can be accom- modated by a functional component or group of components during an analysis period at a given service level. SHORT HAUL. Flights less than 500 mi long. Aircraft on short-haul routes may be able to operate with very short turnaround times compared with those on long-haul routes. STRUCTURED PARKING. Multilevel building for automobile parking at airport. TERMINAL. Building with facilities for passenger processing and boarding of aircraft or groups of such buildings (unit terminals, often used by a single airline) within a terminal area. Terminals are often classified into four configurations by the system used for horizontal movement of pas- sengers: linear, pier, satellite, and transporter (see Figure G-1). TERMINAL CURB. Passenger interface between ground access and termi- nal. Passengers arrive or depart in private automobiles, hotel and rental-car vans, limousines and buses, and transit vehicles. The curb system may include direct rapid transit and rail system links to the airport, although Stations are typically located elsewhere and linked by bus or pedestrian paths to terminal buildings. TRANSFER. Passenger changing planes at an airport en route to the final destination. Such passengers, either on-line or off-line transfers, typically are at the airport for a relatively short period of time and use fewer landside facilities and services than passengers starting or ending their journey at the airport. Airline hub-and-spoke operations sharply increase the number of transfers at the hub airport. TRANSPORTER. Mobile vehicle for carrying passengers between terminal and hardstand. TURNAROUND TIME. Scheduled time required between aircraft arrival and departure for passenger unloading, ground handling, and boarding.

161 Cu,b Curb (a) Unbar (example: Dallas- Ft. Worth International) -*---------- * Apmr Curb () Satellite (example: Atlanta, Harisfield International) () Pier (example: New York, LaGuardIa) Apron + Curb (d) Transporter (example: Washington, D.C., Duties International) FIGURE G-1 Airport landside terminal configurations. USER. Broadly understood to include passengers, airlines, cargo shippers, concessionaires, and others who use airport landside facilities and services. In this report, passengers are the principal users, and are served by airlines and airport operations. WIDEBODY. High-passenger-capacity jet aircraft such as the Airbus, Boeing 747, McDonnell-Douglas DC- 10, and Lockheed L- 1011. Physical dimen- sions of such aircraft, also termed jumbo jets, may be incompatible with airport gate and apron areas designed for smaller narrowbody aircraft.

Appendix A Framework for Defining Airport Landside Service-Level Targets There are no generally accepted definitions or targets for acceptable landside service levels at U.S. airports. A comprehensive set of such targets is a logical prerequisite for consistent nationwide assessments of landside capacity. This study was undertaken with the hope of recommending target levels, but the committee concluded that existing data are simply inadequate to support even a suggestion of targets that might be applied nationwide. Nevertheless, a guiding framework for developing landside service-level targets can be rec- ommended and is described briefly here. The framework has four principal dimensions: Type of airport, Type of air transport service, Prototypical service-level target, and Landside functional component. Other dimensions that may influence selection of service-level targets at a particular airport include geographic location of the airport, its role within a regional and national airport system, and management characteristics such as financing arrangements. Research is needed to extend the framework proposed here beyond com- mercial passenger capacity. Cargo services offered by passenger airlines and dedicated air cargo transporters place significant demands on landside facili- ties at many commercial service airports and should be considered in any determination of airport landside capacity. 162

163 TYPE OF AIRPORT Airport size is a key factor influencing passenger experience. A physically small airport takes less time to negotiate than a large airport and offers a narrower range of passenger amenities. The FAA's classification of commer- cial service airports into four categories by number of annual passenger enpianements may be a reasonable basis for setting targets (1). For example, there is a strong correlation between this measure of airport type and mini- mum interline flight connection times accepted by the airlines serving a particular airport.1 A similar correlation may be found between size and such service-level indicators as baggage claim service time, gate occupancy time, and total time required for passengers to travel through the landside. Overall layout of the airport terminal area may influence capacity as well. For example, multi-building terminals typically make interline passenger and baggage transfers more difficult than more centralized configurations, but may offer advantages for apron and gate area utilization. TYPE OF AIR TRANSPORT SERVICE Whether an airline hub-and-spoke operation is centered at a particular airport is another important factor. The daily pattern of passenger peak loads changes substantially when an airline hub begins operation, particularly with respect to the number of peaks. An airline hub typically increases the number of peaks and raises average daily utilization of gates and holdrooms. Predominant stage lengths of flights originating at an airport (e.g., long versus short haul) and travel market served typically influence times required for aircraft ground handling, passenger access requirements, passenger bag- gage loads, and amenities airlines offer their passengers. A significant number of long-haul international flights at an airport, commuter operations, low-fare budget airlines, charter operations, and airlines serving primarily business travel or vacation and tourist destinations each have a particular influence on service-level targets. PROTOTYPICAL SERVICE-LEVEL TARGET A generally applicable and easily understood description of service levels is a valuable basis for setting service-level targets at any particular airport. The broad international acceptance of the highway level-of-service descriptions on a scale of A through E from "free flow" to "breakdown," presented in TRB's Highway Capacity Manual demonstrates the poinL

