Rolling Stock Rostering Optimization Under Maintenance Constraints

This article presents a mixed-integer linear-programming formulation for integrating short-term maintenance planning in a network-wide railway rolling stock circulation problem. This is a key problem in railway rostering planning that requires covering a given set of services and maintenance works with a minimum amount of rolling stock units. In our formulation, a rostering solution is viewed as a minimal cost Hamiltonian cycle in a graph with service pairings, empty runs, and short-term maintenance tasks. We use a commercial MILP solver to compute efficient solutions in a short time. Experimental results on real-world scenarios from the main Italian railway company Trenitalia show that this integrated approach can reduce significantly the number of trains and empty runs when compared with the current rolling stock circulation plan.