Generating robust schedules for train maintenance staff

Publication date

DOI

Document Type

Master Thesis

Collections

Open Access logo

License

CC-BY-NC-ND

Abstract

Train units are cleaned, inspected, and parked on shunting yards when not needed in service. One of the components of automatically generating plans for such a train service yard is to assign all tasks that need to be performed to personnel. These tasks are subject to release dates, deadlines, precedence constraints, and are situated at different locations at the shunting yard. We propose an approach using a greedy heuristic to obtain an initial solution, and a local search improvement step, optimizing a combination of the flexibility, fairness, and walking distances. The method is able to consistently find succesful results for realistic scenarios.

Keywords

staff scheduling;local search;flexibility;robustness;simple temporal network;metaheuristic;maintenance

Citation