164 Although proposals have been made to adapt the six highway level-of- service prototypes to the airport landside (2), the committee believes that such a scale may be overly detailed and proposes three prototype service levels: Level 1: Passengers are unlikely to encounter delays, queues, or crowd- ing. Aircraft operations are not affected by landside conditions. Level 2: Passengers may encounter some delay at isolated locations or for limited times during peak periods. Some queueing and crowding are observed. Aircraft operations are not affected by landside conditions. Level 3: Passenger delays are likely. Queues and crowding are observed generally throughout peak periods. Aircraft operations may be affected by landside conditions. These prototypical service levels are based on the committee's experience with conditions typically encountered at U.S. commercial service airports. Research is needed to determine specific criteria for likelihood of occurrence of delays, queues, and crowding, and their effect on aircraft operations. LANDSIDE FUNCTIONAL COMPONENT A comprehensive set of service-level standards should address each of the 11 individual types of functional components and the landside system as a whole, as discussed in Part II of this report. Unique conditions at some airports may justify establishing service-level targets for other landside components as well, for example, to monitor interairport connecting time in a multiairport region. NOTE 1. This relationship is discussed with respect to assessment of Connecting Passenger Transfer (Chapter 16). REFERENCES National Plan of Integrated Airport Systems 19 84-1993. Federal Aviation Admin- istration, U.S. Department of Transportation, Aug. 1985. Guidelines for Airport Capacity/Demand Management. Airport Associations Coor- dinating Council, Geneva, Switzerland, and International Air Transport Associa- tion, Montreal, Quebec, Canada, Nov. 1981.

Appendix B Airport Landside Capacity Analysis Methods A brief review is given of a range of tools and procedures that may be useful to analysts and decision makers in assessments of airport landside capacity. The specific tools and procedures presented are a cross section of the field and are not recommended as preferred for any particular situation. A variety of rules of thumb, mathematical relationships, and computer- based simulations are available. Much of the research and many of the resulting analysis tools for terminal buildings and curb use the mathematical framework of queueing theory. Problems of motor vehicle access and parking are sometimes addressed by using techniques of highway transportation and traffic engineering. The airport landside capacity analyst choosing among alternative analysis methods must typically strike a balance between simplicity, speed, and ease of use, on the one hand, and more detailed representation of the facilities and services of interest, greater need for data and technically trained analysts, and cost, on the other. Although methods employing greater detail and more data are generally presumed to yield more reliable results, this does not hold true for forecast data, which are inherently uncertain. Analysis tools are most likely to be useful if they help analysts and decision makers to understand better the sources of current problems and the possible consequences of selecting among alternative solutions to these problems. Any airport's landside system is complex. There is some flexibility for the airport user to accommodate to current or anticipated conditions at the airport and to travel through the system even when queues are growing. Given the difficulty of representing such a situation, simple rules of thumb based on 165

observation at airports in operation may often be as useful as more complex mathematical analyses for assessing capacity. Such rules of thumb are invaria- bly easier to apply. Nevertheless the framework of queueing theory gives a useful structure to the analysis process. The airport terminal may be thought of as a series of active processors (e.g., baggage check, passenger security check, ticket coun- ter, baggage claim, and parking area) tied together by a series of holding and circulation areas (e.g., passenger waiting areas and corridors).1 As long as the average rate of arrival and the accumulated number of passengers wishing to pass through the processing element or holding or circulation area does not exceed the rate at which they can be accommodated, subject to standards defining this accommodation over the period of time in question, capacity is adequate. Using queueing theory, the interaction among various functional compo- nents of the terminal may be explored. If a capacity problem is relieved in one component, congestion and delay may then be observed in another component downstream. Queueing models help predict when such a progression of the problem is likely to occur. An alternative to the use of analytic models is simply comparing seemingly similar airports. The comparison may be a straightforward analogy or may use sophisticated statistical procedures to identify trends in data for many airports. Analysis methods have been developed by using both approaches. Al- though many of them were developed for terminal planning and design of new facilities, they may often be adapted to terminal landside capacity analysis. Landside capacity analysis often requires the services of a technically qualified professional. Many airports have such professionals on their staffs. Other airport operators may need the assistance of a consultant. This appendix is intended to aid the nontechnical user of these guidelines to work with technical analysts. The methods summarized here represent levels of detail ranging from broad planning to detailed design of terminal facilities. Assessment of a potential capacity problem may require several successively more detailed levels of analysis to identify specific problems and to suggest likely solutions. Staff or consultant analysts usually use several methods of the types reviewed here in the course of assessing the landside capacity of a particular airport. Methods are grouped into the following categories: Gate and apron utilization; Terminal buildings and connectors; Ground access, terminal curb, and parking; and Terminal system as a whole.

167 GATE AND APRON UTILIZATION Many analysis methods begin with a schedule of daily flight operations that gives scheduled flight arrival and departure times by airline and aircraft turnaround times. This information is typically presented in a ramp chart. Simplified procedures compress this information into average daily statistics. Analysis methods may yield a measure of how intensively a group of gates is utilized or determine whether additional flights can be accommodated. For larger gate complexes the problems of aircraft-gate compatibility become sufficiently complex to warrant computer simulation. The following methods are reviewed here: Direct calculation of gate needs Square-root rule Parsons gate-enplanement curve Average-to-peak utilization correction Gate-capacity graphic analysis Parsons apron area capacity estimate Ramp chart hourly utilization analysis Aggregate apron utilization efficiency Gate management simulation models Canadian gate assignment model Direct Calculation of Gate Needs (1) Procedure The peak-day aircraft fleet mix is listed, categorized by expected gate occupancy time. A weighted average service time is calculated. The single-gate-capacity index, in aircraft per minute per gate, is calcu- lated as the inverse of the weighted, average service time: c= weighted service time Overall capacity, in number of aircraft, is the product of this index and the total number of gates at the terminal:

