Hierarchically Structured Scheduling and Execution of Tasks in a Multi-agent Environment

In a warehouse environment, tasks appear dynamically. Consequently, task management system that matches them with the workforce too early (e.g., weeks in advance) is necessarily sub-optimal. Also, rapidly increasing size of action space such consists significant problem for traditional schedulers. Reinforcement learning, however, suited to deal issues requiring making sequential decisions towards long-term, often remote, goal. this work, we set ourselves on presents itself hierarchical structure: task-scheduling, by centralised agent, dynamic multi-agent environment and execution one schedule, decentralised agents only partial observability thereof. We propose use deep reinforcement learning solve both high-level scheduling low-level schedule execution. The topic contribution relevant operations research scientific communities directed future real-world industrial applications.