E-Thesis 296 views 102 downloads
The Sustainable Adoption of Industry 4.0 and Machine Learning in the Automotive Industry / JAMES FLYNN
Swansea University Author: JAMES FLYNN
-
PDF | Redacted version - restricted access
Copyright: The Author, James Flynn, 2023.
Download (27.33MB)
DOI (Published version): 10.23889/SUthesis.63690
Abstract
The automotive industry is undergoing a major transformation. New environmental legislation, changing consumer requirements, Industry 4.0 technologies, and advancements in battery technologies, have contributed to an industry-wide shift towards electric powertrain. To remain competitive in this ra...
| Published: |
Swansea, Wales, UK
2023
|
|---|---|
| Institution: | Swansea University |
| Degree level: | Doctoral |
| Degree name: | Ph.D |
| Supervisor: | Giannetti, Cinzia. |
| URI: | https://cronfa.swan.ac.uk/Record/cronfa63690 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract: |
The automotive industry is undergoing a major transformation. New environmental legislation, changing consumer requirements, Industry 4.0 technologies, and advancements in battery technologies, have contributed to an industry-wide shift towards electric powertrain. To remain competitive in this rapidly changing environment, automotive manufacturers must ensure high levels of technical and organisational innovation to transition towards digital and data-driven business practices. This research aims to address these growth opportunities and manage ongoing change in three steps. First, the literature on machine learning applications in automotive manufacturing is critically reviewed and the barriers to developing and implementing machine learning are discussed. Secondly, a structured framework is developed to assess the industry 4.0 maturity of automotive manufacturing operations and guide digital transformation at the factory level. In the third and final step of this research, two machine learning projects identified by the assessment are presented in detail. The first case study presents an anomaly detection solution to identify process errors in engine assembly. This research introduces multiple advancements in anomaly detection in manufacturing, including the introduction of the Anomaly No Concern class. The second case study is a greenfield project to explore new digital value chains to add value to EV customers and explore dataas-a-service business models. This case study uses a combination of Google Street View data and GIS data to identify houses suitable for EV charging and represents a major advancement towards fully automated remote surveying of the built environment. To conclude, multiple advancements are presented that contribute to the academic literature. Clear stepwise frameworks support the proposed industrial solutions to develop, implement and replicate these solutions across the business. These contributions have been used to support ongoing digitalisation efforts, implement state-of-the-art anomaly detection solutions, and explore new data-as-a-service business models in one of the world's largest automotive companies. |
|---|---|
| Item Description: |
A selection of third party content is redacted or is partially redacted from this thesis due to copyright restrictions. |
| Keywords: |
Machine Learning, Industry 4.0, Anomaly Detection, Image Classification, Maturity Model, Change Management |
| College: |
Faculty of Science and Engineering |
| Funders: |
Ford Motor Company, UK Engineering and Physical Sciences Research (EP/S001387/1 and EP/L015099/1), M2A European Social Fund via the Welsh Government (c80816), |