This report is based on the first phase of a review of models and other techniques for estimating travel demand. A critique identifies the strengths and weaknesses of the `traditional' four-stage model, and assesses the available alternatives in terms of demand forecasting and policy analysis. Among the alternatives, particular attention is paid to three currently operational models - the Dutch National Model (Landelijk Model) based on disaggregate structures, MEPLAN which is a fully integrated land use/transport model designed for practical application to a wide range of transport as well as land use issues, and MASTER (Micro Analytic Simulation of Travel Employment and Residences) which is based on the explicit modelling of interactions at the individual and household level. The report also reviews a number of other techniques of data capture and analysis which have the potential to address some of the weaknesses identified in the traditional approach. Some of these techniques are outside the mainstream of modelling methodology but are considered to have made recognizable contributions to the understanding of travel behaviour. These include stated preference (SP) techniques which are useful when addressing choice situations which are not amenable to direct observation; activity based approaches which are usually associated with detailed analyses of daily patterns of travel behaviour; and `dynamic' methods which are characterised either by `path dependency' or incorporate lagged recursions over time.

Want to know more about this project?