Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Xiangjie Xiao, Cong Zhang, Wen Song, Zhiguang Cao

TL;DR

This work introduces \textsc{ReSched}, a minimalist DRL framework that rethinks both the scheduling formulation and model design of FJSP, and outperforms classical dispatching rules and state-of-the-art DRL methods on FJSP.

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.

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

TL;DR

This work introduces \textsc{ReSched}, a minimalist DRL framework that rethinks both the scheduling formulation and model design of FJSP, and outperforms classical dispatching rules and state-of-the-art DRL methods on FJSP.

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.
Paper Structure (71 sections, 1 theorem, 17 equations, 6 figures, 12 tables, 2 algorithms)

This paper contains 71 sections, 1 theorem, 17 equations, 6 figures, 12 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Work
  3. Preliminary
  4. Flexible Job Shop Scheduling Problem (FJSP)
  5. Transformer Block
  6. Methodology
  7. State Representation
  8. Revisiting the Scheduling Formulation
  9. MDP Formulation
  10. State: Minimal Representation
  11. State: Subproblem
  12. State: Features
  13. Action
  14. Transition
  15. Reward
  16. ...and 56 more sections

Key Result

Proposition 1

For any two scheduling trajectories $\tau_1$ and $\tau_2$ that reach the same state $\mathcal{S}_t$, the corresponding remaining subproblems share an identical feasible solution set.

Figures (6)

  • Figure 1: (a) Illustration of the formulation for a 2‑job, 2‑operation, 2‑machine (2-2-2) FJSP instance. (b) Changes in topology and O2M/O2O Connection between two steps for a 3-3-3 instance.
  • Figure 2: The ReSched framework. The state consists of four information, resulting in four key features (available time and minimum duration for operation; available time for machine; duration), and incorporates three (underlined) architectural enhancements for Transformer-based network. For decision-making module, we concatenate the embeddings of each feasible operation–machine pair, feed them into an MLP to obtain a score.
  • Figure 2: Results on Taillard Benchmark for JSSP.
  • Figure 3: Running Time on FJSP Benchmark
  • Figure 4: The illustration of JSSP and FFSP instances, respectively.
  • ...and 1 more figures

Theorems & Definitions (4)

  • Definition 4.1
  • Proposition 1: State-dependent Optimality in Scheduling
  • Remark 1
  • proof