Semantic rule modelling for intelligent supplier discovery

Abstract

Formal representation of manufacturing capabilities is a critical requirement for autonomous deployment of agile supply chains in distributed environments. In this context, manufacturing ontologies play a key role by providing the required means for explicit knowledge representation. Manufacturing Service Description Language (MSDL) is an OWL (Web Ontology Language)-based ontology developed with the purpose of modelling manufacturing capability in a service-oriented framework. This paper presents the rule-based extension of MSDL that enhances the ontology semantically and enables advanced ontological reasoning. The particular focus of this paper is on analysing the reasoning and inference patterns used by human experts during the supplier discovery process and formally representing them, to the possible extent, using Semantic Web Rule Language (SWRL). Two categories of rules, namely, property inference and classification rules are introduced and implemented in this research. In order to capture the rules and evaluate the effects of the encoded rules on the performance of the MSDL search engine, an experimental approach is followed in this work. The obtained results support the hypothesis that semantic enrichment of the ontology through introducing semantic SWRL rules enhances the performance of the semantic search process.