i1rj Capacity = C x (gates) x 60 mm For exclusive gate use, the previous step is repeated for each group of gates under exclusive use. Total terminal capacity is then the sum of individ- ual group capacities. Commentary Use of this method implies 100 percent utilization and perfect schedule mesh. A slightly longer average occupancy time would allow for uncertainties and schedule mismatch. The maximum passenger capacity, if relevant, might be calculated by multiplying the flight capacity by the FAA-style equivalent aircraft (EQA) seating number for the airport's fleet mix computed with an average aircraft load factor of 1.0 (all seats full) [see Parson's manual (2)]. Although the calculation is simple and fast, the method yields indicative results only and application is limited to assessments. Square-Root Rule (3) Procedure A rule of thumb is cited regarding numbers of extra gates needed to allow for scheduling flexibility: Total gates required = (gates required by current schedule) + (square root of gates required by current schedule) This rule implies a margin of approximately 30 percent for a 10-gate airport, declining to 8 percent (12 gates) for a large 140-gate airport. An "effective" number of gates could be calculated from the total gates at an airport with this rule of thumb, which would then be used in the direct calculation method to determine practical capacity. Commentary The rule appears not to consider the degree of impact that exclusive gate use might have, but may be adequate as an approximation of current practice.

169 The rule would have to be applied separately for widely separated unit terminals or satellites. Computation would presumably start with a schedule of flights at the airport. Parsons Gate-Enplanement Curve 2) Procedure Plot of annual enplanements versus estimated gate positions may be inverted to read capacity for a given number of gates. Above approximately 4 million annual enplanements, relationship be- comes linear at 300,000 annual enplanements/gate. Curve appears approximately parabolic below 4 million annual enplane- ments, so that capacity = 12,000 x (gates)2 annual enplanements. Commentary Translation to peak-hour capacity would presumably be made with stan- dard peak-to-annual relationships, ideally specific to the airport in question. Relationship may be tested against current airport gate utilization data. Preliminary inspection of AOCI 1981 data suggests that curve may have merit compared with current operating practices. An independent verification of capacity limits would be desirable. The Parsons manual as a whole, although frequently criticized as very general and probably Out of date, is nevertheless widely used. Average-to-Peak Utilization Correction (4) Procedure Analysis of data at San Francisco International found that average gate utilization, measured in terms of time occupied on a basis of 6-min intervals, was approximately 84 percent of peak utilization. Such a factor could be used in conjunction with other procedures that yield annual enpianement capacities.

170 Commentary No claim of transferability can be made. Data from Palm Beach suggest average at approximately 64 percent of peak, perhaps indicative of the higher peaking at smaller or vacation-oriented arports. Easy correction factors similar to this should be developed by using data for other specific airports. Gate-Capacity Graphic Analysis (5) Procedure Given an average gate occupancy time for non-widebody flights and percent of non-widebodies in peak-hour flight schedule, hourly gate capacity operations base is read from graph. Gate size factor is read from graph, based on percentage of widebody aircraft in peak-hour flights and percentage of available gates able to handle widebodies. Gate capacity, in operations per hour, is computed as C = (operations base) x (size factor) x (number of gates) Commentary Full gate utilization is assumed. This method would be applied separately to each group of exclusive-use gates. Relatively easy to use, technically adequate, and officially sanctioned, this method may often be a good starting point for preliminary assessment of gate utilization problems. Parsons Apron Area Capacity Estimate (2, 6) Procedure Average space requirements are suggested for aircraft, including an inferred typical "all aircraft parking envelope" at 232 ft x 260 ft.

171 Total space required for parking is projected at 1.41 acres/aircraft, with a range from 1.0 acre for a DC-9 to 3.7 acres for a B-747. Commentary This approach uses straightforward computation and yields seemingly valid results but may seldom be relevant to capacity. Ramp Chart Hourly Utilization Analysis (7) Procedure A ramp chart is prepared for the average day, peak month, showing scheduled flight arrival and departure time by assigned gate. Gate occupancy in each hour is read from the ramp chart, or calculated as the number of aircraft on the ground in the previous hour less the number of departures in the previous hour plus the number of arrivals in the current hour divided by the number of gates. If occupancy is computed to be 100 percent during the governing time period, then effective capacity has been reached for that time period. Commentary If capacity is to be determined on an hourly basis only, this procedure may be unnecessary. Gate capacity is simply 100 percent occupancy with the airport's current ficet mix. Under exclusive gate use, an airline might have flights scheduled such that there is a gate available for a full (peak) hour, so that apparent occupancy would not be 100 percent, even though no other flights could be accommo- dated. This method requires assumptions about turnaround times to compute slot availability. Aggregate Apron Utilization Efficiency (8) Procedure A calculation is made of the plan area of a circle with diameter equal to the length or wingspan of the aircraft, whichever is greater.

172 The average use efficiency parameter, equal to the ratio of aircraft seat load to area of this circle, is calculated for the fleet currently operating at the airport. A similar calculation is made for the anticipated fleet mix. If average parameters (passengers per unit apron space) are significantly lower with the future fleet mix, then other design actions than addition of gates may be required to accommodate future passenger traffic growth. Commentary This may be a useful and quick test of impending space constraints. Gate Management Simulation Models Procedure In a computerized system, airport physical layout, flight schedule, and fleet characteristics are specified by the user in response to prompts. Model assigns flights to gates, either to minimize total gates in use or with preferential assignment. A full 24-hr day is simulated. Graphic outputs include gate utilization statistics and diagrammatic maps of terminal building gate positions with aircraft. Commentary Models are proprietary packages, available only through the consultants. Simulation algorithms and optimization procedures are undisclosed in available literature. Programs may be menu-driven and include graphic output, making them relatively user friendly. Canadian Gate Assignment Model (9) Queueing analysis is applied to the flight schedule, using a "first-arrived, first-assigned" algorithm, but with preference given to larger aircraft, airline gate preferences, and flight sector.

