Solver Based Heuristics for Rolling Stocks Corrective Maintenance Scheduling

Trains have a constrained schedule and are not available on demand. If a malfunction is detected, the moment and the place to fix the problem while keeping the network at its optimal use may not be easy to find, given that there will be multiple problems throughout the network. Deciding of the maintenance requests the solution of a scheduling problem. The goal of this research is to provide an efficient solution to this problem. To achieve this aim, we develop a mixed integer linear programming model, a constraint programming model and multiple heuristic algorithms such as a Local Branching (LB) over the mixed integer linear programming model and a Variable Partitioning Local Search (VPLS) over the constraint programming model.