Existing research seldom focuses on analysis of college students’ travel behavior. As a special travel group, college students can generate massive travel demand during the beginning and end days of the holiday, causing great pressure to the traffic systems around the university. Therefore, this paper studies the travel mode choice behavior of college students by using the MNL model. First, special RP and SP questionnaires are designed to collected college students’ travel behavior data, the questionnaires are designed based on the orthogonal design method. Second, a multi-level nested logit (MNL) model is constructed to study students' travel mode choice behavior. Car, airport bus and metro are selected as the alternatives, and travel cost, in vehicle time and transfer time are selected as the attribute variables of the alternatives. The MNL model analysis the parameters of the model, and estimates the significant factors affecting college students’ travel behavior. Finally, taking the travel mode choice behavior of Chang'an university students to Xi'an Xianyang Airport as an example, the feasibility of the proposed method is verified. Results show that students who choose cars pay more attention to travel expenses than in car time; students who choose airport bus are sensitive to travel cost and in car time; students who choose airport railway cares more about travel cost and travel time punctuality. This study can provide theoretical support for traffic planning and management around universities. Key words: Travel behavior; choice of travel mode; MNL model; RP&SP questionnaire; college student

CICTP