Preparing Peak Period and Operational Profiles—Guidebook (2013)

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.

39 This chapter discusses how to use the Toolbox to calculate existing design day profiles of aircraft operations and passengers and provides guidance for estimating future profiles. This chapter includes sample problems and provides some comments and cautions on the use of the data. Design day profiles are focused on the distribution of aircraft operations or passenger activity throughout the design day. They are useful when the level of analysis is more involved, especially for terminal building and landside analysis. Activity levels at different facilities lead and lag passenger enplanement or deplanement activity by differing amounts depending on the facility in question. Design day profiles permit facility-specific peaks to be estimated more accurately than by using simple peak activity estimates. These profiles can be prepared for either a design day or an AAD. 5.1 Operations 5.1.1 Existing Design Day Operations Profiles An approach to estimating existing aircraft operation design day profiles is provided in Exhibit 5.1. Methods of estimating design day operations are described in Chapter 4. The data input day is a representative day for which complete aircraft operations data, with arrival or departure times identified for each operation, can be obtained. The data input day need not be the same as the design day, since it can be adjusted to represent the design day. Potential sources include noise monitoring data, if collected by the Airport, or Enhanced Traffic Management System (ETMS) or Performance Data Analysis and Reporting System (PDARS) data from the FAA. The input schedule data is organized by hour or other time period and then scaled to represent the design day. Exhibit 5.2 describes an approach for estimating future design day profiles of aircraft operations. The calculation of future design day operations is discussed in Chapter 4. Those operations, in conjunction with the existing design day profile, can be used to estimate a future design day profile of aircraft operations, provided no peak spreading is expected. If peak spreading is anticipated, a peak spreading factor can be applied that lessens the magnitude of both the peaks and valleys in the design day profile, as typically occurs with busier airports. The peak spreading factor is based on the tendency of hourly activity at airports to converge towards the average activity level of an hour during an operational day. See Appendix H in the ACRP WOD 14 for additional discussion. C h a p t e r 5 Preparation of Design Day Profiles

40 preparing peak period and Operational profiles—Guidebook Exhibit 5.1. Process for estimating existing design day profile of aircraft operations. Estimate Arrival and Departure Operations Design Day Operations Profile Operations Data by Time of Day for Data Input Day User Definition of Design Day Existing Design Day Operations Calculate Ratio of Design Day to Data Input Day Operations Data Input User Determined Assumptions Intermediate Output Final Output Exhibit 5.2. Process for estimating future design day profile of aircraft operations. Estimate Future Unadjusted Design Day Profile of Operations Estimate Future Design Day Profile of Operations Adjusted for Peak Spreading Future Design Day Profile of Aircraft Operations Future Design Day Operations Existing Design Day Profile Peak Spreading Factor Data Input User Determined Assumptions Intermediate Output Final Output

preparation of Design Day profiles 41 5.1.2 Use of Toolbox The operations module of the Toolbox provides a quick method of estimating a future profile of design day operations. The following sample problem shows how to use the Toolbox to calculate a forecast profile of design day operations. Sample Problem 5.1 Problem Statement: Validate Design Day Flight Schedule prepared by planning consultant Airport: Large Commercial Airport in Mountain Time Zone Forecast Element: Design Day Profile of Operations Toolbox Module: Operations In this instance, an airport planner needs to review a forecasted design day schedule prepared by a consultant. The consultant has assured the planner that the schedule incorporates a reason- able amount of peak spreading to ensure that future capacity issues are not overstated. The planner needs an independent crosscheck of this assumption and uses the Toolbox to prepare a design day profile that incorporates peak spreading as a benchmark to test the design day schedule. The airport planner wants to check peak spreading for both arriving and departing operations and therefore runs the Toolbox twice, once in arrival mode and once in departure mode. The example in Exhibit 5.3 shows arrival mode (A). The planner also inputs the forecast of annual operations for the year (B) and the design day definition (10 percent in this case) (C). The peak period definition, and begin/end time designation (D) can be ignored since no peak period analysis is required for this effort. The airport planner then moves to the C. Base Year Data worksheet and imports arrivals and departures by day of the year and arriving and departing operations by time of day for a represen- tative day (see Exhibit 5.4). In this instance, operations by day of the year were obtained from the FAA’s ETMS database and the schedule data for the representative day were obtained from the Air- port’s Noise and Operations Monitoring System (see Appendix B for more detail on data sources). The F. Design Day Profile spread worksheet provides the forecasted design day profile incorporating peak spreading. The outputs for both arrivals and departures are provided in Exhibits 5.5 and 5.6. With the above output the planner can proceed with evaluating the forecasted gated flight schedule, paying particularly close attention to the arrival peak hour, which occurs at 20:00–20:59 (8:00–8:59 PM) and falls from 9.8 percent to 9.1 percent of the total. The planner also closely examines the departure peak hour which occurs at 17:00–17:59 (5:00–5:59 PM) and falls from 9.0 percent to 8.5 percent of the total. 5.2 Design Day Passenger Profiles 5.2.1 Existing Design Day Passenger Profiles The general approach to estimating existing passenger profiles is similar to the approach for estimating aircraft operation profiles. Since hourly data on passenger levels is typically unavailable, additional steps are required for the estimate. Exhibit 5.7 shows an approach for estimating existing design day passenger profiles.

