Self-Evaluation for Job-Shop Scheduling

2025-02-12

AI Summary

Job Shop Scheduling (JSP), an NP-hard combinatorial optimization problem, suffers from error accumulation in conventional sequential decision-making approaches, degrading solution quality. This paper proposes the first self-evaluation-driven parallel solving framework for JSP: it abandons incremental construction and instead jointly models task–resource dependencies via a heterogeneous graph neural network integrated with a Transformer, enabling concurrent learning of policy generation and solution quality assessment. A novel self-evaluation mechanism enables parallel generation and filtering of task-assignment subsets, optimized end-to-end via reinforcement learning. Evaluated on multiple classical benchmarks, the method achieves average makespan improvements of 3.2%–7.8% over state-of-the-art approaches, demonstrating the effectiveness, robustness, and generalizability of the self-evaluation paradigm in combinatorial optimization.

Technology Category

• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Search and Optimization: Combinatorial Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic 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: Cost models of using LLMs in production systems

Abstract

Combinatorial optimization problems, such as scheduling and route planning, are crucial in various industries but are computationally intractable due to their NP-hard nature. Neural Combinatorial Optimization methods leverage machine learning to address these challenges but often depend on sequential decision-making, which is prone to error accumulation as small mistakes propagate throughout the process. Inspired by self-evaluation techniques in Large Language Models, we propose a novel framework that generates and evaluates subsets of assignments, moving beyond traditional stepwise approaches. Applied to the Job-Shop Scheduling Problem, our method integrates a heterogeneous graph neural network with a Transformer to build a policy model and a self-evaluation function. Experimental validation on challenging, well-known benchmarks demonstrates the effectiveness of our approach, surpassing state-of-the-art methods.

Problem

• Addresses NP-hard job-shop scheduling challenges
• Overcomes error accumulation in sequential decision-making
• Proposes self-evaluation framework for scheduling optimization

Innovation

• Neural Combinatorial Optimization
• Heterogeneous Graph Neural Network
• Self-Evaluation Transformer