Airport Aviation Activity Forecasting (2007)

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.

3 BACKGROUND The purpose of this synthesis study is to provide an overview of current practices and methods in airport aviation activity forecasting in the United States. The study reviews academic and professional literature on forecasting as well as airport master plans, state airport system plans, and other sources of information showing how aviation decision makers, airport managers, sponsors, and their consultants actually produce forecasts of aviation activity. This study does not necessarily reflect the views or require- ments of the U.S. FAA with respect to airport aviation activ- ity forecasting. In practice, however, it is recognized that many airport forecasts follow guidance directives issued by FAA to qualify for funds from the agency’s Airport Improve- ment Program (AIP), which is the primary federal funding mechanism for public-use airport improvements. Further dis- cussion of forecasting produced under these circumstances is included in the section Forecasting Under FAA Guidance. Air transportation is an integral part of the global econ- omy. Scheduled commercial passenger transport alone has grown over the past half century to more than 27 million flights in 2005 across the world, and more than 11.6 mil- lion flights to or from the United States alone. Other sec- tors, including cargo, air taxi, and general aviation, also have expanded rapidly. Airports are obviously a fundamental part of modern air transportation systems—some serve as hubs for extensive national and international transportation networks, whereas others function as the complementary spokes; still others serve as parts of multi-airport regional systems; and some serve as important connections to the outside world for otherwise isolated populations. Airports serve both commer- cial and private aviation—often referred to as air carrier and general aviation. The construction, operation, and future expansion of air- ports can require substantial initial and ongoing investments, a large share of which is usually paid for with public monies. Consequently, for any individual airport, it is important to be able to forecast future demands for aviation services to assess the potential need for further investments in capacity or ser- vices to meet those demands. Accurate forecasts are essential for effective airport planning and decision making, and for the efficient provision of capacity. The type of forecast and level of effort required to produce it depends importantly on the purpose for which the forecast is being made. Short-term aviation forecasts (typically refer- ring to projections no more than five years into the future) are needed to support operational planning and often are used to assess personnel requirements at an airport or the need for incremental improvements or expansions of landside facilities and terminal areas, air cargo facilities, general aviation hangar space, etc. Since the terrorist attacks of September 11, 2001 (9-11), new security requirements may also require changes to landside facilities. Depending on the size of the airport, intermediate-term (6–10 years out) and/or long-term forecasts (11–20 years) are used to plan major capital investments, such as land acquisi- tion, new runways and taxiways, extensions of existing run- ways, and new terminal or tower facilities. Forecasts beyond 20 years are sometimes undertaken to assess the need for addi- tional airports or other regional aviation facilities. Although forecasts over the short term can also be used in planning for large capital projects, longer-term projections are typically required to adequately assess the costs and benefits of such investments. Another important part of the forecasting process is the assessment of uncertainty that is inherent in any forecast. Predictions are often presented as a single set of numbers, which give no indication as to their likely accuracy. By directly addressing and, if possible, quantitatively measuring the uncertainty associated with a given forecast, forecasters can give decision makers the ability to plan different strate- gies based on the range of uncertainty and to explore how the uncertainty may depend on particular assumptions built into the forecast. FORECASTING METHODS The majority of airport and regional and state aviation activ- ity studies use fairly simple methods to produce forecasts, and address forecast uncertainty only in informal and nonsystem- atic ways. Data availability, particularly reliable time series for historical aviation activity, often dictates what forecasting techniques can be employed. In general, the level of sophisti- cation depends on a variety of factors, including data avail- ability, intended use of the forecast, and the level and types of activity at the airport. CHAPTER ONE INTRODUCTION

