A comparison of three heuristics on a practical case of sub-daily staff scheduling

Sub-daily personnel planning, which is the focus of our work offers considerable productivity reserves for companies in certain industries, such as logistics, retail and call centres. However, it also creates complex challenges for the planning software. We compare particle swarm optimisation (PSO), the evolution strategy (ES) and a constructive agent-based heuristic on a set of staff scheduling problems derived from a practical case in logistics. All heuristics significantly outperform conventional manual full-day planning, demonstrating the value of sub-daily scheduling heuristics. PSO delivers the best overall results in terms of solution quality and is the method of choice, when CPU-time is not limited. The approach based on artificial agents is competitive with ES and delivers solutions of almost the same quality as PSO, but is vastly quicker. This suggests that agents could be an interesting method for real-time scheduling or re-scheduling tasks.