NEURAL-NETWORK ANALYSIS OF TRAVEL BEHAVIOR - EVALUATING TOOLS FOR PREDICTION

This article explores the application of neural networks to a behavior al transportation planning problem. The motivation for adding neural n etworks as a new modeling methodology stems from its apparent relevanc e to problems requiring large scale, highly dimensional, data analysis , such as travel related behavior. Neural networks provide a tool to a nalyze the data in which we can model our intuition, and they provide that capability without the complication of having to formalize all th e complex causal variables and relationships which other models requir e. The transportation issue explored, upon which the neural network me thodology is tested, is a comparison of travel demand patterns of men and women in Israel. The information base is the Traveling Habits Surv ey (Central Bureau of Statistics, Israel, 1984, Statistical Abstract o f Israel, No. 35) commissioned by the Israel Ministry of Transport; co mbined with demographic and socioeconomic data of the 1983 Population and Housing Census. As extensive as such surveys are, the neural netwo rks imply that additional categories of data are necessary to predict how these elements relate to travel behavior. This article concentrate s on the extent to which neural networks can combine the relative simp licity of aggregate transportation models, with the theoretical advant ages and level of detail of disaggregate transportation models, withou t the latter's complexity. We describe the various directions we took in analyzing complex travel related data with feed forward, backpropag ation trained, neural networks. Copyright (C) 1996 Elsevier Science Lt d