42 preparing peak period and Operational profiles—Guidebook Exhibit 5.3. Aircraft operations mode.

preparation of Design Day profiles 43 To estimate design day passenger profiles, the user must provide design day passenger levels (see Chapter 4) and a distribution of scheduled seat arrivals and departures from the OAG (see Appendix B for detail on data sources). Ideally the OAG distribution of seat arrivals and departures would be obtained from the date corresponding to the existing design day identified from the design day analysis (Chapter 4). Referring back to Sample Problem 4.2 in Chapter 4, the user selected the 10 percent threshold to represent the design day, and the D. Design Day worksheet (see Exhibit 4.12) identified August 16, 2009 as representing the design day in the base year. Therefore, an OAG schedule for August 16, 2009 should be selected for the design day profile analysis in this case; however, the Toolbox is designed to adjust seat distributions from alternative days if data for the specific design day are not available. Once the design day schedule information is entered, the average load factor for the data input day is then applied to the scheduled seat arrivals and departures to generate passenger distributions by time of day, which are then scaled so that the daily sum matches design day passenger levels. If reliable data on load factor distributions by time of day are available, the design day profile can be further defined (See Appendix E in ACRP WOD 14 for additional discussion). The design day passenger profile can be converted to a design day O&D passenger profile by applying a ratio of originations to enplanements. When planning for facilities driven by local Exhibit 5.4. Base year data.

44 preparing peak period and Operational profiles—Guidebook Exhibit 5.5. Design day profile: arrivals.

preparation of Design Day profiles 45 Exhibit 5.6. Design day profile: departures.

46 preparing peak period and Operational profiles—Guidebook (O&D) passenger demand such as security, check-in, baggage claim, parking, and roadways, additional refinements to estimate design day profiles of O&D demand may be required. Absent data from the airline, the following approaches can be used to refine estimates of the O&D share of total passengers by hour: • Segregate the passenger data by airline and then estimate hourly O&D traffic for each airline. This way, hours that are dominated by airlines that do not provide connecting service will reflect the high O&D share that characterizes these airlines. • If there are no redeye flights, flights that depart prior to the first arrival bank will have virtually no connecting traffic. Flights that arrive after the last departing bank will likewise have virtually no connecting traffic. • If the design day profile is derived from a design day schedule, additional market variations in O&D shares can be incorporated. See Section 6.3 for additional detail. • At “spoke” airports where virtually all passenger traffic is O&D, local passenger traffic will essentially be the same as enplaning and deplaning traffic, and the passenger and O&D passenger design day profiles will be very similar. Estimate Arriving and Departing Passengers by Hour Design Day Passenger Profile Scheduled Seat Data by Time of Day for Data Input Day User Definition of Design Day Existing Design Day Passengers Calculate Ratio of Design Day to Data Input Day Passengers Load Factor Adjustment by Time of Day Lead/Lag Passenger Distributions Derivative Design Day Passenger Profile Origination to Enplanement Ratio Design Day O&D Passenger Profile Derivative Design Day O&D Passenger Profile Data Input User Determined Assumptions Intermediate Output Final Output Exhibit 5.7. Process for estimating existing design day passenger profile.

