Short-term rolling stock rostering and maintenance scheduling optimization.

This paper describes an optimization framework for rolling stock rostering and maintenance scheduling at a busy workshop. A new mixed-integer linear-programming formulation is provided for the maintenance scheduling problem faced by Trenitalia (train operating company) managers, with input data taken from the rolling stock rostering plan. The computational results are carried out on a Trenitalia's maintenance site located in Naples. The solutions computed via a commercial MILP solver are compared with practical solutions. A relevant cost reduction is possible by using the proposed framework, involving both rostering and maintenance scheduling. We also show how the proposed method can be used as an effective tool to absorb real-time timetable perturbations while respecting the agreed level of service. [ABSTRACT FROM AUTHOR]