EFFICIENT METHODS FOR TRIP GENERATION AND DISTRIBUTION FORECASTS: A CASE STUDY

Recently, great attention has been given to analysis of travel in the peak period, especially peak hour trip generation and distribution. For trip generation analysis, either the cross-classification or the linear regression method is employed. While trip distribution is often accomplished through the use of growth-factor methods, intervening opportunity models, gravity models, and discrete choice models (i.e. various logit, probit, and dogit models). Much of the literature discusses the results based on the comparative studies of different methods for trip generation and distribution forecasts. These studies indicate that each method has its advantages and disadvantages. The selection of methods depends on data sources, contents of study, level of detail, and accuracy of forecasting, and even time and budget, and available software. This paper describes the efficient and practical methods to predicate peak hour travel patterns under the circumstance of limited data sources