With these provisos in mind, the forecasting methods con- sidered here can be grouped into four categories: • Market share forecasting—local activity calculated as a share of some larger aggregate forecast. • Econometric model forecasting—aviation activity tied to other economic measures. • Time series model forecasting—trend extrapolation of existing activity. • Simulation—a separate method used to provide high- fidelity “snapshot” estimates of how traffic flows across a network or through an airport. The studies prepared for smaller general aviation airports tend to rely on trend extrapolation or market share analyses. Many such forecasts do not attempt to assess uncertainty, although multiple scenarios may be considered by assuming different growth rates. These methods are also used by many regional and state planning agencies where aggregate fore- casts of aviation activity across large numbers of airports are required. Forecasts prepared for larger commercial airports are more likely to use formal statistical methods such as econometric model forecasting, reflecting better data availability. Uncer- tainty is still typically dealt with in an informal (nonstatistical) way by creating high and low forecasts based on alternative assumptions about how certain explanatory variables in the econometric model (or other external factors) may change in the future. FORECASTING UNDER FAA GUIDANCE The primary purpose of FAA’s AIP program is to provide grants for public-use airport improvements. For facilities defined as large and medium primary hub airports, an AIP grant covers 75% of eligible costs (or 80% for noise program implementation). For small primary, reliever, and general avi- ation airports the grant covers 95% of eligible costs. For those forecasts produced to support requests for AIP funding, the FAA’s Advisory Circular on Airport Master Plans (2005) contains specific guidance on the entire fore- casting process. Other FAA guidance documents, in particular Forecasting Aviation Activity by Airport (2001) and Revision to Guidance on Review and Approval of Aviation Forecasts (2004), provide further practical information that planners use in determining how to produce forecasts that meet FAA requirements. In addition, the Advisory Circular on the Airport System Planning Process (2004) provides guidance for state and regional planners; this document is discussed separately later in this chapter. This overview of FAA guidance related to aviation fore- casting does not constitute official FAA policy or guidelines. Airport sponsors should refer to the above-cited documents 4 and contact FAA directly when preparing forecasts as part of the master planning process. It is important to understand that preparing aviation fore- casts is only a small part of the overall master planning process. An airport master plan is a comprehensive study that describes the short-, medium-, and long-term development plans at the facility to meet future aviation demand. It incor- porates many specific elements that go well beyond the scope of the forecasting topics covered in this study including: • Preplanning • Establishment of a public involvement program • Environmental considerations • Analysis of existing conditions • Aviation forecasts • Assessment of facility requirements • Development and evaluation of alternatives • Airport layout plans • Facilities implementation plan • Financial feasibility analysis. From a technical standpoint, it should be noted that AIP grant assurances require only that an airport sponsor have an approved airport layout plan, not necessarily an airport master plan. This discussion focuses only on the development of avia- tion forecasts. When forecasts are produced as part of airport master plans to be submitted for FAA approval, specific fore- cast elements (identified in the FAA Advisory Circular on Air- port Master Plans, AC 150/5070-6B, 2005) must be included, as shown in Table 1. Because of FAA’s major role in funding airport projects that are identified in master plans and the role that these plans play in the distribution of AIP funds, these are the measures found in many airport forecasts. Although the aviation activity elements listed in Table 1 specifically refer to annual estimates, FAA also requires mas- ter plan forecasts to include “appropriately defined peak period activity levels for facilities planning.” It is usually the case that peak activity will be more relevant for facilities planning at an airport than overall annual totals. Depending on the situation, the appropriate measure of peak activity may refer to seasonal, monthly, daily, and/or time-of-day demands. The master plan Advisory Circular also specifies that fore- casts should be prepared for the short term (up to five years), medium term (six to ten years), and long term (beyond ten years). In practice, most forecasts cover a 20-year period from the base year. Sponsors are expected to present a “baseline” forecast that represents the most likely estimate of activity over the 20-year period. In addition, they may provide a range of “scenario” forecasts to assess the impact of higher and lower activity levels on development plans at the airport. There are other FAA guidelines specified in the Advisory Circular including:

