Defining and Measuring Aircraft Delay and Airport Capacity Thresholds (2014)

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.

47 This chapter provides recommendations for delay and capacity analyses methods and metrics to use at particular phases during airport development. A very direct, logical, and quantifiable relationship exists between capacity, demand, and delays, depicted in Figure 4-1. Capacity is the resource available; throughput measures how that resource is utilized (considering operational rules, procedures, etc.); delay is an output variable of demand and capacity, and there will be delays if/when demand is greater than capacity. Delay evalu- ates effects of a specific flight demand as it operates on the resource. Although flight schedule patterns affect delays, the schedules should not alter capacity. Specifically, if one knows two of the three factors—demand, capacity, delay—the other can be calculated. Guidance and recommendations are provided regarding the relevance of particular delay and capacity measures by air- port type, airport characteristics, and project lifecycle phase. The approach taken in this report is to suggest what tools are most appropriate at various points in the project development cycle, for specific items in each element, and for different types of airports. After presenting an overview of the relevant char- acteristics, the recommendations are presented as • Measuring delays based on project lifecycle (Table 4-1); • Operational delays that result from applying particular delay thresholds, based on airport type (Table 4-2); and • Measuring capacity during project lifecycle (Table 4-3). However, it is not practical to have one threshold that can be applied to all airports, and this report recommends against using acceptable delay as a term or measure. Rather than establishing a standard for what level of delay should trigger development, the metrics will provide a standard language that should be understandable to all the stakeholders. Tables 4-1 through 4-3 contain the recommended approaches for measuring delay and capacity, as well as threshold metrics. 4.1 Airport Characteristics Affecting Capacity/Delay Analyses The methods and metrics to analyze capacity and/or delay may differ depending on the type of planning scenario and certain airport characteristics. This section presents catego- ries that will be used in this chapter to recommend analy- sis methods and metrics. Many of these recommendations are based on information reviewed from various airports, as described in Section 3.4. 4.1.1 Based on Point in Project Lifecycle Airport development projects typically progress through a project lifecycle (Figure 4-2) involving several stages of airport and environmental planning, followed by design and con- struction activities, and resulting in project commissioning. The planning component of this lifecycle is typically where the majority of capacity and delay analysis is performed. The planning process often begins with initial or strategic planning, moves into master or comprehensive planning, and then into environmental planning and documentation. For larger, more sophisticated airports, planning may further progress through more detailed planning and programming (often referred to as advanced planning or preliminary design), which typically yield detailed project-specific documents with programming, scheduling, phasing, and costing. As the planning efforts move from initial planning through detailed planning and design, or to EIS preparation, the need for better detail in capacity and delay analysis increases. High- level analyses may be conducted for initial planning, but more detailed analyses often are needed as planning progresses to gain support from airline users and local citizens. More thorough analysis is used to support benefit-cost analysis, pro- vide quantitative data for items like air quality impact evalu- ations, and also can be used to better determine when project design and construction activities should begin. C H A P T E R 4 Recommendations

48 A second important variable in determining the need for more detail and sophistication in capacity and delay analy- sis revolves around the type of airport being studied. Major hub airports need a more detailed capacity and delay analysis, typically involving computer simulation, to help define proj- ect benefits and delay savings and to support good financial resources to fund these activities. On the other end of the scale, small general aviation airports are addressing very different Figure 4-1. Relationship of capacity, throughput, and delay. Measuring Delay During the Project Lifecycle Point Airport Type Large Commercial Service/Cargo Service Airports Small Commercial Service Airports General Aviation & Reliever Airports Initial Planning/Strategic Planning Can use FAA and/or BTS databases (e.g., OPSNET) to compare historical delays. Spreadsheet and queuing models. Report average delays or overall total delays. If airport included, can use FAA and/or BTS databases (e.g., OPSNET) to compare historical delays. Spreadsheet and queuing models. Report average delays or overall total delays. Spreadsheet and queuing models, AC 150/5060-5, Chapter 3. Master (or Comprehensive) Planning Spreadsheet and queuing models if capacity need is