Establishing a knowledge graph-based recommendation system for product family reconfiguration and redesign

• Main Author: Harimane, Nishanth
• Other Authors: Chen Chun-Hsien
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Manufacturing::Product engineering; Engineering::Computer science and engineering::Computer applications::Computer-aided engineering
• Online Access: https://hdl.handle.net/10356/168201
• Institution: Nanyang Technological University

Knowledge graphs are an upcoming method of storing information and understanding relationships between various entities. It has mainly been used in the technology sector to simplify and quicken processes to search relations between entities compared to other conventional methods. The application of knowledge graphs to other industries such as engineering and manufacturing has been recent and mainly focused on its applications in storage. This research paper will explore a newer application of knowledge graphs in engineering and manufacturing applications. This paper explores the possibility of the applications of knowledge graphs in the concept of product family design, a commonly used concept in engineering and manufacturing. The aim of this paper is to develop a knowledge graph system that could make the product family redesign process more efficient and effective so that the entire process can be streamlined. After developing a knowledge graph of the object in question, methods to determine the component that would need to go through redesign based on specific scenarios are explored. It was determined that in either method that was tested, information about the scenario was necessary to identify the component required to be redesigned. This would mean that without the specific information about requirements for the scenarios, the knowledge graph would not be effective in determining the component that would require redesign. In this paper, due to the lack of specification, it was not possible to determine the component that required redesign unless the requirements were clearly present in the knowledge graph itself. However, how the graph was traversed to find components was optimised by categorizing the parts that could go through product family redesign thus making it more efficient. This paper as such would be able to have implications about how the knowledge graph for a product is constructed to allow for the efficiency of identification of component fit for redesign.