Skip to Main Content
This paper examines a real-life maintenance scheduling problem in a power system. Due to the large search space, it proved difficult to reach an optimal solution using a traditional genetic algorithm (GA). In order to improve the GA performance, the problem was divided into several layers, each layer representing a part of the initial problem. A modified GA was used to built sub-schedules, satisfying specified criteria. The sub-schedules were used as single genes to built up a complete schedule.