beyond 10 years. Simulation modeling with SIMMOD, TAAM, or other model if project need is within 10 years. At a minimum, report average delays, but more detailed delays useful for gaining airline support. Spreadsheet and queuing models if capacity need is beyond 10 years. Simulation modeling with SIMMOD, TAAM, or other model if project need is within 10 years. Report average delays. Spreadsheet and queuing models, AC 150/5060-5, Chapter 3. Report average delays. Advanced Planning/Preliminary Design Simulation modeling with SIMMOD, TAAM, or other model. Review hourly delays and travel times in all wind/weather configurations when comparing alternative layouts. Simulation modeling with SIMMOD, TAAM, or other model. Average delays from major wind/weather configurations. Spreadsheet and queuing models, AC 150/5060-5, Chapter 3. May use simulation for comparing design alternatives. Average delays from major wind/weather configurations. Benefits-Cost Analysis Simulation modeling with SIMMOD, TAAM, or other model. Average annualized taxi times and delays may be sufficient, depending on the project. Overall total delays also can be used. Spreadsheet or simulation models. Average annualized taxi times and delays may be sufficient, depending on the project. Overall total delays also can be used. Spreadsheet and queuing models, AC 150/5060-5, Chapter 3. Average annualized taxi times and delays should be sufficient. Environmental Planning (EA/EIS) Simulation modeling from master plan or new/updated simulation modeling. Spreadsheet or simulation models. Average annualized delays Spreadsheet and queuing models, AC 150/5060-5, Chapter 3. Average annualized delays may be sufficient for analyses. Peak hour delays can be useful to gain public support. Simulation output also can provide data for air quality analyses. should be sufficient. Average annualized delays should be sufficient. Table 4-1. Recommended methods to measure delays based on project lifecycle point.

49 Construction Phasing Simulation modeling with SIMMOD, TAAM, or other model. Report hourly delays and taxi times in all wind/weather configurations. Compare phasing delays to current delays to gain airline agreement. Simulation modeling with SIMMOD, TAAM, or other model. Report average delays in major wind/ weather configurations. Spreadsheet or simulation models. Average annualized delays should be sufficient. May focus on specific wind/weather configurations. Operational Performance Can use FAA and/or BTS databases (e.g., OPSNET) to compare historical delays. Simulation modeling with SIMMOD, TAAM, or other model to analyze future schedules, fleet mix changes, or new procedures. Delay analysis depends on the issue being evaluated, but will likely require hourly delays. If airport included, can use FAA and/or BTS databases (e.g., OPSNET) to compare historical delays. Simulation modeling with SIMMOD, TAAM, or other model to analyze future schedules, fleet mix changes, or new procedures. Delay analysis depends on the issue being evaluated; average delays may be sufficient. Spreadsheet or simulation models. Delay analysis depends on the issue being evaluated; average delays may be sufficient. Table 4-1. (Continued). Major Capacity/Weather Characteristics Airport Type Major Connecting Hub Major O&D Airport Medium/Small Hub Air Carrier Airport Examples ATL, IAD, CLT, DFW DCA, SEA COS, ALB, ORF Typical/high incidence of IMC IMC capacity similar to VMC capacity Average delay of 5 minutes ≈ max delays of 40 minutes in VMC or 90 minutes in IMC Average delay of 5 minutes ≈ max delays of 30 minutes in VMC or 60 minutes in IMC Average delay of 5 minutes ≈ max delays of 15 minutes in VMC or 60 minutes in IMC Examples ORD, PHL, EWR, MSP JFK, SFO, BOS CHS, PBI Typical/high incidence of IMC IMC capacity significantly less than VMC capacity Average delay of 5 minutes ≈ max delays of 45 minutes in VMC or 120 minutes in IMC Average delay of 5 minutes ≈ max delays of 30 minutes in VMC or 100 minutes in IMC Average delay of 5 minutes ≈ max delays of 15 minutes in VMC or 60 minutes in IMC Examples SAN MCO, FLL, TPA TUS, JAX Low incidence of IMC IMC capacity similar to VMC capacity Average delay of 5 minutes ≈ max delays of 30 minutes in VMC or 45 minutes in IMC Average delay of 5 minutes ≈ max delays of 20 minutes in VMC or 30 minutes in IMC Average delay of 5 minutes ≈ max delays of 15 minutes in VMC or 30 minutes in IMC Examples PHX, IAH LAX, LAS LGB, ABQ Low incidence of IMC IMC capacity significantly less than VMC capacity Average delay of 5 minutes ≈ max delays of 30 minutes in VMC or 120 minutes in IMC Average delay of 5 minutes ≈ max delays of 20 minutes in VMC or 60 minutes in IMC Average delay of 5 minutes ≈ max delays of 15 minutes in VMC or 45 minutes in IMC Examples RSW AGS, ASE • • • • • • • • • Airport with concentrated seasonal traffic Average delays should be calculated for both peak and non-peak times. Delays in peak times are more relevant than annualized average delays. Average delays should be calculated for both peak and non-peak times. Delays in peak times are more relevant than annualized average delays. Average delays should be calculated for both peak and non-peak times. Delays in peak times are more relevant than annualized average delays. Table 4-2. Delay thresholds for various airport types.

