Application of Knowledge-Augmented Generative Agents in Dynamic Scheduling for Discrete Manufacturing Workshop

View at publisher Summarise with VIRAT
Authors: Yanpeng Wang, Yuanjian Zhong, Xu Sun
Cited by: 0
References: 15

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Discrete manufacturing workshops face dynamic events such as order disruptions and energy price fluctuations, necessitating rapid adjustments to production scheduling. Traditional approaches rely on predefined rules or optimization model reconstruction, limiting responsiveness and adaptability. This paper proposes a knowledge-enhanced generative agent that constructs a domain knowledge base containing process rules, equipment parameters, and historical strategies. This base provides contextual information to large language models, enabling automatic generation of scheduling solutions from disturbance commands. In an energy disturbance response scenario at an automotive component assembly workshop, the agent generated a scheduling solution within seconds. Peak-hour loads were significantly reduced, production tasks were completed on schedule, and overall profitability increased.

Fields of Study & Subjects

Scheduling and Optimization Algorithms Impact of AI and Big Data on Business and Society Digital Transformation in Industry

Publication details

Journal / source
International Conference on Smart City and Information System (ICSCIS)
Pages
23–26
Publication date
30 January 2026
Language
EN