A Customizable Simulator for Artificial Intelligence Research to Schedule Semiconductor Fabs

Benjamin Kovacs, Pierre Tassel, Ramsha Ali, Mohammed El-Kholany, Martin Gebser, Georg Seidel

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

Abstract

Optimal scheduling of semiconductor fabs is a huge challenge due to the problem scale and complexity. New dispatching strategies are usually developed and tested using simulators of different fidelity levels. This work presents a scalable, open-source tool for simulating factories up to real-world size, aiming to support the research into new scheduling algorithms from prototyping to large-scale experiments. The simulator comes with a declarative environment definition framework and is out of the box usable with existing reinforcement learning methods, priority-based rules, or evolutionary algorithms. We verify our tool on large-scale public instances and provide proof-of-concept demonstrations of the reinforcement learning interface's usage.

Original languageEnglish
Title of host publication2022 33rd Annual SEMI Advanced Semiconductor Manufacturing Conference, ASMC 2022
PublisherIEEE
ISBN (Electronic)9781665494878
DOIs
Publication statusPublished - 2022
Event33rd Annual SEMI Advanced Semiconductor Manufacturing Conference: ASMC 2022 - Saratoga Springs, United States
Duration: 2 May 20225 May 2022

Publication series

NameASMC (Advanced Semiconductor Manufacturing Conference) Proceedings
Volume2022-May
ISSN (Print)1078-8743

Conference

Conference33rd Annual SEMI Advanced Semiconductor Manufacturing Conference
Abbreviated titleASMC 2022
Country/TerritoryUnited States
CitySaratoga Springs
Period2/05/225/05/22

Keywords

  • benchmarking
  • gym environment
  • reinforcement learning
  • scheduling
  • semiconductor manufacturing
  • simulation

ASJC Scopus subject areas

  • General Engineering

Fingerprint

Dive into the research topics of 'A Customizable Simulator for Artificial Intelligence Research to Schedule Semiconductor Fabs'. Together they form a unique fingerprint.

Cite this