50 Identifying and Measuring Capacity During the Project Lifecycle Airport Type Major Connecting Hub Major O&D Airport Medium/Small Hub Air Carrier Airport GA Airports Initial Planning/Strategic Planning Likely need hourly capacity estimates. Likely need hourly capacity estimates. Likely need hourly capacity estimates. FAA ASV calculation, new ACRP 03-17 model. Master (or Comprehensive) Planning Likely need hourly capacity estimates, perhaps for both peak arrival hours vs. peak departure hours. Likely need hourly capacity estimates, perhaps for both peak arrival hours vs. peak departure hours. Likely need hourly capacity estimates. FAA ASV calculation, new ACRP 03-17 model. Advanced Planning/Preliminary Design Likely need peak capacity in 10- or 15-minute increments, for peak arrivals and peak departures. Likely need peak capacity in 10- or 15-minute increments, for peak arrivals and peak departures. Likely need hourly capacity estimates, for peak arrival hours and peak departure hours. FAA ASV calculation, new ACRP 03-17 model. Benefits-Cost Analysis Likely need hourly capacity estimates, perhaps for both peak arrival hours vs. peak departure hours. Likely need hourly capacity estimates, perhaps for both peak arrival hours vs. peak departure hours. Likely need hourly capacity estimates. FAA ASV calculation, new ACRP 03-17 model. Environmental Planning (EA/EIS) Likely need hourly capacity estimates, perhaps for both peak arrival hours vs. peak departure hours. Likely need hourly capacity estimates. Likely need hourly capacity estimates. FAA ASV calculation, new ACRP 03-17 model. Construction Phasing Likely need peak capacity in 10- or 15-minute increments, for peak arrivals and peak departures. Likely need peak capacity in 10- or 15-minute increments, for peak arrivals and peak departures. Likely need hourly capacity estimates, for peak arrival hours and peak departure hours. FAA ASV calculation, new ACRP 03-17 model. Operational Performance Likely need peak capacity in 10- or 15-minute increments, for peak arrivals and peak departures. Likely need peak capacity in 10- or 15-minute increments, for peak arrivals and peak departures. Likely need hourly capacity estimates, for peak arrival hours and peak departure hours. Hourly capacity. Table 4-3. Measuring capacity during project lifecycle. Figure 4-2. Typical project lifecycle.

