Skip to main navigation Skip to search Skip to main content

FiLM-Conditioned Cross-Attention Multi-Objective Reinforcement Learning Algorithm for Dynamic Seru Production System Reconfiguration

  • Zihan Lan
  • , Yaoxin Zhang
  • , Hongbo Jin
  • , Dongni Li*
  • *Corresponding author for this work
  • Beijing Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Seru production systems (SPSs), enhancing flexibility and responsiveness compared to traditional assembly lines, face challenges in addressing demand uncertainty characterized by unpredictable and time-varying customer orders. This study proposes a novel bi-objective optimization model for the dynamic seru reconfiguration problem, aiming to simultaneously minimize tardiness and operational cost. We develop a multi-objective reinforcement learning algorithm incorporating a FiLM-conditioned cross-attention mechanism. Computational results demonstrate that our method significantly improves the handling of demand uncertainty, advancing practical production decision-making in dynamic SPS environments.

Original languageEnglish
Title of host publication38th Chinese Control and Decision Conference, CCDC 2026
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages3540-3546
Number of pages7
ISBN (Electronic)9798331550707
DOIs
Publication statusPublished - 2026
Externally publishedYes
Event38th Chinese Control and Decision Conference, CCDC 2026 - Nanjing, China
Duration: 15 May 202618 May 2026

Publication series

Name38th Chinese Control and Decision Conference, CCDC 2026

Conference

Conference38th Chinese Control and Decision Conference, CCDC 2026
Country/TerritoryChina
CityNanjing
Period15/05/2618/05/26

Keywords

  • Multi-objective reinforcement learning
  • Reconfiguration
  • Responsiveness
  • Seru production system
  • Tardiness cost

Fingerprint

Dive into the research topics of 'FiLM-Conditioned Cross-Attention Multi-Objective Reinforcement Learning Algorithm for Dynamic Seru Production System Reconfiguration'. Together they form a unique fingerprint.

Cite this