173 Model projects ahead to determine whether vacant gates should be held open for "more deserving" approaching aircraft. Up to a full operating day can be simulated according to user-specified assignment strategy and aircraft-gate compatibility to minimize aircraft and passenger delays. Commentary The model is programmed for IBM-compatible microcomputers. The model can be calibrated to accommodate common-use, exclusive- use, or preferential-use assignment strategies or combinations of these strat- egies. Currently the model is available for Transport Canada applications. TERMINAL BUILDINGS AND CONNECTORS Queueing theory underlies many of the models and rules of thumb for terminal buildings and connectors. Departing and arriving flows are generally considered separately, and attention must be given to those who accompany passengers through parts of the system. Capacity is determined in all cases essentially by the number of people who can be accommodated in a period of time. The following methods are reviewed here: Terminal concept capacity ranges Parsons manual planning standards Simple queueing formula Airport Terminal Simulation Model (ATSIM) and Terminal Area Capac- ity Simulation Model Canadian Terminal Simulation Model Terminal Concept Capacity Ranges (2, 6) Procedure Parsons manual and FAA circular give broad guidance on relevance of terminal concepts for airports expected to operate at various broad volume levels. See Figure B-i.

I I Recommended for Greater Than I— I I I I 75 perCflt Passenger Originations Cl) L - - LU I Transporter z I 0 I 0 I Satellite z ( I Multilevel Telnal 4)I I I Frontage and Building 0 I I LU LU Multilevel Connector ( and Boarding A I I 25,000 200,000 500,000 1,000,000 3,000,000 ANNUAL ENPLANEMENTS FIGURE B-i Capacity aspects of terminal concepts (2, 8).

175 If future operations are likely to grow beyond the transition levels where alternative concepts are workable, a capacity problem might be indicated. It is suggested that 1 million annual enplanements represents a basic transition of airport type, and that landside capacity issues are more likely to be substantial as traffic grows through the range of 0.5 to 1.0 million annual enplanements. Commentary Information should perhaps be considered within context of relationships between terminal siting and runway configuration (Figure B-2). As a simple broad indicator of generally accepted conventional wisdom this method may be a useful starting point for consideration of capacity problems. Parsons Manual Planning Standards (2, 6) Procedure Planning relationships and standards may be used to establish whether a particular airport's current operation is consistent with what had been planned to meet current demand. For example, planned gate utilization is permitted to grow to as much as 300,000 to 450,000 annual enplanements per gate. If the most heavily loaded 10 percent of gates at an airport exceed these levels annually, perhaps capacity is being approached. Standards may be inferred for the following additional elements of the system: - Baggage processing and claim areas; - Baggage claim equipment type and passenger service perimeter where bags are taken; - Concessions and lobby areas; - Ticket counter frontage and queueing areas; - Customs and immigration queueing area, manual processing positions, and total area; and - Gross terminal building floor area related to number of gates.

it -J <U) Widely spaced parallel Widely spaced parallel Z Single or closely placed parallel runways; limiting Runways with inlersecling axes; limiting apron terminal runways with intersecting crosswind runway or taxiway; runways with no intersecling crosswind runway; limiting apron z apron terminal on one side on two sides limiting apron terminal terminal on two sides, except cc on three sides as limited by taxiways o_ w U U) ,0 LU 0 Possible access from two o 3 Access from single point using one-way loop road Access from single point using one-way loop road Access from single point using one-way loop road points using two-way axial road with one-way loop roads serving each apron terminal area 0< ri z z U) Runway and roadway limit Runway and roadway limit Runway and roadway limit Runways limit expansion to 0 z< expansion to two directions expansion to two directions expansion to two directions two directions a-ui 0 Usually (but not limited to) Small, medium, or large Medium or large volumes Medium or large volumes 0 small or medium volumes volumes <0 Terminal Ea Apron . Runway ---- Ground access FIGURE B-2 Relationships between terminal site and runway configuration (2, 6).

177 Commentary Entire manual could in principle be "run in reverse" to compute annual and hourly enplanements from given physical facilities measures. However, age of material makes some of the relationships particularly open to question. Simple Queueing Formula (12) Procedure Major service points in the system are treated as single-channel or multichannel service facilities modeled by queueing theory. Passengers arrive at some known rate, are given service at some rate (generally assumed to be fixed at some average service time), and move to the next element of the system. Queues form if arrivals rate exceeds service rate. The queue length, expected total service time (including any waiting in a queue), and time for queues to clear may be calculated, and depend on the underlying mathematical distributions of arrival and service times assumed to apply. If passenger arrivals are assumed to be random, the Poisson process may be used. With service times assumed to be described by an exponential model, average delay time at the processing point and average number of people waiting for service (queue length) are calculated as follows: T= a s(s - a) N= a 2 s(s - a) where T = average waiting time, N = average number of people waiting, a = arrival rate, and s = service rate. Field observations or assumptions based on the design of the airport and its flight schedule may be used to estimate the parameters s and a during the peak hour. Calculated wait time and the ratio of calculated number of people waiting to the available waiting area may be compared with standards.