51 – Primary airports are commercial service airports that have more than 10,000 passenger boardings each year. Hub categories for primary airports are defined as a per- centage of total passenger boardings within the United States in the most current calendar year ending before the start of the current fiscal year. • Cargo service airports are airports that, in addition to any other air transportation services that may be available, are served by aircraft providing air transportation of only cargo with a total annual landed weight of more than 100 million pounds. An airport may be both a commercial service and a cargo service airport. • Reliever airports are airports designated by the FAA to relieve congestion at commercial service airports and to provide improved general aviation access to the overall community. These may be publicly or privately owned. • The remaining airports are commonly described as gen- eral aviation airports. This airport type is the largest single group of airports in the U.S. system. The category also includes privately owned, public-use airports that enplane 2,500 or more passengers annually and receive scheduled airline service. Airport Classifications This study has taken the broad FAA airport categories and created subcategories within the commercial service type for purposes of identifying the best ways to measure capacity and delay. The following discreet categories will be used to iden- tify and discuss delay thresholds: • Commercial service/cargo service airports – Major connecting hub—one or more airlines has a sig- nificant presence at the airport with a business model for transferring passengers from each inbound flight to multiple outbound flights. Example airports include ATL, DEN, DFW, and PHX. – Major O&D airport—passengers are typically arriving or departing this large airport rather than connecting from one flight to another. While there is some connect- ing traffic, the vast majority of the passengers are O&D. Example airports include BOS, DCA, and MCO. – Medium/small/non-hub (“spoke” airport)—nearly all passengers are originating or terminating at this airport. Examples include COS, JAX, PBI. • Reliever and general aviation airports Several major airports serve both as a connecting hub and as O&D. Many O&D airports serve long-haul markets where the timing of flights is very sensitive, and where overseas arrival slot times may get missed if flights get significantly delayed. One must consider the overall traffic mix and the issues that seldom revolve around runway capacity constraints and typically can use more simple capacity and delay tools. The following section discusses the relevant characteristics of the various types of airports for which capacity tools might apply. 4.1.2 Relevant Airport Characteristics The level of average annual delay to use as a threshold for determining capacity at a given airport varies greatly. The following three principal factors that categorize airports into specific groups for purposes of providing general guidance about how to measure delays and determine reasonable levels of delay have been identified in this effort: • Airport type, • Good weather/bad weather capacity ratio, and • Good weather/bad weather occurrence. Airport Type Put simply, different types of airports (major connecting hubs vs. small general aviation airports) can tolerate different levels of delay. Major connecting hubs that operate continu- ously throughout the day like Hartsfield-Jackson International Airport in Atlanta have a great need for schedule reliability. In contrast, local origin and destination (O&D) airports in major metropolitan areas that are difficult to expand where customers have no other or limited airport choices (San Diego or La Guardia in New York, for example) have a different delay threshold. This section identifies how delay thresholds might be approached for a range of airport types. For consistency with FAA planning guidance, the FAA categorization of airports is first presented below. Then, a number of subcategories for purposes of measuring delay is proposed, based on operational characteristics and how they relate to delay levels. FAA Definition of Airport Categories The FAA classifies airports into several broad categories, including commercial service, primary, cargo service, reliever, and general aviation airports. These airport categories are briefly summarized as follows: • Commercial service airports are publicly owned airports that have at least 2,500 passenger boardings each calendar year and receive scheduled passenger service. Passenger boardings refer to revenue passenger boardings on an air- craft in service in air commerce, whether or not in scheduled service. Passenger boardings at airports that receive sched- uled passenger service are also referred to as enplanements. – Nonprimary commercial service airports are commer- cial service airports that have at least 2,500 and no more than 10,000 passenger boardings each year.

52 affecting schedule reliability because of the rarity of bad weather. On the other hand, airports with high incidence of poor weather (e.g., Seattle) cannot tolerate very high delays in poor weather and still maintain schedule reliability while keeping airline operating costs reasonable. In Figure 4-3, the darker bars depict the percent of time the airport typically experiences poor visibility—ceilings less than 1,000 feet and/or visibility less than 3 miles, while the lighter bars show the capacity reduction from optimal conditions and IFR conditions, as reported in the FAA’s 2004 Benchmark Report. Categorizing airports by their good weather/bad weather incidence will further help to identify what delay threshold to apply at airports in each group. 4.2 Estimation of Delay for Different Purposes After reviewing numerous airports’ planning study reports for various phases or points along the airport lifecycle (described in Section 3.4), recommendations are included for delay analyses. No single delay metric can be applied to all airports. Characteristics of the airport and traffic demand profile as well as the project lifecycle step all affect the delay analysis. Also, certain delay statistics will be more useful to different audiences. Recommendations are presented in the following subsections for delay analyses methods and metrics for effective planning. importance of such operational items as tight connections, frequency of ground holds to other airports, and criticality of on-time performance at downline stations in conducting delay analyses at these airports. Good Weather/Bad Weather Capacity Ratio Airports that have similar capacities in good and bad weather are very different than those airports with dramatic differences in capacities in good and bad weather. Where capacities are similar, for example, an average annual delay of 10 minutes per operation generally means just that: average delays are 10 min- utes, day in and day out, despite the weather, yielding good pre- dictability and schedule reliability. In contrast, an average delay of 10 minutes at airports where capacities in good weather can be double those in bad weather might mean that average delays are 4–8 minutes in good weather, but delays of 45–60 min- utes might accrue in bad weather, severely impacting service reliability. Categorizing airports by their good weather/bad weather capacity ratio will help identify what delay threshold to apply at airports in each group. Good Weather/Bad Weather Occurrence Airports with very low incidence of bad weather (e.g., Phoenix Sky Harbor International) can tolerate high delays in very limited poor weather conditions without adversely Source: American Aviation Institute (AAI) Figure 4-3. Airport capacity loss vs. occurrence of inclement weather conditions.

