A thesis submitted in partial fulfillment of the requirement for the degree of Master of Engineering, School of Civil Engineering

Thesis (M.Eng.) - Asian Institute of Technology, 1996

Two models for travel demand forecasting are formulated based on consumer behavior theory and considering value of time saving. Techniques for estimation of model parameters are also illustrated using social trip data. The first model predicts the demand share of each travel mode from each origin to each destination simultaneously. Second model is total demand model which predicts the total number of trips generated in a given origin. Combination of these two models gives the actual number of trips by each mode from each origin to each destination. All the demand functions include the policy variables such as travel time and travel cost which facilitate to assess the effect of policy change on travel demand. Prediction accuracy of models is checked comparing the predicted and observed demands. Results show that predictions are fairly accurate with high correlation coefficient between the predicted and observed demands. Sensitivity analysis is carried out. Results show that demands are highly sensitive to the variables considered in the model. Predictions and sensitivity's of the said models were compared with those of conventional models. Results prevailed that proposed models are better that the conventional one in terms of accuracy and sensitiveness