Modeling on Mode Choice Behavior Based on Trip Chaining: A Case Study in

A Multinomial logit model is adopted to analyze mode choice behavior based on household travel survey data in 2010 in Zhongshan, which is a medium city of Guangdong province in China. The total mode share includes walk, bicycle, e-bike, motorcycle, car and bus. Since travelers often chain multiple activities together in a trip chain for greater efficiency and convenience, the influence of trip chain on mode choice is considered in this research by classifying trips to single trip, simple trip chain and complex trip chain. The model estimation results show that trip chaining has significant effects on mode choice and travelers scheduling more activities in their trip chain are more likely to choose flexible modes such as automobile, motorcycle, e-bike, and bicycle. These findings have important implication in increasing public transportation service and improving activity-based modeling system.