preparation of Design Day profiles 47 As noted in Chapter 2, the peak passenger impact on many terminal and landside facilities occurs at a different time than the enplanement and deplanement peaks. Enplaning passengers will stress these facilities for a period of time prior to scheduled aircraft departure while deplaning passengers will stress these facilities for a period of time after aircraft arrival. The lead and lag times will differ depending on the facility, and there will be a probability distribution associated with each lead and lag time. If estimates of these distributions are available, derivative design day passenger profiles tailored to each facility can be estimated from the original design day passenger profile. 5.2.2 Forecast Design Day Passenger Profiles Exhibit 5.8 shows an approach for estimating future design day passenger profiles. The pro- cess can be used to estimate future design day passenger profiles, O&D passenger profiles, and derivative passenger profiles. The calculation of future design day passengers is discussed in Chapter 4. Those passenger projections, in conjunction with the existing design day profile, can be used to estimate a future design day profile of aircraft operations, provided no peak spreading is expected. If peak spreading is anticipated, a peak spreading factor can be applied that reduces the magnitude of both the peaks and valleys in the design day profile, as typically occurs with busier airports. The peak spreading factor is based on the tendency of hourly activity at airports to converge towards the average activity level of an hour during an operational day. See Appen- dix H in the ACRP WOD 14 for additional discussion. 5.2.3 Use of the Toolbox The passenger module of the Toolbox provides a quick method of estimating a future profile of design day passengers. The following sample problem shows how to use the Toolbox to calculate a derivative design day passenger profile. Estimate Future Unadjusted Design Day Passenger Profile Estimate Future Design Day Passenger Profile Adjusted for Peak Spreading Future Design Day Passenger Profile Future Design Day Passengers Existing Design Day Profile Peak Spreading Factor Data Input User Determined Assumptions Intermediate Output Final Output Exhibit 5.8. Process for estimating future design day passenger profile.

48 preparing peak period and Operational profiles—Guidebook In this case, an environmental planner needs to quickly estimate future airport vehicular traffic for air quality dispersion analysis. After coordination between the planner, the FAA, and local air agency, it was agreed that because the pollutant of concern focused on the frequency that an air quality standard is exceeded, a dispersion analysis would be conducted for the design day of the facility. The planner knows that O&D passengers are the primary determinant of airport related vehicular traffic but also knows that departing passengers arrive at the airport well before their flight leaves and that there is also a lag between the time an aircraft arrives and the deplan- ing passenger leaves the airport. Therefore, airline schedules by themselves are not an accurate determinant of the hourly distribution of vehicular traffic. The Toolbox is used to generate a derivative design day profile that reflects the differences in the hourly distribution of activity at the airport roadways versus the airport gates. The planner goes to the B. User Parameters worksheet and enters data for forecast annual passengers (see Exhibit 5.9). Since the planner is interested in both arriving and departing Exhibit 5.9. User parameters worksheet. Sample Problem 5.2 Problem Statement: Estimate future daily distribution of airport-related vehicular traffic for air quality dispersion analysis. Airport: Large Commercial Airport in Pacific Time Zone Forecast Element: Derivative Design Day O&D Passenger Profile Toolbox Module: Passenger

