RESCHED: Rethinking Flexible Job Shop Scheduling from a Transformer-based Architecture with Simplified States

2026-03-07

AI Summary

This work addresses the limited generalization capability of existing approaches to the flexible job shop scheduling problem (FJSP), which often rely on complex handcrafted features or graph neural networks. To overcome this challenge, the authors propose a lightweight deep reinforcement learning framework. By reformulating the FJSP as a Markov decision process with a state space reduced to four core features, they design a scheduling-oriented, lightweight Transformer architecture incorporating a subproblem-aware state representation, dot-product attention mechanism, and three targeted structural enhancements. The resulting method substantially reduces modeling complexity, outperforms conventional dispatching rules and current deep reinforcement learning approaches on FJSP, and achieves performance comparable to specialized neural models on related variants such as the job shop scheduling problem (JSSP) and the flexible flow shop scheduling problem (FFSP), demonstrating strong cross-problem generalization.

Technology Category

• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Search and Optimization: Learning to Search
• Machine Learning: Deep Neural Architectures and Foundation Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work

Abstract

Neural approaches to the Flexible Job Shop Scheduling Problem (FJSP), particularly those based on deep reinforcement learning (DRL), have gained growing attention in recent years. However, existing methods rely on complex feature-engineered state representations (i.e., often requiring more than 20 handcrafted features) and graph-biased neural architectures. To reduce modeling complexity and advance a more generalizable framework for FJSP, we introduce \textsc{ReSched}, a minimalist DRL framework that rethinks both the scheduling formulation and model design. First, by revisiting the Markov Decision Process (MDP) formulation of FJSP, we condense the state space to just four essential features, eliminating historical dependencies through a subproblem-based perspective. Second, we employ Transformer blocks with dot-product attention, augmented by three lightweight but effective architectural modifications tailored to scheduling tasks. Extensive experiments show that \textsc{ReSched} outperforms classical dispatching rules and state-of-the-art DRL methods on FJSP. Moreover, \textsc{ReSched} also generalizes well to the Job Shop Scheduling Problem (JSSP) and the Flexible Flow Shop Scheduling Problem (FFSP), achieving competitive performance against neural baselines specifically designed for these variants.

Problem

• Flexible Job Shop Scheduling Problem
• deep reinforcement learning
• state representation
• model complexity
• generalization

Innovation

• Flexible Job Shop Scheduling
• Deep Reinforcement Learning
• Transformer Architecture
• State Simplification
• Generalizable Scheduling