178 Similar assumptions yield a model of waiting time at baggage claim: T=t(2)+ nt(3) -1(1) n+1 where T = average waiting time, n = average number of bags per passenger, :(1) = expected length of time from flight arrival for all passengers to arrive in claim area, 1(2) = expected length of time from flight arrival for first luggage to reach claim area, and t(3) = expected length of time between arrival of first bags and last bags at claim area. Commentary Canadian government studies demonstrate how standardized accumula- tion curves could be developed for each waiting area in a particular airport. A major difficulty with this approach is that often passenger arrivals at service points are platooned rather than random. Assessment of a single hour for capacity determination is particularly sensitive to clustered arrivals. Airport Terminal Simulation Model (ATSIM) (Aviation Simulations International) and Terminal Area Capacity Simulation Model (Peat, Marwick, Mitchell and Co.) Procedure The airport terminal layout is blocked off into movement areas and nodes, each characterized by passenger processing statistics. Scenarios de- scribing passenger behavior then are specified for the model's discrete-event simulation of passenger movement through the system. The model routes passengers in response to the basic behavioral scenario and congestion en- countered because of earlier passenger arrivals. An alternative approach is based on time-stepped simulation, which operates relatively faster and with less computer capability required, but at some small loss of accuracy.

179 Commentary These models are examples of proprietary tools developed by a number of consultant organizations. Canadian Terminal Simulation Model (Transport Canada) Procedure Computation procedure not yet available. Commentary The present model will operate only on Transport Canada's mainframe computer. The program is being rewritten to run on IBM-XT-compatible microcomputers and is expected to be completed in 1987. GROUND ACCESS, TERMINAL CURB, AND PARKING Ground access, the terminal curb, and parking serve motor vehicles, and although some simple models and rules of thumb have been devised, detailed assessment of these components is likely to require assistance of a qualified traffic and transportation analyst. Methods of analysis are described in such standard publications as the Highway Capacity Manual (10), published by the Transportation Research Board, and the Transportation and Traffic Engineer- ing Handbook of the Institute of Transportation Engineers (11). Transportation professionals involved in airport planning and design have adapted some standard procedures to the specific needs of the airport. The terminal curb has been a particular focus of such work. Simulation models have been used to analyze parking lot operations. The following methods give an overview of the range of planning and design procedures that may be adapted to capacity assessment: Access capacity-to-demand index Parking requirement planning curve

180 Curbside queueing model Taxi Operations Simulation (TAXISIM) Curbside level-of-service planning method Access Capacity-to-Demand Index (13) Procedure A passenger demand index is calculated as PDI = [1.5 (daily passengers minus interline transfers) + 2 (no. of airport employees)}/1 ,000 This index is meant to approximate the number of trips per day made to the airport. An access supply index is calculated as PCI = 3.1 [effective lane capacity (vehicles/hr)]/1,000 This calculation implies an assumed average vehicle occupancy rate of 3.1 persons/vehicle. If the ratio PDIIPCJ is greater than 1.0, there is a potential capacity problem. Commentary Such indexes reflect the underlying principles for access capacity anal- ysis. The multipliers used could be changed to suit conditions at an individual airport. However, such indexes can only be first approximations, useful for initial screening for problems. Parking Requirement Planning Curve (3) Procedure Planning standards reflected in relationships such as that shown in Figure 12-1 may be "inverted" to estimate capacity of facilities in place. Given a number of available parking spaces, the implied range of accept- able passenger loads may be estimated.

181 Commentary Such relationships may be useful first indicators of problems, but may not fit the specific conditions at a particular airport. Curbside Queueing Model (14) Procedure The curbside operation is described as a standard queueing model and standard tables and graphs are used to solve for queue times or space requirements. Vehicle arrival rates and mean dwell times are required to calculate required curb length for given demand. Commentary The approach illustrates the analytical benefits of being able to use a simple single-channel queueing model to represent a part of the terminal landside. Nomographs are available or can be constructed for solving problems. Taxi Operations Simulation (15) Procedure Average passsenger arrival and trip characteristics are input to this computerized simulation model, along with fare and fee data. Taxis in use and average fleet utilization statistics are projected. Financial analysis of the projections indicates whether improved fleet profitability can be achieved without reduction of passenger service levels. Commentary Such models are typically developed for specific applications.

182 Curbside Level-of-Service Planning Method (6) Procedure Given peak-hour enplariing or arriving passengers and available curb frontage, level of service is projected by the graph shown in Figure 11-2. Levels of service are analogous to those in the Highway Capacity Manual. Underlying formulas and adjustment factors are used to respond to specific conditions at an airport, such as vehicle dwell time, arrival rate, and vehicle fleet mix. Commentary The procedure is based on an assumed steady rate of arrival of vehicles and passengers during the peak hour and is consistent with queueing models. TERMINAL SYSTEM AS A WHOLE Analyses of individual component capacity may fail to recognize important functional linkages within the system. Analysts have tried to overcome this problem by constructing simulations of the terminal as a whole. These models are typically complex computer simulations, reflecting the complexity of the terminal landside system. Similar results can in principle be achieved by linking separate component models together, but interaction among compo- nents may then be poorly represented. The following methods are reviewed here: Airport Landside Simulation Model (ALSIM) Spreadsheet equivalent capacity analysis Performance simulation using SLAM Airport Landside Simulation Model (ALSIM) (17) Procedure Details of flight schedule, ground transport demand and schedules, airline operations, airport operations, and passenger charcteristics are input to a

183 mainframe computer program. Mean service times and standard deviations are required for most items. The airport is modeled as a Markov network, with transition probabilities calculated from input data. Defaults are available for many parameters. Iterative simulation of passenger movements through the terminal land- side system produces estimates of expected peak-hour mean passenger delay and cumulative delay (in passenger-minutes) and annual cumulative delay. Statistics are shown for individual components as well as for the system as a whole. Commentary This model, developed by the FAA, was one of the most comprehensive attempts to model the terminal landside as a whole. It was intended to be generally applicable and was in the public domain. The model was not widely used and is now effectively dormant. Spreadsheet Equivalent Capacity Analysis (9) Procedure Each major component of the airport is represented as a simple processor or hold area; average service rate or holding capacity is specified. The system is presented in a standard accounting spreadsheet format. Maximum theoretical hourly capacity of each component is calculated and used to estimate a maximum theoretical airport service volume. Commentary This straightforward accounting framework is an example of the type of work made practical by interactive use of microcomputers. Although the underlying model is simplified compared with stochastic simulations, the results may be equally useful.

