Reinforcement Learning Approach for Multi-Agent Flexible Scheduling Problems

TL;DR

This paper introduces a Reinforcement Learning method for multi-agent flexible scheduling problems, providing a new environment and heuristic-guided Q-Learning solution that achieves state-of-the-art results in job shop scheduling.

Contribution

It presents an open-source gym environment and a novel RL approach with heuristics for improved scheduling performance.

Findings

Achieved state-of-the-art performance in multi-agent scheduling

Developed an open-source environment for scheduling problems

Demonstrated effectiveness of heuristic-guided Q-Learning

Abstract

Scheduling plays an important role in automated production. Its impact can be found in various fields such as the manufacturing industry, the service industry and the technology industry. A scheduling problem (NP-hard) is a task of finding a sequence of job assignments on a given set of machines with the goal of optimizing the objective defined. Methods such as Operation Research, Dispatching Rules, and Combinatorial Optimization have been applied to scheduling problems but no solution guarantees to find the optimal solution. The recent development of Reinforcement Learning has shown success in sequential decision-making problems. This research presents a Reinforcement Learning approach for scheduling problems. In particular, this study delivers an OpenAI gym environment with search-space reduction for Job Shop Scheduling Problems and provides a heuristic-guided Q-Learning solution with state-of-the-art performance for Multi-agent Flexible Job Shop Problems.