53 Delay thresholds truly need to be customized to the air- port’s operations. When the traffic demand pattern has sharp peaks, such as at connecting hub airports, the maxi- mum delays or variances can be quite high. In such cases, the average delays may only be 5 minutes, but the actual opera- tional delays during good weather can be as much as an hour and can reach 2 to 3 hours during IMC. If the delays reach a level that exceeds the length of schedule lulls between peaks, then the delays affect not just the current peak/bank but also subsequent peaks throughout the day. At airports where the traffic demand pattern does not have very high peaks, the delays may be more constant throughout the day. Similarly, at airports with high occurrences of IMC traffic, the maximum operational delays can be very high. These air- ports may choose to focus their planning and capacity stud- ies on IMC configurations if they contribute the bulk of the delays. However, if IMC does not occur very frequently, the planning may only focus on good weather conditions, know- ing that the operational effects may be irrecoverable during those few hours when severe weather occurs. In general, there are very low delays at general aviation (GA) airports because the demand is typically not constant throughout the day, week, or year. If spreadsheet analysis shows more than 2 minutes of average delay at a GA airport, then a significant percent of the traffic would be experiencing some delay either in the air or on the ground (e.g., waiting to depart) or not be able to operate at their desired time. Since delays are seldom experienced at most GA airports, that air- port category is excluded from this table. 4.3 Capacity Metric Recommendations Chapter 3 discussed various elements of airport capac- ity measurements. Analysts can apply different methods to measure capacity and report the metric in terms of annual or hourly throughput or some other measure. Recommen- dations for what capacity metrics to use for various air- port types and project lifecycle phases are summarized in Table 4-3. Note that ACRP Report 79 includes a detailed decision sup- port tool and checklists for choosing model sophistication based on airport characteristics and the issue being studied. Readers are referred to that report to further identify their appropriate type of capacity model. Similar to the delay analysis recommendations above, more fine-tuned analyses are recommended as the planning progresses and to gain acceptance by certain stakeholders. For GA airports or those where capacity is not a concern, simple annual estimates will suffice. 4.2.1 Based on Point in Project Lifecycle Recommendations for the method or tool to use for delay analysis at each point in the project lifecycle are presented in Table 4-1. Appropriate delay measures are also proposed. In general, less-detailed analysis can be used early in the plan- ning processes. Although the recommendations primarily are to use analytical tools that would measure unimpeded delays, initial planning also can be effectively conducted with analysis of some of the operational delay databases described in Section 2.2. However, as the planning progresses, analyses should be more detailed and provide more detailed delay statistics. Suggestions also are provided for using certain delay metrics to gain acceptance of capacity enhancement options by stakeholders and the local community. It is critical that an analysis specify whether the delay was analyzed from operational delay databases (e.g., FAA OPSNET, BTS) or from simulation/spreadsheet tools. For example, OPSNET only includes delay values for flights delayed 15 min- utes or more such that average delays are quite high, whereas a simulation counts every second of delay and average delays may be a few seconds to a few minutes. Note that if initial measurements are reported from delay databases, there could be confusion if subsequent simulation analyses report much lower delay values. 4.2.2 Based on Airport Characteristics Delay logically would be an output of an analysis of the capacity and demand, not an input that defines capacity— but it can be set to help determine when capacity would be needed to accommodate a certain level of demand. Delay thresholds are used to • Define capacity, • Establish operational goals, and • Justify improvements/development. The average annualized delay from a spreadsheet or simu- lation model provides no indication of the extremely high delays that may be experienced by individual flights at that airport under different weather conditions. Since average annual delays are still the most common method to com- pare alternatives, especially when multiple runway configu- rations are used throughout the year due to wind/weather conditions, Table 4-2 provides estimates of the operational flight delays that may be experienced when certain simulated/ calculated delay values are simulated. Maximums have been estimated from numerous analyses such as those shown pre- viously in Figure 2-4, based on typical traffic patterns at that airport type.