Automatic Generation of Maintenance Worker Schedules at Railway Rolling Stock Depots by Tabu Search

This paper proposes a hierarchical optimization framework for automatic generation of maintenance worker schedules at rolling stock depots. Maintenance worker schedules at rolling stock depots are daily schedules assigned to maintenance worker teams such as cleaning rolling stock and performing inspection during turnaround operations of superior trains. To evaluate the effectiveness of the proposed method, the schedules generated by the proposed method are compared with actual schedules, those generated by a mathematical programming method, and local search. The results indicate that the proposed method can generate maintenance worker schedules satisfying non-relaxable constraints and high priority relaxable constraints within a practical computation time of three minutes with levelized and reduced worker loads.