---
title: "Preference-Guided Multi-Policy Optimization for Flexible Job Shop Scheduling"
authors:
  - "Inguk Choi"
  - "Woo-Jin Shin"
  - "Sang-Hyun Cho"
  - "Hyun-Jung Kim"
doi: "10.24963/ijcai.2026/685"
url: "https://ijcai.org/proceedings/2026/685"
pdf: "https://ijcai.org/proceedings/2026/0685.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "6156-6165"
track: "Main Track"
keywords:
  - "Planning and Scheduling: Search in planning and scheduling"
  - "AI: Planning and Scheduling"
  - "Planning and Scheduling: Scheduling"
  - "Planning and Scheduling: Learning in planning and scheduling"
---

# Preference-Guided Multi-Policy Optimization for Flexible Job Shop Scheduling

Inguk Choi, Woo-Jin Shin, Sang-Hyun Cho, Hyun-Jung Kim

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 6156-6165. DOI: [10.24963/ijcai.2026/685](https://doi.org/10.24963/ijcai.2026/685)

## Abstract

Under the shift toward Industry 4.0, mass-customized manufacturing systems have introduced complex scheduling problems, such as the Flexible Job Shop Scheduling Problem (FJSSP). Recent Deep Reinforcement Learning (DRL)-based heuristics have shown promise, yet existing methods often suffer from two key limitations: they typically rely on single-policy optimization, which limits exploration, and on imprecise reward functions, which fail to accurately reflect decision quality. To address these challenges simultaneously, we propose PGMPO (Preference-Guided Multi-Policy Optimization), a novel learning framework consisting of (1) a simple but effective multi-policy modeling approach that allows a single network to represent multiple decision-makers, and (2) a preference-driven model optimization method that effectively guides policies to learn diverse and specialized problem-solving strategies without the need for explicit reward functions. Experimental results demonstrate that PGMPO substantially boosts the performance of existing neural solvers across several benchmarks.