Variational Approach for Job Shop Scheduling

TL;DR

This paper introduces a variational graph-based framework for job shop scheduling that improves generalization and training stability by decoupling representation learning from policy optimization, outperforming existing methods especially on large, complex instances.

Contribution

The paper pioneers the application of variational inference to JSSP, creating a probabilistic framework that enhances robustness and generalization in scheduling solutions.

Findings

Superior zero-shot generalization on benchmark instances

Enhanced training stability and robustness

Outperforms state-of-the-art DRL and traditional methods

Abstract

This paper proposes a novel Variational Graph-to-Scheduler (VG2S) framework for solving the Job Shop Scheduling Problem (JSSP), a critical task in manufacturing that directly impacts operational efficiency and resource utilization. Conventional Deep Reinforcement Learning (DRL) approaches often face challenges such as non-stationarity during training and limited generalization to unseen problem instances because they optimize representation learning and policy execution simultaneously. To address these issues, we introduce variational inference to the JSSP domain for the first time and derive a probabilistic objective based on the Evidence of Lower Bound (ELBO) with maximum entropy reinforcement learning. By mathematically decoupling representation learning from policy optimization, the VG2S framework enables the agent to learn robust structural representations of scheduling instances through a variational graph encoder. This approach significantly enhances training stability and robustness against hyperparameter variations. Extensive experiments demonstrate that the proposed method exhibits superior zero-shot generalization compared with state-of-the-art DRL baselines and traditional dispatching rules, particularly on large-scale and challenging benchmark instances such as DMU and SWV.