184 Performance Simulation Using SLAM (18) Procedure The terminal landside system is represented as a multi-channel queue service facility, using exponential and Erlang distributions of service times. A standard simulation language, in this case SLAM, is used to project consequences of various service volumes in terms of delay times (per pas- senger and cumulative), queue lengths, and utilization levels of servers. Commentary This approach is similar to that underlying the ALSIM model. In contrast to ALSIM, the user must be familiar with the simulation language and the technical assumptions involved in selecting service time distributions. NOTE 1. Transport Canada and IATA guidelines adopt a three-part characterization: pro- cessors, reservoirs, and links among the first two types of components. REFERENCES N. Ashford and P. H. Wright. Airport Engineering. Wiley and Sons, New York, 1979. Ralph M. Parsons Company. The Apron and Terminal Building Planning Report. Report FAA-RD-75-191. FAA, U.S. Department of Transportation, July 1975. R. deNeufville. Airport Systems Planning—A Critical Look at Methods and Experience. Macmillan Press, London, 1976. R. D. Belshe. A Study of Airport Terminal Gate Utilization. Graduate Report. Instiwte of Transportation and Traffic Engineering, University of California, Berkeley, Aug. 1971. Techniques for Determining Airport Airside Capacity and Delay. Report FAA F4RD-74-124. FAA, U.S. Department of Transportation, June 1976. Planning and Design Considerations for Airport Terminal Building Development. Advisory Circular 150/5360-7A. FAA, U.S. Department of Transportation, undated. F. X. McKelvey. Palm Beach International Airport: Interim Airport Operating and Use Plan. Aviation Planning Associates, Cincinnati, Ohio, 1984. W. Hart. The Airport Passenger Terminal. Wiley and Sons, New York, 1985. S. G. Hamzawy. Management and Planning of Airport Gate Capacity: A Micro- computer-Based Gate Assignment Simulation Model. Presented at 65th Annual Meeting of the Transportation Research Board, Washington, D.C., 1986.

185 Special Report 209.' Highway Capacity Manual. TRB, National Research Council, Washington, D.C., 1985. Transportation and Traffic Engineering Handbook, 2nd ed. Institute of Transporta- tion Engineers, Washington, D.C., 1982. R. Horonjeff and F. X. McKelvey. Planning and Design of Airports, 3rd ed. McGraw-Hill, New York, 1983. E. M. Whitloclç H. M. Mirsky, and F. LaMagna. Ground Transportation Planning Implications of Airline Shuttle Passengers. In Transportation Research Record 499, TRB, National Research Council, Washington, D.C., 1974, pp. 47-57. R. Tiles. Curb Space at Airport Terminals. Traffic Quarterly, Oct. 1973, pp. 5 63-5 82. R. A. Mundy, C. J. Langley, and L. Stulbert. Determination of the Appropriate Number of Taxicabs to Serve an Airport. In Transportation Research Record 1025, TRB, National Research Board, Washington, D.C., pp. 14-22. P. B. Mandle, E. M. Whitlock, and F. LaMagna. Airport Curbside Planning and Design. In Transportation Research Record 840, TRB, National Research Coun- cil, Washington, D.C., 1982, pp. 1-6. L. McCabe and M. Gorstein. Airport Landside. Vols. 1-5. Report DOT-TSC- FAA-82-4-(1-5). FAA, U.S. Department of Transportation, 1982. S. Mumayiz and N. Ashford. Methodology for Planning and Operations Manage- ment of Airport Terminal Facilities. In Transportation Research Record 1094, TRB, National Research Council, Washington, D.C., 1986, pp. 24-35.

Study Committee Biographical Information MARJORIE BRINK, Chairman, has had 25 years' experience as an airport planning consultant. She graduated from Indiana University and was a mem- ber of Phi Beta Kappa. She has prepared basic research studies leading to the airside-landside separation concept used for developing the world-renowned passenger terminal facilities at Tampa International Airport, which was opened in 1971. (The terms "airside" and "landside" were created as a part of this research project and have since become standard terminology in the industry.) Ms. Brink is responsible for a number of airport master plans that involved comprehensive analysis of landside capacity, including those for major airports in Tampa, Portland, Tucson, and Nashville. The author of many reports and articles, she is a principal with Peat Marwick Airport Consulting Services. MARGARET M. BALLARD is a Senior Planner with the Sverdrup Corporation. She received a B.A. from Goucher College and an M.S. in Urban Planning from Johns Hopkins University. Before joining the Sverdrup Corporation, she was Manager of Planning and Environmental Services with the Maryland State Aviation Administration. Ms. Ballard is a member of the American Planning Association and the Women's Transportation Seminar. GEORGE J. BEAN is Director of the Hillsborough County Aviation Authority, which owns the Tampa International Airport. His civil aviation career began after World War II when he joined Northeast Airlines in Worcester, Mas- sachusetts. Appointed Manager of Worcester Airport in 1953, he later became 186

