DFL-based hybrid integration of ontology and rule for product development engineering

Abstract

In order to support decision makings in product development, both product data knowledge base and product design knowledge base are necessary. As a product data modeling language, DL-based ontology language has been considered suitable in the collaborative environment. From past, logic programming languages has been applied to build rule-based system capturing design expertise of product development. However, most of the previous researches have focused either on building product data model, using ontology language, or on building rule-based expert system to represent product design knowledge. A few researches on product development have applied the integrated approach of ontology and rule just using OWL/SWRL but no one performed the analysis of expressivity requirements for Product Data Model Ontology (PDMO) and Product Design Rule Base (PDRB). Thus, this paper investigates the necessity of combining ontology and rule for PDMO and PDRB. Furthermore, this paper derives the requirements that PDMO needs OWL 1.1 expressivity and constraint checking mechanism which works under CWA. For PDRB, it is realized that non-logical features of LP and Description Frame Logic (DFL) integration approach are essential. In order to reflect the analysis, we introduce an integrated knowledge representation framework. In the framework, based on DFL approach, we combine OWL 1.1 as an ontology language, Prolog as a rule language and Constraint for Description Logic which works under CWA. In addition, we describe the architecture of the proposed approach, enhanced expressiveness, and reasoning interactions between DL reasoner, Prolog engine, and constraint checking. Finally, we illustrate a detailed case study on process planning from process selection to cutting tool condition selection to show how PDMO and PDRB are represented in the framework and how reasoning results are inferred from them.