Solving the combinatorial puzzle

To find optimal schedules for maintenance is a difficult combinatorial puzzle that belongs to the class of so-called NP-hard problems, for which only algorithms with an exponentially growing running time are known. Researchers in the CIMPLO project managed to model maintenance scheduling as a multiobjective problems and devised algorithms that in feasible time can produce near optimal schedules. This resulted in two algorithms that were published and presented in top international conferences in computer science held in New Zealand and, resp., Prague this year. The first algorithm was developed by Yali Wang in her PhD research, and she investigated the maintenance scheduling of car fleets. The second algorithm was a so called tabu-search approach which combines exploration and exploitation and developed by Marios Kefalas in his PhD research. A next step will be to integrate statistical predictions of remaining useful lifetime of components in the modelling, to make the schedules more realistic and robust.Wang, Y., Limmer, S., Olhofer, M., Emmerich, M. T., & Bäck, T. (2019, June). Vehicle Fleet Maintenance Scheduling Optimization by Multi-objective Evolutionary Algorithms. In 2019 IEEE Congress on Evolutionary Computation (CEC) (pp. 442-449). IEEE.