- Wednesday 25 October 2023, 1400-1500
- Hybrid: Charles Thackrah 1.05 OR Online - book below
- Peizhi Shi, CDR, University of Leeds
The integration of machine learning for decision-making in intelligent manufacturing has revolutionised the industrial sector. This innovation has not only optimised production processes, reduced costs, and enhanced product quality but has also fundamentally reshaped how decisions are made in manufacturing operations. In this presentation, Dr. Peizhi Shi will highlight his research findings on the application of machine learning in decision-making in two critical areas of manufacturing: feature recognition in traditional manufacturing and part orientation in additive manufacturing. Feature recognition involves automatically identifying features on manufactured parts, such as holes, slots, and chamfers, which can improve manufacturing efficiency and quality. Dr. Shi's research has focused on developing machine learning algorithms to recognise these highly overlapping features accurately and efficiently, eliminating the need for manual inspection and expediting decision-making processes. Part orientation in additive manufacturing refers to determining the best orientation of a 3D printed part to ensure the highest strength and accuracy. Dr. Shi's research has aimed to optimise part orientation through machine learning techniques, resulting in better efficiency in determining part orientation. Additionally, the challenges that come with implementing a learning-based decision-making system in manufacturing will also be presented.
Dr. Peizhi Shi is a Lecturer in Applied Artificial Intelligence at the Centre for Decision Research, University of Leeds. In March 2019 he completed his Ph.D. in Computer Science from the Machine Learning and Optimization Group, Department of Computer Science, University of Manchester. From 2019 to 2023, he served as a research fellow in Machine Learning at the Maths Group of the EPSRC Future Advanced Metrology Hub. His research interests are machine learning and its applications in decision-making and intelligent systems development. He has more than ten years of research experience in machine learning, focusing on computer vision, 3D object detection, deep learning, semi-supervised learning, small-sample learning, and generalisable reinforcement learning. His current research topics cover various aspects of Industry 4.0, which involve expanding the role of machine learning in manufacturing, engineering management, and operational research. Additionally, he would also like to expand his research focus to include broader areas within business and social contexts.