Integrated resolution of public transport planning problems: focus on timetabling and vehicle scheduling with heterogeneous fleet

In this paper, we propose a new Integer Linear Programming model, based on a time-space network that integrates the timetable generation problem and the vehicle scheduling problem with heterogeneous fleet. A difference of this approach consists in considering the demand for the timetable redefinition and the vehicle scheduling, factor rarely applied in optimization models of the transportation system. We applied real and large random instances. The results indicate that the model may contribute to optimizing the public transport planning leading to significant savings in terms of scheduled vehicles. Moreover, as the timetable changes are fairly short, it is slightly modified, minimally modifying the passengers routine, which enables the application of these approaches to real context.