A Petri net model for vehicle scheduling problem

Abstract

Vehicle scheduling problem consists in assigning vehicles to a set of time-tabled trips, wishing to minimize the objective cost function. Two aprroaches of the problem were studied: single-depot vehicle scheduling problem (SDVSP) and multiple-depot vehicle scheduling problem (MDVSP). There are known several methos of solving this problem based mainly on liniar programming, branch and bound, graph colouring thechniques. SDVSP is solvable in polynomial time, while MDSVP is NP hard. The main propose of this paper is to present a model based on Petri nets for the vehicle sceduling problem. We defined priced timed workflow nets as a suitable model for SDVSP. We tried to emphasize the advantages of using Petri nets in order to describe this kind of planning problems.