| Abstract | This study dealt with the mode choice attitudes of travelers toward motorized and nonmotorized public transport, minibus and paratransit, bus and car in Khon Kaen, Thailand. Also, this research dealt with the determination of a stochastic disaggregate demand model that can explain mode choice of individual commuters. Firstly, variables for probabilistic model were selected by discriminance function analysis for system characteristic variable models. Secondly, user characteristic variables were introduced together with the system characteristic variables in the model. The variables were selected based on discriminance function analysis. These estimated models were then used to predict the probability of individual mode choice as well as diversions of commuters from one mode to another. In this study, 750 questionnaires concerning work journey trips were distributed to current employees of different offices in Khon Kaen. For minibus, cycle rickshaw, and auto rickshaw travelers, 1000, 250, and 200 questionnaires were distributed, respectively. Data related to work journey trip was collected by random work places interviews and home interview survey at different residential areas. Moreover, on-board surveys were also conducted on minibuses and as well as in shopping complexes, schools, colleges, and bus terminals. Cycle rickshaw and auto rickshaw questionnaire surveys were conducted solely onboard. Results obtained from the developed models based on system as well as user characteristics indicated that the probabilities of selecting different modes were quite similar to that of the probabilities obtained from actual questionnaires. Predicted probabilities of private vehicle vs public transport were 69.85% vs 30.15%, minibus vs city bus were 73.30% vs 26.70%, car vs bus were 66.03% vs 33.97%, motorized public transport vs non-motorized public transport were 78.92% vs 21.08%, and motorized paratransit vs non-motorized paratransit were 56.70% vs 43.30% obtained from the developed models. Probabilities of private vehicle vs public transport were 70.40% vs 29.60%, minibus vs city bus were 73.82% vs 26.18%, car vs bus were 71.48% vs 28.52%, motorized public transport vs non-motorized public transport were 87.02% vs 12.98%, and motorized paratransit vs non-motorized paratransit were 57.05% vs 42.95% obtained from the actual questionnaires. To evaluate the impact on mode choice of travelers of Khon Kaen, several alternatives were proposed and tested with the predicted models. In case of free public transport was introduced, 6.19% of travelers would divert from private veh\cle to public transport. If travel time of private vehicle was increased by 6 minutes, 2.75% of travelers would divert from private vehicle to public transport. If travel cost of motorized paratransit decreased by 20%, then 0.26% would shift from non-motorized paratransit to motorized paratransit. Results of the study indicated that the decrease in travel cost and increase in travel time of private vehicle did not make any difference to the travelers of Khon Kaen. Though travel time and travel cost are the important factors for the modal split analysis which were not true in case ofKhon Kaen travelers. |