Modeling of stratified urban trip distribution

Trip distribution analysis involves determination of the pattern of trip interchanges between land-use parcels or traffic zones using an appropriate analytical technique. The different analytical techniques available for trip distribution can be broadly grouped into two types: growth-factor methods and synthetic methods. The gravity model is one of the commonly used synthetic models for trip distribution. This paper deals with the estimation of travel deterrence for trips made for different purposes and by different modes through calibration of two types of gravity model formulations, using the data set pertaining to a medium-size city with a complex travel environment in India. Stratified data of home-based trip distribution for three trip purposes, and trips made by five different modes, were used for calibrating the model formulations. Models of type I formulation were calibrated by the cell-by-cell comparison procedure and those of type II formulation as per the Bureau of Public Roads (BPR), United States of America method. The parameter values obtained for the models of both formulations were found to be logical and statistically significant. Modeling of the stratified trip distribution, apart from yielding models of better predictive power, has provided insight into the relative effect of travel deterrence on trips made for different purposes and by different modes.