This book provides an updated, concise summary of forecasting air travel demand methodology. It looks at air travel demand forecasting research and attempts to outline the whole intellectual landscape of demand forecasting. It helps readers to understand the basic idea of TEI@I methodology used in forecasting air travel demand and how it is used in developing air travel demand forecasting methods. The book also discusses what to do when facing different forecasting problems making it a useful reference for business practitioners in the industry.

This book provides an updated, concise summary of forecasting air travel demand methodology. It looks at air travel demand forecasting research and attempts to outline the whole intellectual landscape of demand forecasting. It helps readers to understand the basic idea of TEI@I methodology used in forecasting air travel demand and how it is used in developing air travel demand forecasting methods. The book also discusses what to do when facing different forecasting problems making it a useful reference for business practitioners in the industry.

Introduction: One of the key problems in the analysis and planning of any transport properties and facilities is estimating the future volume of traffic that may be expected to use these properties and facilities. Estimates of this kind are now being made regularly as the transport system continues to expand. The future planning, implementation and operation of a successful transportation system requires accurate and realistic forecasts of traffic volumes. To achieve optimal policies, the planner needs to be able to predict the effect of alternate decisions. Although the planning process involves much more than a forecast of the future traffic statistics, these statistics provide the essential quantitative dimensions for the planning process. Forecasts of expected traffic are an essential prerequisite to long-range planning. The link between planning and forecasting lies in recognizing that in order to bring an expected situation under control, the planner must be provided with the entire spectrum of situations that could be anticipated and, hence, could be planned for. The reasonableness and reliability of these traffic statistics is, therefore, of vital importance to the planner. This study investigates the North Atlantic passenger travel demand. The final goal is to make a forecast of the passenger traffic on this route. It is believed that such a forecast would prove to be a critical tool for long-range planning of transport properties and facilities on both sides of the Atlantic. For this reason, it is important to be well informed about the technical and economic factors which will determine and limit the travel volume, especially for manufacturers of aircraft, domestic and international airlines, and the government. Governments, for example, must be provided with traffic forecasts if they are to provide adequate ground facilities and air traffic control systems.

TRB’s National Cooperative Highway Research Program (NCHRP) Report 716: Travel Demand Forecasting: Parameters and Techniques provides guidelines on travel demand forecasting procedures and their application for helping to solve common transportation problems.

We analyze whether it is better to forecast air travel demand using aggregate data at (say) a national level or whether one should aggregate forecasts derived for individual airports using airport-specific data. We compare the U.S. Federal Aviation Administration's (FAA) practice of predicting the total number of passengers using macro economic variables with an equivalently specified AIM (aggregating individual markets) approach. The AIM approach outperforms the aggregate forecasting approach in terms of its out-of-sample air travel demand predictions for different forecast horizons. Variants of AIM, where we restrict the coefficient estimates of some explanatory variables to be the same across individual airports, generally dominate both the aggregate and the AIM approaches. The superior out-of-sample performance of these so-called quasi-AIM approaches depend on the trade-off between heterogeneity and estimation uncertainty. We argue that the quasi-AIM approaches efficiently exploit the heterogeneity across individual airports without suffering from as much estimation uncertainty as the AIM approach.