To analysis the factors that influence the mode choice behavior for commuters in the metropolitan area, a questionnaire was designed from considering individual attributes, family attributes, and travel attributes. The Nested Logit (NL) model was proposed to examine commuter travel characteristics. At the same time, to verify the evaluation effect, the Support Vector Machine (SVM) model was adopted to compare with the NL model from the accuracy of traffic mode prediction. Base on analyzing the significant influence factor in the metropolitan area, the traffic mode changes after policy adjustments were studied by using the model with high prediction accuracy. The results show that the commuter travel time, travel costs and transfer times are negative in the NL model coefficients, and the effect is significant, the average travel mode prediction accuracy of the NL model is 70.7%, the SVM model is more substantial 90.1%. The SVM model predicts the travel mode and calculates the changes after four policy adjustments respectively. The data shows that the average proportion of buses, subway and train has increased by 5.68%, 0.74% and 4.43%, the car has decreased by 7.23% after comprehensive policy adjustment, which indicates that policy adjustments can effectively improve the percentage of public transportation. It is worth noting that the proportion of buses has declined by 6.54% in Langfang after travel time policy adjustment only, which means policy can not always play a good role for each district to optimize the proportion structure, and policy need to be considered comprehensively.

CICTP