Machine Learning with Heuristic Search-based Hybrid Framework for Cycle Time Optimization in Semiconductor Production - Paper metadata

Aug 5, 2026

Upload by Eden

This paper optimizes cycle time (CT) in semiconductor wafer production, a key driver of operational efficiency and competitiveness in semiconductor manufacturing. A hybrid methodology combines statistical analysis with machine learning to determine the best KPI configuration for individual tools and to minimize CT. Hyperparameter tuning and model optimization are carried out using Sequential Quadratic Programming (SQP), Particle Swarm Optimization (PSO), and Genetic Algorithm (GA), including constraint handling for KPI limitations via a hierarchical decomposition approach. Experimental results indicate that random forest with GA delivers the strongest CT reduction.

LanguageEnglish

CategoryTechnology

FormatPDF

Pages7

Reading Time18 min

AccessFree to download

Abstract

Introduction

Methodology

KPI selection and hybrid modeling

Hyperparameter tuning and optimization (SQP/PSO/GA)

Constraint handling and hierarchical decomposition

Experimental results

Algorithm comparison for CT reduction

Machine Learning with Heuristic Search-based Hybrid Framework for Cycle Time Optimization in Semiconductor Production Rahman, M. S. , Islam, M. M. M. , Prasad, G. , Bhattacharyya, S. , McCreadie, K. , Reiter, T. , & Parker, N. (2025) . Machine Learning with Heuristic Search-based Hybrid Framework for Cycle Time Optimization in Semiconductor Production. In 2025 IEEE Conference on Artificial Intelligence (CAI) (pp. 1286-1291) . IEEE. Advance online publication. [https://doi.org/10.1109/CAI64502.2025.00224](https://doi.org/10.1109/CAI64502.2025.00224) Link to publication record in Ulster University Research Portal Published in: 2025 IEEE Conference on Artificial Intelligence (CAI) Publication Status: Published online: 07/07/2025 DOI: 10.1109/CAI64502.2025.00224 Document Version Author Accepted version For Author Accepted Manuscripts (AAM) published under Ulster University's Rights Retention Policy for Scholarly Works (RRPSW) When citing an AAM published under Ulster University's RRPSW please use the following citation structure: Author, A. A. (Year) . Title of article. Journal Name,[Accepted Author Manuscript] . PURE Portal URL. Licensed under CC BY 4.0. General rights The copyright and moral rights to the output are retained by the output author(s), unless otherwise stated by the document licence. Unless otherwise stated, users are permitted to download a copy of the output for personal study or non-commercial research and are permitted to freely distribute the URL of the output. They are not permitted to alter, reproduce, distribute or make any commercial use of the output without obtaining the permission of the author(s) . If the document is licenced under Creative Commons, the rights of users of the documents can be found at [https://creativecommons.org/share-your-work/cclicenses/](https://creativecommons.org/share-your-work/cclicenses/) . Take down policy The Research Portal is Ulster University's institutional repository that provides access to Ulster's research outputs. Every effort has been made to ensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in the Research Portal that you believe breaches copyright or violates any law, please contact [pure-support@ulster.ac.uk](pure-support@ulster.ac.uk) Download date: 05/08/2026 Machine Learning with Heuristic Search-based Hybrid Framework for Cycle Time Optimization in Semiconductor Production Mohammad Sharifur Rahman Intelligent Systems Research Centre Ulster University Londonderry, BT48 7JL, UK [rahman-m11@ulster.ac.uk](rahman-m11@ulster.ac.uk) Karl McCreadie Intelligent Systems Research Centre Ulster University Londonderry, BT48 7JL, UK [k.mccreadie@ulster.ac.uk](k.mccreadie@ulster.ac.uk) Girijesh Prasad Intelligent Systems Research Centre Ulster University Londonderry, BT48 7JL, UK [g.prasad@ulster.ac.uk](g.prasad@ulster.ac.uk) M.M. Manjurul Islam Intelligent Systems Research Centre Ulster University Londonderry, BT48 7JL, UK[m.islam@ulster.ac.uk](m.islam@ulster.ac.uk) Tamas Reiter Seagate Technology Londonderry, Northern Ireland, BT48 0BF, UK [tamas.reiter@seagate.com](tamas.reiter@seagate.com) Saugat Bhattacharyya Intelligent Systems Research Centre Ulster University Londonderry, BT48 7JL, UK [s.bhattacharyya@ulster.ac.uk](s.bhattacharyya@ulster.ac.uk) Nuala Parker Seagate Technology Londonderry, Northern Ireland, BT48 0BF, UK [nuala.a.parker@seagate.com](nuala.a.parker@seagate.com) Abstract—This paper aims to optimise cycle time (CT) in semiconductor wafer production, a critical factor for enhancing operational efficiency and competitiveness in the semiconductor manufacturing industry. A hybrid methodology, based on statistical analysis and machine learning (ML) techniques, is developed to identify the optimal combination of key performance indicators (KPIs) for individual tools to minimise CT. To achieve this, hyperparameter tuning

FAQs

What is the main goal of the proposed framework?

The framework aims to reduce cycle time (CT) in semiconductor wafer production by selecting and optimizing the best KPI combinations for individual tools.

Which optimization techniques are used to tune and optimize the models?

Model tuning and optimization use Sequential Quadratic Programming (SQP), Particle Swarm Optimization (PSO), and Genetic Algorithm (GA).

How do the experiments evaluate performance, and which approach performs best?

Experiments assess CT reduction across techniques and report that random forest with GA significantly outperforms other methods for reducing CT.

Aug 5, 2026

Upload by Eden