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.

84 This chapter shows how physical or policy constraints may affect the magnitude of the peaks and the distribution of daily activity and provides guidance on how to model these aspects. One of the more difficult aspects of estimating future peak period activity or operational profiles is how to best incorporate the impact of airport constraints. Addressing constraints in annual forecasts is beyond the scope of this guidebook, so the focus will be on how to evaluate the impact of constraints on a daily or less-than-daily basis. Even if an airport fully intends to expand to accommodate increased demand, constrained forecasts are often necessary to identify no-action (without project) operational conditions as a baseline which can be used to measure project benefits and environmental impacts. Constraints can be imposed by insufficient physical infrastructure or by policy restrictions. Physical constraints can include: • Airfield constraints such as the lack of runway, taxiway, and queuing capacity; • Terminal constraints such as the lack of gate capacity, or chokepoints within the terminal building, such as security clearance, people mover systems, and baggage processing; and • Landside constraints, such as limits on roadway, curbside, and automobile parking capacity. Policy constraints can include: • Slot controls; • Peak period pricing (higher landing fees during busy times); and • Nighttime noise restrictions or curfews. These differing constraints will affect peak or operational activity in different ways. 9.1 Airfield Constraints The hourly throughput capacity of an airfield needs to be estimated before the impact of airfield constraints can be analyzed. This is best done using an airfield simulation model such as SIMMOD or TAAM. The result will not be a single capacity number, but rather a relationship that can be graphed between aircraft arrival capacity and departure capacity. Increased arrival capacity will come at the expense of decreased departure capacity, and vice versa. Chen and Guldingx authored a paper, “Assessment of System Constraints for Producing Constrained Feasible Schedules,” that provides a means of estimating the impact of airfield throughput capacity on airline schedules. The authors identified the existing relationship between scheduled demand and existing visual meteorological conditions (VMC) capacity at John F. C h a p t e r 9 Application of Constraints

application of Constraints 85 Kennedy International (JFK) and Newark Liberty International (EWR) Airports. The authors noted from previous FAA work that although scheduled demand could exceed capacity for short periods of time, significant excesses of demand over capacity were not sustainable over long periods of time. For example, the maximum demand/capacity ratio could be as high as 1.41 for a 15-minute period, but fell to 1.21 for a one-hour period, 1.14 for a two-hour period, and 1.06 for a three-hour period (see Exhibit 9.1). Assuming airport throughput capacity estimates are available, these ratios can be used to establish upper limits on the number of aircraft operations scheduled for each period. See Example 9.1 for a sample calculation. Multiplying the throughput capacity by the ratio will provide a control total for operations that will limit the extent of the peaks in constrained daily profiles and design day schedules. If airfield capacity cannot accommodate unconstrained demand, airlines have several potential options for reducing flights: • Reschedule flights to off-peak hours; • Increase the size of the aircraft serving a market, while reducing frequency; • Increase load factors so that more passengers can be flown with a given amount of seat capacity; • At connecting airports, divert connecting passengers through other hubs or gateways in their network; • Increase fares to reduce the number of O&D passengers to be accommodated; or • Cease or reduce service to certain markets. Exhibit 9.1. Maximum demand/capacity ratio by peak period definition. Peak Period Definition Maximum Demand/Capacity Ratio (Aircraft Operations) 15 minutes 1.41 1 hour 1.21 2 hours 1.14 3 hours 1.06 Example 9.1. Applying demand/capacity ratios. Assumption: Departure Throughput Capacity = 80/hr 15 minute maximum departures = 80 dept/hr × (15 min/60min) × 1.41 = maximum of 28.2 scheduled departures in a 15 minute period. 1 hour maximum departures = 80 dept/hr × (60 min/60min) × 1.21 = maximum of 96.8 scheduled departures in a 1 hour period. 2 hour maximum departures = 80 dept/hr × (120 min/60min) × 1.14 = maximum of 182.4 scheduled departures in a 2 hour period. 3 hour maximum departures = 80 dept/hr × (180 min/60min) × 1.06 = maximum of 254.4 scheduled departures in a 3 hour period.

