Optimization of annual planned rail maintenance

Research on preventative rail maintenance to date majors small or artificial problem instances, not applicable real-world use cases. This article tackles large, scheduling problems. Maintenance costs and availability of the infrastructure need be optimized, while adhering a set complex constraints. We develop compare three generic approaches: an evolution strategy, greedy metaheuristic, hybrid two. As case study, we schedule major preventive full year in complete Netherlands, one busiest networks Europe. Empirical results two datasets show approach delivers high-quality schedules.