Page 186 - Planning and Design of Airports
P. 186

For ecasting for Airport Planning    153


                    Time series analysis essentially involves extrapolating or project-
                 ing existing historical activity data into the future. Market share fore-
                 casting is a simple top-down approach, where current activity at an
                 airport is calculated as a share of some other more aggregate measure
                 for which a forecast has been made (typically a regional, state, or
                 national forecast of aviation activity). Econometric modeling is a
                 multistep process in which a casual relationship is established
                 between a dependent variable (the item to be forecast) and a set of
                 independent variables that influence the demand for air travel. Once
                 the relationship is established, forecasts of independent variables are
                 input to determine a forecast of the dependent variable. These tech-
                 niques can also be referred to as a bottom-up forecast. Simulation
                 models are often used when one needs very detailed estimates of air-
                 craft, passengers, or vehicles. These models impose precise rules that
                 govern how passengers, aircraft, or vehicles are routed, and then
                 aggregates the results so that planners can assess the needs of the
                 network or a component of the airport to handle the estimated
                 demand. Typically the outputs from the other forecasting methods
                 are used as inputs to simulation models. Forecasts from simulation
                 models represent snapshots of how a given amount of traffic flows
                 across a network or through an airport, rather than a monthly or
                 annual estimate of total traffic.
                    An important element which should be utilized in any forecast-
                 ing technique is the use of professional judgment. A forecast prepared
                 through the use of mathematical relationships must ultimately with-
                 stand the test of rationality. Frequently a group of professionals
                 knowledgeable about aviation and the factors influencing aviation
                 trends are assembled to examine forecasts from several different
                 sources, and composite forecasts are prepared in accordance with the
                 information in these sources and the collective judgment of the group.
                 In some cases, judgment becomes the principal approach used with
                 or without an evaluation of economic and other factors that are
                 believed to affect aviation activity. A common approach being uti-
                 lized more often today for preparing forecasts by judgment is known
                 as the Delphi method. In this method a panel of experts on a particu-
                 lar subject matter is asked to rate or otherwise prioritize a series of
                 questions or projections through a survey technique. The results of
                 the survey are then distributed to the members of the panel and an
                 opportunity is provided for each member to reevaluate the original
                 rating based upon the collective ratings of the group. The reevalua-
                 tion process is often sent through several iterations in order to arrive
                 at a better result. In the Delphi method, the results of the technique do
                 not have to represent a consensus of the panel and, in fact, it is often
                 quite useful to have a forecast which indicates the spread of the panel
                 in reaching conclusions on a particular issue.
                    The preparation of judgmental forecasts which reflect the collec-
                 tive wisdom of a broad range of professionals has proven to be very
   181   182   183   184   185   186   187   188   189   190   191