Towards a formal ontology to support knowledge sharing across product design and manufacturing

Abstract

The ability to seamlessly share knowledge across different product lifecycle domains is a crucial enabler for decision making. It dictates the competence of a manufacturing enterprise. This transcends over to Information and Communication Technology (ICT) systems, which are increasingly becoming an integral part of design and manufacturing stages. In today’s competitive manufacturing world, these systems are required to seamlessly share knowledge for better, faster and cheaper production. However, different manufacturing domains have different data structures and syntaxes leading to knowledge sharing issues. Furthermore, the loosely defined semantic of the contributing concepts and relations lead to different, sometimes contradicting interpretations. Thus, the knowledge sharing capability of such systems across design and manufacturing domains are impeded. A computationally interpretable ontology can resolve these issues by providing a basis for common understanding across these domains.

In this thesis, a unique solution in the form of a Product Lifecycle Ontology (PLO) is proposed that facilitates semantic knowledge sharing across product design and manufacture. The proposed ontology supports this by providing a common semantic base that provides a route to link domains and enable knowledge sharing. The research work demonstrates sharing of knowledge from Machining, Welding and Inspection with Design. This is achieved by defining a set of concepts and relations with rigorous formal semantics. An approach to specialise these concepts at multiple new levels to capture the varying depth of meanings with higher granularity has been presented. This has further been utilised to develop a novel model for classification of joining and welding processes that facilitates reconciliation of international welding standards. Similarly, an innovative way to categorise different types of manufacturing operations and efficiently model their sequences was unveiled.

The ontology is verified experimentally and through and an industrial case study. The research work has shown the potential to reduce the number of design revisions by capturing the manufacturing specific knowledge and share it with product design. Further, the recommendation of this work is ready to be fed into the technical committees overseeing the welding standards to improve them for better interoperability. Thus, the proposed ontology expands previous works and fills in the existing research gaps within the area of formal manufacturing reference ontologies.

Date of Award	Jun 2021
Original language	English
Awarding Institution	Coventry University
Sponsors	Institute for Advanced Manufacturing Engineering
Supervisor	Weidong Li (Supervisor), Nazaraf Shah (Supervisor) & Zahid Usman (Supervisor)