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.
This synthesis study provides a review of airport aviation activity forecasting in the United States. Forecasts of airport aviation activity have become an integral part of transportation planning. Most airport-specific forecasts are prepared on behalf of airport sponsors and state or regional agencies. The type and method of forecasting can depend importantly on the purpose for which the forecast is being made. For example, there may be sharp variations between forecasts used to support an annual budget versus a long-term facilities expansion. In practice, an important factor affecting many forecasts is that they are developed in sup- port of the master planning process that is used by FAA to identify capital projects that may qualify for funds from the agencyâs Airport Improvement Program. This is the primary federal funding mechanism for public-use airport improvements. The primary statistical methods used in airport aviation activity forecasting include market share analysis, econometric modeling, and time series modeling. These methods can be used to create forecasts of future airport activity over time. Simulation models are a separate method of analysis used to provide snapshot estimates of traffic flows across a network or through an airport. There are several activity measures typically included in airport aviation activity forecasts; the two most commonly used for commercial airports are aircraft operations and passenger enplanements. Based aircraft counts are important at general aviation airports because they drive the need for hangars, fueling, and other facilities. One of the most important requirements in preparing forecasts is to obtain accurate historical data on aviation activity. There are many useful data sets available from the federal government and other sources. For forecasts involving econometric modeling, it is also necessary to obtain historical data and future estimates of the explanatory variables to be included in the model. The market share approach to forecasting is a top-down method where activity at a par- ticular airport is assumed to be tied to growth in some aggregate external measure (typically a regional, state, or national aviation growth rate). For this method to produce reasonable pre- dictions, it is important that the presumed relation between airport activity and the larger aggregate measure be relatively constant over time. Many airport forecasts use econometric methods that utilize explanatory variablesâfactors thought to explain changes in the demand and/or supply of aviation activities. These factors can be broadly grouped into macroeconomic and demographic factors, airline market fac- tors, air transport production costs and technology, regulatory factors, infrastructure con- straints or improvements, and potential substitutes for air travel. Although econometric modeling is potentially a very sound and powerful method, there are many ways in which the specific model can go wrong, and it is not always obvious how best to proceed when sta- tistical tests or data issues indicate a problem. Time series modeling is another forecasting approach that involves some form of extra- polating existing data into the future. In its simplest form, it is based only on values of the variable being forecast and projects the future based on current or past trends. Because one SUMMARY AIRPORT AVIATION ACTIVITY FORECASTING
does not need to collect data on other variables, it can be a low-cost method compared with econometric modeling. Although the approach is conceptually simple, specific statistical tech- niques that are employed to make it more accurate can be quite sophisticated. This method can be useful when there are unusual conditions that make the relationship between local activity and other external factors unstable. Simulation models are a separate method of analysis used to obtain high-fidelity snapshot forecasts of traffic flows in a network or at an airport. Such models impose precise rules that govern how passengers or aircraft are routed, and then aggregate the results so that planners can assess the infrastructure needs of the network or airport to be able to handle the estimated traffic. Airport forecasting studies often neglect the issue of uncertainty. Most often, forecasts are presented only as point estimates, although it is common to also present alternative âhighâ or âlowâ estimates that are based on differing assumptions about external factors thought to affect the forecast. Although this can provide a reasonable range of estimates, there are additional sources of uncertainty related to the statistical properties of the models employed that are often neglected entirely. Accuracy is another often-neglected aspect of forecast evaluation, largely because it can only be done after the factâwhen values can actually be measured and compared with their forecast estimates. This problem is particularly relevant for long-term aviation forecasts where accuracy cannot be fully assessed for many years. Once the data are accessible, there are a variety of metrics available to measure forecast accuracy. There is a potential for optimism bias in airport forecasting that is countered by the issuance of FAA guidance documents, requirements for master planning, and other rules that local spon- sors must follow when applying for grants. In cases where more than one forecast is available for consideration, a number of alterna- tive approaches can be pursued. These include critical analysis of each individual forecast to help identify possible errors or mistakes, consideration of each forecastâs predictions by experts in the field who may possess significant domain knowledge regarding current and future air- port activities, and combining multiple forecasts to yield consensus averages. Several avenues for future research are suggested by this study including investigation into the reliability of data collection (particularly at smaller airports), detailed study of common sta- tistical and data problems associated with econometric forecasting models, the potential use of time series models and how their predictions compare with other methods, and formal studies of how well typical aviation forecasts project future activity. 2