86 preparing peak period and Operational profiles—Guidebook The airlines’ ability to pursue the above options is limited by the following factors: • Existing and planned aircraft fleet; • Network requirements, especially the type of aircraft needed to serve all their markets—not just the airport under study; • A competitive environment, which would affect both the ability to reduce frequency and to raise fares without losing passengers and revenue; and • Market characteristics such as length of haul, which would affect passengers’ ability to use alternative transportation modes. Based on empirical analysis and surveys conducted at several airports,xi the airlines would take the following actions, ranked in order of likelihood: • Increase fares to take advantage of reduced competition and to cover increased operating costs, thereby reducing the number of passengers from unconstrained levels; • Reschedule some flights to less busy connecting banks or off-peak hours subject to market requirements; • Increase the average size of the aircraft serving the market, provided the required aircraft are in their fleet. • Cease or reduce service to certain markets, starting with the lowest revenue markets (typically short-haul and leisure markets). Although it is counterintuitive, the evidence from these studies suggests that there will be little change in load factors. Minimal analysis has been done on the impact of airfield constraints on operations besides those of scheduled passengers, but the following observations are pertinent: • Passenger charter and all-cargo operations generally occur off-peak and would be unaffected. They also tend to have greater flexibility in adjusting schedules or using available supplemental airports if they have adequate runway length. • General aviation operations by smaller aircraft (piston and turboprop) often avoid busy com- mercial airports even when there are no significant capacity constraints. They can typically be persuaded to use reliever airports. • There is often a core of more sophisticated business jet operations that is difficult to dislodge. They are attracted to the all-weather instrument capability, access to air service, and often access to downtowns. Although they are often reluctant to change airports, they do have some flexibility to change flight times. Because of their complexity, the above factors are best addressed by incorporating them into a constrained design day schedule. The design day schedule can then be used to prepare constrained daily profiles and peak period estimates as needed. 9.2 Terminal Constraints The effects of terminal constraints will differ depending on whether the chokepoints are at the gates or at other terminal facilities. The impact of gate constraints on aircraft operations and passengers depends on the characteristics of the airlines serving the airport. Gate utilization for hub carriers is limited by the number of connecting banks, and large hub operations such as Delta at Atlanta can sustain more connecting banks than small hub operations, such as Delta at Memphis. Note that average gate utilization is typically less than the number of connecting banks, for various reasons. Not all connecting banks are of equal size, and therefore some gates will be unoccupied during the smaller banks. Also, some aircrafts,

application of Constraints 87 especially in overseas international service, may sit at a gate for more than one connecting bank. And, as was mentioned in Chapter 6, there is a need for excess gate capacity (spare gates) to accommodate disrupted schedules. Even at spoke airports, airlines with hub-and-spoke operations will be constrained by the timing of the connecting banks at the origin or destination airport. Airlines without formal hub-and-spoke networks, like Southwest Airlines, have more schedule flexibility and can achieve higher gate utilization rates. ACRP WOD 14 (Appendix O) provides some examples of the relationship between peak period operations and the number of gates at selected airports where gates are used intensively. These relationships may be used as a general guide to the maximum design day departures possible when the number of gates is constrained. Airline reactions to terminal constraints will be similar to their reactions to airfield constraints with one exception. Except at slot-controlled airports, such as Washington’s Reagan National Airport, airfield capacity is common-use, whereas gates are typically exclusive-use or preferential-use. Since individual airlines have more control over gate capacity than airfield capacity, they have more flexibility to “right-size” utilization to meet operational concerns without the risk of competing airlines backfilling the freed up capacity. Airlines therefore have more ability to exploit gate constraints to raise fares than they have with airfield constraints. Non-gate terminal constraints modify passenger activity differently than gate constraints. Non-gate terminal constraints, such as security or people-mover systems, directly affect passengers and then indirectly affect the airlines. Gate constraints directly affect the airlines and then indi- rectly affect passengers. Since passengers will modify their behavior to accommodate potential chokepoints by increasing lead or lag times, airlines generally do not feel compelled to modify their schedules. ACRP Report 25 contains several mini-queuing models that can be used to estimate the impact of constraints in ticketing, passenger screening, and Customs and Border Protection on passenger flow rates, in effect providing peak period constrained forecasts. 9.3 Landside Constraints Landside constraints are similar to non-gate terminal constraints in that they directly affect the passenger and only indirectly affect the airlines. If the throughput capacity is less than peak demand, knowledgeable passengers will extend their lead times. If possible and reasonably convenient, some passengers will change their transportation mode to a mode less affected by the constraint. Note that automobile parking demand and any associated constraints will have a different profile than other landside facilities. Parking is a function of cumulative demand rather than peak demand. Therefore, at many airports with a strong business passenger component, parking demand (number of spaces occupied) tends to peak during the middle of the week when the majority of resident business travelers are away, even though passenger enplanement and deplanement activity tends to be low at that time. 9.4 Policy Constraints Policy constraints can be grouped into two general categories, direct and indirect. Examples of the former would be slot controls or noise curfews, wherein activity levels are directly spelled out. Indirect constraints would include restrictions that limit, but do not eliminate, the ability of airlines

