A system for distributed sharing and reuse of design and manufacturing knowledge in the PFMEA domain using a description logics-based ontology

Abstract

Potential Failure Modes and Effects Analysis in Manufacturing and Assembly Processes (PFMEA) is an important preventive method for quality assurance, and through it the decisions based on the severity levels and probabilities of occurrences and detection of the failure modes can be planned and prioritized, seeking to improve the quality of the manufactured products. This activity generates a valuable source of knowledge about the manufacturing processes in the company. However, the sharing and reuse of this knowledge is a challenge, because usually the knowledge is not semantically organized, and therefore its meaning depends on the understanding of the specialists involved. Also, there is a high fragmentation and distribution of knowledge along the production chain. Considering this scenario, this paper presents the development of a system for distributed knowledge sharing and reuse in the PFMEA domain using an ontology based on description logics, which intends to allow knowledge inference and retrieval in manufacturing environments with distributed resources. The results show that the proposed approach is adequate to support knowledge sharing and reuse, allowing: (a) knowledge representation and organization; (b) distributed knowledge inference and retrieval; (c) management of organizational knowledge on manufacturing environments with distributed resources.

Introduction

Nowadays, the social and economical environment is characterized by the appearance of new forms of industrial organizations, caused by complex factors such as market globalization, product life cycle reduction, high demand variability, need of high flexibility and responsiveness, and fast development of the technologies of information and communication [1]. In this perspective, new forms of organizational structures have been recognized by the scientific community and other professionals of this area, which include: extended enterprise, virtual enterprise, virtual organization, supply chain management and enterprise clusters [1], [2].

The key problem in these environments consists of integrating the distributed resources that contribute to production, since these new organizational structures are geographically distributed, composed by different commercial partners, each of them with its own specialization and resources directed to specific functions in the product life cycle [2], [3].

In this context, new challenges are also imposed on traditional models of management and quality improvement, which should be able to encompass not only internal processes of a single company [4], but also extend to cases involving external interconnected companies [3], [5], [6]. In these new environments, in particular, the solution of nonconformance problems is characterized by knowledge intensive activities heavily based on experience, which in complex cases can go beyond the knowledge and experience of members of a single company. Given this scenario, this paper proposes an architecture in which agents are used to share and retrieve knowledge resulting from the solution of previous nonconformance problems, together with the Potential Failure Mode and Effects Analysis in Manufacturing and Assembly Processes (PFMEA) method. PFMEA is a quality engineering method that is thoroughly used in improvement processes, which contributes to gathering information on manufacturing. Thus, PFMEA could be used to provide valuable knowledge that could be shared among the different links that compose the manufacturing chain.

The developed distributed system uses an ontology based on description logics (DL) for the knowledge representation in the PFMEA domain. This system seeks to provide means to share and reuse current knowledge in PFMEA tables in support to the management of the organizational knowledge regarding processes in manufacturing environments with distributed resources.