Phenyane, Siphamandla Sifiso2025-03-182023Phenyane, Siphamandla Sifiso. (2023). Designing an AI-based Predictive Maintenance Framework to Improve OEE for an Automotive Manufacturer in South Africa [Master’s dissertation, University of the Witwatersrand, Johannesburg].WireDSpace.https://hdl.handle.net/10539/44355https://hdl.handle.net/10539/44355A research report submitted in partial fulfillment of the requirements for the degree of Master of Management in the field of Digital Business to the Faculty of Commerce, Law, and Management, Wits Business School, University of the Witwatersrand, Johannesburg, 2023This study presents a comprehensive approach to enhancing Overall Equipment Effectiveness (OEE) in an automotive manufacturing setting by integrating artificial intelligence and Internet of Things technologies with people and processes. The research applies the Design Science Research (DSR) methodology to develop, deploy, and evaluate an AI-driven predictive maintenance framework. The first phase involves a detailed exploratory data analysis to understand the current state of OEE and identify critical bottlenecks within the production line, particularly operating below OEE industry standards. The second phase builds on the insights gathered from a semi-structured survey conducted among field experts, leading to the formulation of a cohesive framework that synergises the socio-technical aspects of the manufacturing environment. The core of the study revolves around the design and development of a Bi-directional Long Short-Term Memory (Bi-LSTM) model artefact, capable of analysing sequential data to predict the Remaining Useful Life (RUL) of machinery, thereby pre-empting production halts. The model's predictive capability is rigorously tested and validated using historical IoT data, demonstrating a high degree of accuracy across different spot-welding locations. Overall, the study highlights the critical role of AI in transforming manufacturing processes, emphasising the need for continuous adaptation and improvement of predictive models to maintain operational efficiency. The proposed framework aims to serve as a strategic tool in lean manufacturing, contributing to smoother operations and improved OEE in automotive manufacturing settingsen© 2025 University of the Witwatersrand, Johannesburg. All rights reserved. The copyright in this work vests in the University of the Witwatersrand, Johannesburg. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of University of the Witwatersrand, Johannesburg.Automotive manufacturingDowntimeFault prognosisInternet-of-ThingsKnowledge-based methodsUCTDDissertationUniversity of the Witwatersrand, JohannesburgSDG-8: Decent work and economic growthSDG-12: Responsible consumption and production