88 preparing peak period and Operational profiles—Guidebook to operate during sensitive timeframes such as times of congestion or nighttime. Examples of indirect constraints include peak period pricing or noise budgets. 9.4.1 Direct Policy Constraints At first glance, peak period or operational profiles would appear to be easy to estimate by simply defining the peak period as the regulated limit. However, even with rigorous hourly slot systems, airlines have latitude to schedule at different times within an hour, so peak periods may vary within an hour. As noted earlier, the peak 60-minute period often differs from the peak clock hour, and slot restrictions typically apply to the clock hour. In these instances, it is advisable to use the approach discussed in Chapter 8 for peak period operations. The effects of the slot system will be reflected in the distribution of daily operations; the remaining steps will be the same as in unconstrained airports. Note that in some instances, the restrictions do not apply to off-peak hours. Strict noise curfews restrict the operations occurring during sensitive nighttime hours which may or may not correspond to the INM definition of night (10 PM to 7 AM). Typically, noise curfews do not apply until 11:00 PM or later, so there is still potential for nighttime operations. The approaches in Chapter 7 can be used to estimate day/night splits even with a noise curfew, but the impact of the noise curfew will need to be evaluated when developing assumptions on the expected increase or decrease in the percentage of nighttime operations. Airports with slot restric- tions constrain activity during peak hours which typically occur during the day and therefore are more likely to shift activity to the nighttime than at an unconstrained airport. This should be taken into account when preparing assumptions on the increase or decrease of nighttime operations. 9.4.2 Indirect Constraints The impact of congestion pricing initiatives on peak hour operations and operational profiles can theoretically be modeled by estimating airline elasticities of aircraft operations to operational costs. In this instance, the elasticity represents the percentage decrease in aircraft operations with each one percent increase in operational costs including landing fees. This may not work in reality, since cost concerns associated with a single aircraft operation may be superseded by network concerns and competitive factors. Assuming the anticipated congestion pricing initiative has adequate flexibility, identifying or estimating the congestion reduction goals and assuming that the pricing mechanism will be adjusted to meet those goals may be a more effective way of estimating peak period activity levels. Noise budgets typically involve a fixed noise energy allowance which airlines cannot exceed, but have the flexibility of accommodating with either a few relatively noisy aircraft or a greater number of quieter aircraft. Forecasting operations in the times affected by the noise budget will require knowledge of future airline fleet plans and the noise characteristics of individual aircraft types. It is recommended that the planner consider surveying or interviewing the airlines to obtain their input. 9.5 Comments and Cautions When preparing constrained operational profiles or peak period estimates, the following should be considered: • At present, slot controls are the only demand management tools in use in the United States, and only at a very limited number of airports because they must be authorized by the federal

application of Constraints 89 government. As a result of recent regulatory revisions, airports will soon have a greater menu of options to choose from when managing demand. The impacts on aircraft operation and passenger profiles will vary depending on the type of tool used. As of yet, there is no empirical base that can be used to estimate and calibrate airline and passenger reactions to new demand management tools. • The FAA is becoming increasingly sophisticated in the real-time management of demand. For example, gate holds (aircraft instructed to remain at the gate by FAA air traffic control) are now employed when the FAA anticipates more airspace congestion than can be accommodated. • In adverse weather, airlines will often cancel flights rather than accept extensive delay. This means that in some instances, operational profiles during adverse weather (IFR conditions) will contain fewer operations than operational profiles in good weather (VFR conditions) during the same time of year. Tower data can be used to estimate which types of flights tend to be cancelled under these conditions. Addressing these issues is complex and highly dependent on the particular airport and con- straint type. It may be beneficial to have professional consulting services analyze those types of changes.