Using Neural & Fuzzy-Neural Approaches in School Trip Distribution Modeling

Trip distribution is considered as the second step in urban transportation planning.Many models have been presented for this purpose. Trip distribution traditionallymodels with the deterministic variables although it seems affective variables in trip distribution molding are based on human perceptions. Since perceptions of peoplevary from one person to another, thus variables are imprecise and vague, so modeling the distribution of trips between zones is complex and dependent on the quality andavailability of field data. Neural networks and neuro-fuzzy systems are suitable toolsto modeling non-deterministic variables. This paper develops and presents a new neural network approach to model trip distribution. Neural networks are organized indifferent architectures and the results have been compared in order to determine the best fitting one. Different models were trained, validated and tested with a realdatabase obtained from Tehran and then compared with Frater model made for school trip distribution in Tehran. The results of case study show that fuzzy modelcan be improved in order to accurately predict trip distribution regard to Frater model

Keywords:

Trip distribution , neural trip distribution model , Frater Model , neuro-fuzzy systems