187 manager of the Greater Wilmington Airport and then of the Tampa Airport. Mr. Bean is Past President of the American Association of Airport Executives and the Airport Operators Council International. He has received the AAAE President's Award for his outstanding contributions to the airport management profession. Mr. Bean is a past member of the TRB Executive Committee. FRANK T. BISHOP is currently Manager, Properties, at Houston Intercontinen- tal Airport. A graduate of Westchester Community College in Valhalla, New York, he was a flight operations specialist in the U.S. Army until becoming Supervisor of Administration for Stewart Airport in Newburgh, New York. He then became Assistant Director of Aviation at the Austin Municipal Airport and served for a time as Acting Director. He is a board member of the American Association of Airport Executives and President of the South Central Chapter. A Past President of the Austin Society of Public Administra- tion, Mr. Bishop is also a certified commercial pilot with multi-engine and instrument ratings. GEORGE W. BLOMME is Manager of Aviation Marketing for the Port Au- thority of New York and New Jersey. He received his B.S. in Civil Engineer- ing and an M.S. in Transportation from Northwestern University. He has been with Port Authority for 26 years, involved in airport access, landside planning, and airport maintenance. He serves as Chairman of TRB 's Committee on Airport Landside Operations and is Past Chairman of the Committee on Passenger Terminals for the American Society of Civil Engineers. Mr. Blomme coauthored Airport User Traffic Characteristics for Transportation Planning for the Institute of Transportation Engineers and Technology of Urban Transportation for the Automobile Manufacturers Association, THOMAS H. BROWN is Manager of Facilities Planning and Design with United Airlines. After receiving a B.S. from Oklahoma State University, an M.S. from the University of Southern California, and an M.S.C.E. from Massachusetts Institute of Technology, he served with the U.S. Air Force for five years. Before joining United Airlines, he was with the Right Transporta- tion Laboratory at MIT, Chief Aviation Planner for the Chicago Area Trans- portation Study, and Senior Managing Associate and Vice President of Arnold Thompson Associates, Lester B. Knight & Associates, Inc. WILLIAM C. COLEMAN is Vice President and New England Regional Man- ager of Public Finance for Smith Barney. He received a B.A. and an M.B.A. from Harvard University. He has been on the management staff of the Department of Health and Human Services with the Commonwealth of Massachusetts, Manager of Clinical Services with Massachusetts General

188 Hospital, and Director of Administration and Finance and then Director of Aviation of the Massachusetts Port Authority, which owns and operates Boston's Logan International Airport. He is on the Board of Directors of the Airport Operators Council International. KENNETH McK. ELDRED is Director of Ken Eldred Engineering in Concord, Massachusetts. He received his B.S. from Massachusetts Institute of Technol- ogy and was connected with Bolt Beranek and Newman for 8 years as Vice President, Principal Consultant, and Director of the Architectural Technology and Noise Control Division. Mr. Eldred is a specialist in evaluating and finding practical solutions to problems in vibrations and acoustics, particularly environmental noise. In addition to being a member of the National Academy of Engineering, he chairs the American National Standards Institute Executive Standards Council and is active in many professional societies. JOHN GLOVER is Director of Strategic and Management Planning at the Port of Oaldand, California. He received a B.A. in architecture and an M.S.C.E. from the University of California at Berkeley. Mr. Glover directs long-range planning, research, financial planning, and special projects for aviation, mar- itime, and properties management activities of the Port. During his 17 years with the Port, he has been responsible for all areas of airport planning, including master planning, terminal planning and programming, noise studies, environmental analyses, and ground access studies and projects. He is a member of the American Planning Association, Airport Operators Council International, Aircraft Owners and Pilots Association, and The Planning Forum. ADIB KANAFANI is Professor of Transportation Engineering and Director of the Institute of Transportation Studies at the University of California, Berkeley. He received his B.S. in Engineering at the American University in Beirut and his M.S. and Ph.D. in Civil Engineering and Transportation at the University of California, Berkeley. Over the past 20 years Dr. Kanafani has been involved in many aspects of transportation research with a special expertise in aviation issues. He has studied all aspects of aviation: aviation demand, airport feasibility studies, capacity analysis, and airline economics. Dr. Kanafani has also consulted directly with airports all over the world. He is Associate Editor of Transportation Research and two other transportation journals, and has authored or coauthored numerous articles on aviation. He received the Walter Huber Research Prize from the American Society of Civil Engineers in 1984. PETER B. MANDLE is a Senior Consultant in the airport consulting practice of Peat Marwick Main & Co. in San Francisco. After receiving a B.S. and an

189 M.S. in Civil Engineering from Clarkson College and the University of Connecticut, respectively, he performed and directed numerous studies in airport planning for curb frontage, parking, roadways, transit access, and traffic circulation projects, including major projects at J. F. Kennedy, Miami, Denver Stapleton, Dallas/Fort Worth, Newark, Orlando, and Tucson Interna- tional Airports and at La Guardia Airport. He has coauthored several technical papers on traffic engineering and is Secretary of the TRB Committee on Airport Landside Operations and Chairman of the Airport Landside Planning Committee of the American Society of Civil Engineers. DORN CHARLES MCGRATH, JR., is Chairman of the Department of Urban and Regional Planning at George Washington University. He received degrees from Dartmouth and Harvard and has had considerable experience in airport land use issues, including serving as consultant to several major airports, participant in an NRC study on the environmental impact of the expansion of J. F. Kennedy International Airport into Jamaica Bay, and member of the Office of Technology Assessment Airport Systems Development Study. He has authored papers on airport noise and land use planning and is a member of the TRB Working Group on Airport Macrosystems for the United States. He is Chairman of the Committee of 100 on the Federal City, is active in the American Institute of Certified Planners, and is a TRB University Representative. FRANCIS X. MCKELVEY is Associate Professor of Civil Engineering at Michigan State University. He received a B.S. and an M.S. in Civil Engineer- ing from Manhattan College and New York University, respectively, and a Ph.D. from Pennsylvania State University. Closely involved in airport airside and landside issues for the last 10 years, he has participated in numerous research projects and written many articles. He is coauthor of the third edition of Planning and Design of Airports and is a registered engineer in several states. ROBERT S. MICHAEL is General Manager for the Regional Airport Authority of Louisville and Jefferson County, Kentucky. After receiving a B.A. from Dartmouth College, Mr. Michael served as Airport Manager in Billings, Montana; Airport Director for Milwaukee County, Wisconsin; and Director of Aviation for the City and County of Denver, Colorado. He is an executive member of the American Association of Airport Executives, Past President of the Airport Operators Council International, and Past Chairman of the Board of Examiners of the AAAE. RAY A. MUNDY is Professor, Department of Marketing and Transportation, University of Tennessee, and Executive Director of Airport Ground Transpor- tation. He received a B.A. and an M.B.A. from Bowling Green State