preparation of Design Day profiles 49 Exhibit 5.10. User parameter worksheet: enplanements and load factors. passengers, “both” is selected from the drop down menu (A) when selecting between arriving/ departure type. After consultation with relevant agencies, the 10 percent busiest day is decided as an appropriate busy day for the air quality analysis. Therefore, the Design Day is not defined as the Average Day of the Peak Month and a 10 percent definition is selected instead (B). Since peak period information is not needed, the peak period input parameters are skipped, but the planner provides data on average monthly enplanements (C) and day-of-the-week load factor adjustments (D) from Appendix C (see Exhibit 5.10). Since connecting passengers don’t generate ground vehicle trips, the planner specifies that the Toolbox is to calculate O&D traffic (E) which accounts for 75 percent of total passenger traffic (F) at her airport. Finally, the planner incorporates the lead (G) and lag (H) distributions from a recent passenger survey conducted by the airport (see Exhibit 5.11). The planner moves on to the Base Year Data worksheet (Exhibit 5.12) and provides OAG schedule data by flight for a representative day, and scheduled seat arrivals and departures for each day of the year as shown previously. The environmental planner believes the airport will experience an average amount of peak spreading, and therefore moves past worksheet F. Derivative Profile—NS, which shows the deriva- tive O&D passenger distribution without peak spreading and looks to worksheet H. Derivative Profile—PS (Exhibit 5.13), which does incorporate peak spreading, for results. Using ratios of passengers to vehicle trips obtained from a previous airport survey, the planner now has the information needed to generate an hourly distribution of airport-related vehicle trips. It’s noted that the derivative peak hour occurs at 11:00 AM and is projected to account for 7.6 percent of daily activity. In passing, a glance at the non-derivative design day profile [G. Design Day Profile—PS worksheet (not shown)] shows that the airside peak hour occurs at 12:00 PM and only accounts for 6.8 percent of design day activity. Without accounting for the different distributions of landside and airside activity, the planner would have underestimated the landside peak.

50 preparing peak period and Operational profiles—Guidebook Exhibit 5.12. Base year data worksheet. Exhibit 5.11. User parameter worksheet: lead and lag distribution.

preparation of Design Day profiles 51 Exhibit 5.13. Derivative profile worksheet.

52 preparing peak period and Operational profiles—Guidebook After reviewing the analysis, the Airport Director asks whether a change in security procedures could affect the arrival distribution sufficiently to change the results. Since future security proce- dures and their impacts on arrival distributions are difficult to predict, the planner and the planner’s staff meet and use their professional judgment to develop a wide range of potential lead distributions for scenario testing. The Toolbox allows them to quickly test each of the distribution scenarios. They test many scenarios, two of which are illustrated here. Scenario A assumes the terrorism threat declines significantly and security procedures are streamlined to reflect the reduced threat. The assumed lead distribution for this scenario is shown in Exhibit 5.14 and the results are shown in Exhibit 5.15. Scenario B assumes a much more stringent security environment that requires passengers to arrive much earlier than they do now. The assumed lead time distribution for this scenario is presented in Exhibit 5.16 and the results are presented in Exhibit 5.17. Peak hour passenger O&D traffic at the curb ranges from 7,677 under Scenario A to 7,640 under Scenario B. These numbers compare to 7,844 with the baseline distributions. After examining the remaining scenario results, the planner reports to the Director that changes in security procedures are unlikely to materially change the peak hour impacts and in fact may reduce them slightly. 5.3 Comments and Cautions When preparing design day profiles the following comments and cautions should be noted: • If the data and resources are available, consider segmenting the categories when preparing forecasts of design day profiles. For example, domestic and international passengers have and Exhibit 5.14. Lead distribution scenario.

preparation of Design Day profiles 53 Exhibit 5.15. Lead distribution results.

54 preparing peak period and Operational profiles—Guidebook are expected to continue to grow at different rates. Moreover, international traffic, because of the more limited windows, has a very different and more pronounced profile than domestic traffic. When aggregated, categories with differing growth rates and differing profiles can result in a marked change in the overall profile. • The Toolbox assumes the relationship between the number of total passengers and O&D passengers is the same throughout the day. In reality, the distribution of O&D traffic at airline connecting hubs will differ from the distribution of total passenger traffic. Data on the hourly distribution of local/connecting percentages is very scarce and would need to be obtained from the airlines. Approaches to estimating local/connecting percentages by time of day are provided in more detail in Section 6.3. • Also note that it may be necessary to distinguish between resident and non-resident O&D traffic. This may have some relevance for terminal facilities, because resident passengers will likely be more familiar with the airport and have shorter lead times. The distinction is much more relevant for landside facilities and ground transportation planning. Residents and non-residents tend to use very different transportation modes. Residents mostly use their private automobile to arrive at the airport and are the predominate user of parking facilities. Non-residents are the predominant user of rental cars. Exhibit 5.16. Assumed lead time distribution.

preparation of Design Day profiles 55 Exhibit 5.17. Assumed lead time results.