Customized Commuting Bus Route and Timetable Optimization Using Mobile Internet Data

Customized commuting bus (CCB) is an attractive and direct transit service that satisfies the long-distance demand of commuters with similar origins and destinations. However, existing CCB routes are mainly manually generated with the aid of prebooking data. This study proposes a CCB route and timetable design framework that is applicable to smartphone-based mobile internet data. The data source records individual travelers' trip information via map-based navigation software. Potential CCB demands are spatially and temporally clustered in the first phase. On the basis of passively collected demand data, the bus route and timetable optimization problem can be formulated as a mixed integer linear program for each cluster in the second phase. The logistic regression model is also employed to capture commuters' mode choice behavior. A case study is conducted on three-month travel query data in the Zhangjiang region, Shanghai, China. Result demonstrates that the proposed framework can quickly and conveniently yield feasible and satisfactory solutions.