190 University and a Ph.D. in Transportation and Logistics from Pennsylvania State University. He has directed several major research projects in public transportation and has served as consultant to several airports in determining techniques for offering private delivery systems for public transportation. The author of several reports and articles, he is also Editor of Airport Ground Transportation, Problems and Solutions and Airport Ground Transportation News. E. WAYNE POWER is Deputy Project Manager for development of Terminal 3 at Toronto's Lester B. Pearson International Airport. Formerly, he was Chief of the Airport Development Services Division of Transport Canada. This division specializes in providing airport planning services for the Canadian airport system, including airport master planning, requirements studies, termi- nal capacity/demand analyses, and major capital projects planning. Transport Canada has developed processes and models for Canadian airport planning. J. DONALD REILLY is Executive Director and Secretary General of the Airport Operators Council International. He received a B.S. in Economics from Franklin and Marshall College and an L.L.B. from George Washington University Law School. Mr. Reilly has been with the AOCI since 1961. In 1965 he was appointed Director of Legal Services, a post he held until he was made Executive Director in 1968. He serves as Chairman of the Industry Task Force on Airport Capacity Improvement and Delay Reduction, which reports to the Administrator of the Federal Aviation Administration on many operat- ing procedures and research needs to improve air transportation capacity, with a primary focus on the airside. JAMES W. SPENSLEY is Project Manager for the Stapleton Airport Expansion Project in Denver, Colorado, which includes planning of the only major new airport being actively developed in the United States at this time; President of Spensley & Associates, Ltd., a management and environmental consulting firm; and Adjunct Professor in Environmental Law at the University of Denver. He received a B.S. in Industrial Engineering from Iowa State Univer- sity and a J.D. from George Washington University Law School. A former General Counsel for the University Corporation for Atmospheric Research in Boulder, he also served as Staff Director and General Counsel for several subcommittees of the House Committee on Science and Technology.

The Transportation Research Board is a unit of the National Research Council, which serves the National Academy of Sciences and the National Academy of Engineering. The Board's purpose is to stimulate research concerning the nature and performance of transportation systems, to dis- seminate the information produced by the research, and to encourage the application of appropriate research findings. The Board's program is car- ried out by more than 270 committees, task forces, and panels composed of more than 3,300 administrators, engineers, social scientists, attorneys, educators, and others concerned with transportation; they serve without compensation. The program is supported by state transportation and high- way departments, the modal administrations of the U.S. Department of Transportation, the Association of American Railroads, the National High- way Traffic Safety Administration, and other organizations and individuals interested in the development of transportation. The National Academy of Sciences is a private, nonprofit, sell-per- petuating society of distinguished scholars engaged in scientific and engi- neering research, dedicated to the furtherance of science and technology and to their use for the general welfare. Upon the authority of the charter granted to it by the Congress in 1863, the Academy has a mandate that requires it to advise the federal government on scientific and technical mattters. Dr. Frank Press is president of the National Academy of Sciences. The National Academy of Engineering was established in 1964, under the charter of the National Academy of Sciences, as a parallel organization of outstanding engineers. It is autonomous in its administration and in the selection of its members, sharing with the National Academy of Sciences the responsibility for advising the federal government. The National Acad- emy of Engineering also sponsors engineering programs aimed at meeting national needs, encourages education and research, and recognizes the superior achievements of engineers. Dr. Robert M. White is president of the National Academy of Engineering. The Institute of Medicine was established in 1970 by the National Academy of Sciences to secure the services of cminent members of appropriate professions in the examination of policy matters pertaining to the health of the public. The Institute acts under the responsibility given to the National Academy of Sciences by its congressional charter to be an adviser to the federal government and, upon its own initiative, to identify issues of medical care, research, and education. Dr. Samuel 0. Thier is president of the Institute of Medicine. The National Research Council was organized by the National Academy of Sciences in 1916 to associate the broad community of science and technology with the Academy's purposes of furthering knowledge and advising the federal government. Functioning in accordance with general policies determined by the Academy, the Council has become the principl operating agency of both the National Academy of Sciences and tHe National Academy of Engineering in providing services to the government, the public, and the scientific and engineering communities. The Council is administered jointly by both the Academies and the Institute of Medicine. Dr. Frank Press and Dr. Robert M. White are chairman and vice chairman, respectively, of the National Research Council.

Transportation Research Board National Research Council 2101 Constitution Avenue, N.W. Washington, D.C. 20418 NON-PROFIT ORG. U.S. POSTAGE PAID WASHINGTON, D.C. PERMIT NO. 8970 ADDRESS CORRECI1ON REQUESTED U') CII >-wn 'w-. LL 0 V) — c _Jo . c' _j v, o:x: Zt) —< 'GO'- OcZUJ 0, C —c-c 0 EI