5 • Review of previous airport forecasts, including the FAA’s own Terminal Area Forecast (TAF) for the air- port (the TAF is discussed further in chapter two). • Selection of appropriate forecast method—FAA identi- fies regression analysis (econometric modeling), trend analysis, market share analysis, and smoothing as the most common techniques to be considered. • Application of forecast methods and evaluation of results, including justification or explanation of decreasing or increasing trends in activity, and sensitivity analysis to measure the impacts of changes in the factors that influ- ence activity. More detailed step-by-step guidance on these topics is pro- vided in the FAA document Forecasting Aviation Activity by Airport (2001). In addition, the FAA document Revision to Guidance on Review and Approval of Aviation Forecasts (2004) provides specific guidance on short-term forecasting, including directives regarding the use of historic seasonality patterns to extrapolate departures and passenger enplanements over a two-year forecast period. The guidance and directives from FAA are broadly consis- tent with the forecasting methods, data concerns, and other issues addressed in the present study. There are, however, two major aspects of FAA guidance where forecasters must make specific efforts to satisfy the agency’s requirements. The first involves the activity measures shown in Table 1. Despite the vast changes in the commercial airline industry over the past quarter century, the list of data elements shown in the table has remained virtually unchanged during this time. Of particular concern for airport planners and forecasters are the co-mingling of air taxi and commuter and regional opera- tions, as well as the distinction between air carrier and com- muter activity. The FAA definition of “air taxi” refers to car- riers that operate aircraft with 60 or fewer seats or a cargo payload capacity of less than 18,000 lb, and carries passengers on an on-demand basis only (charter service) and/or carries cargo or mail on either a scheduled or charter basis. “Com- muter” operators as defined by the U.S.DOT are those with scheduled passenger service (five or more round trips per week on at least one route according to published flight schedules) while utilizing aircraft of 60 or fewer seats. Air taxi carriers are governed under Part 135 and commuter carriers are governed under Part 121 of the Federal Aviation Regulations. In terms of airport planning and use, the co-mingling between air taxi and commuter makes little operational sense. The more relevant distinction would be between scheduled and nonscheduled service, which has a primary influence on the type of facilities that must be provided (e.g., terminal ser- vices). For this reason, for airport planners it would be useful to distinguish between air taxi (nonscheduled) operations and commuter (scheduled) operations. Both commuter and air carrier operations involve sched- uled service; the primary distinction is only in terms of the size of the aircraft used. The distinction between air carrier and commuter (regional) operations is becoming even more blurred as commuter markets are served more and more by larger regional jet aircraft with between 60 and 90 seats. This has been accompanied by changes in carriers’ scope clauses with their pilot unions that specify the size of the aircraft that must be flown by union pilots. Recent relaxation of these clauses has allowed carriers’ regional affiliates to fly these Required Included Where Appropriate Operations (annual) Itinerant Air carrier Air taxi and commuter (regional) General aviation Military Local General aviation Military Domestic vs. international Annual instrument approaches IFR vs. VFR operations Air cargo aircraft operations Touch-and-go operations (training) Helicopter operations Average load factor Fuel use Passengers (annual) Enplanements Air carrier Commuter Enplanements Originating Connecting Passenger and cargo data Domestic vs. international General aviation passengers Helicopter Air taxi Other—Number of student pilots/hours flown Aircraft Based aircraft Aircraft mix Critical aircraft Average seats/aircraft Source: Advisory Circular on Airport Master Plans 2005, p. 37. Notes: Cargo data typically include freight and/or U.S. mail tonnage or ton-miles; ìcritical aircraft” refers to identification of equipment based on “substantial use” (defined as either 500 or more annual itinerant operations, or engaged in scheduled commercial service). IFR = instrument flight rules; VFR = visual flight rules. TABLE 1 AVIATION DEMAND ELEMENTS

larger aircraft. Today, the categorization of commuter opera- tions as referring to scheduled service with aircraft of 60 or fewer seats is rather arbitrary and does not correspond with the observed patterns of flying. Nevertheless, as with the com- muter/air taxi co-mingling issue, it is important that forecasts that are to be reviewed and subject to FAA approval meet the necessary requirements regarding user group distinctions. The second major FAA requirement affecting airport forecasting studies that are being conducted as part of mas- ter planning is that the baseline forecasts of operations, pas- senger enplanements (where relevant), and based aircraft must be compared with the FAA’s TAF. They will be con- sidered consistent with the TAF, and therefore approved as part of the master planning process, if the 5-year-ahead fore- cast differs by less than 10% from the TAF and the 10-year- ahead forecast differs by less than 15% from the TAF. (In some cases, these requirements do not have to be met; for example, if it is shown that the forecasts do not affect the tim- ing or scale of an airport project, or in the case of smaller air- ports if the forecasts do not affect the role of the airport.) If the baseline forecasts do not meet these requirements, the differences will need to be adequately explained and resolved with FAA. Even if airport forecasting studies that are con- ducted as part of a master plan are consistent with the TAF, FAA may not automatically approve the plan without addi- tional information and justification. For state and regional planners, the FAA’s Advisory Cir- cular on the Airport System Planning Process (2004) pro- vides specific guidance for forecasts made at the system (multifacility) level. These system plans typically focus less on individual airport forecasts, and much more on defining 6 an airport’s role within the system and prioritizing airport development. The Advisory Circular provides guidance on reviewing individual airport forecasts consideration of inter- actions between airports where appropriate, and advice on estimating activity at nontowered airports. System plan fore- casts that are used in support of AIP projects are subject to the five-year 10% rule discussed earlier for TAF. ROADMAP FOR ANALYSIS The discussions in the following chapters focus on the most common practices and techniques used by analysts to produce airport aviation activity forecasts. Chapter two addresses how airport forecasts are used for varying purposes and identifies common aviation metrics, aviation data sources, issues in data collection and preparation, and special data issues at non- towered airports. It also discusses various drivers of aviation activity. Chapter three provides an overview of available forecast- ing methods and then discusses each of the major methods in more detail. Selection of the appropriate method is also dis- cussed, along with a presentation of some representative examples of actual airport activity forecasting studies. Chapter four discusses the evaluation of forecasts, includ- ing assessments of forecast uncertainty, accuracy, issues of optimism bias, and options for resolving differences when multiple forecasts are available. Chapter five provides concluding remarks and sugges- tions for future research.