Efficient supply chain management (SCM) requires consistent exchange and sharing of information semantics, which is often hindered by semantic clashes among heterogeneous applications. This paper proposes an ontology-based architecture for addressing the problem of semantic integration. A supply chain ontology (SCO) is developed by using the skeletal method to capture concepts and relationships common to SCM. Moreover, it is formally defined in OWL DL, a web-based ontology language, to serve as an interlingua for an architecture enabling information integration of SCM. Syntactic transformation between XML Schema and OWL in the architecture is implemented by syntactic translators that are developed by using XSLT. Furthermore, a rule-based approach is presented to map semantically similar terminologies between SCO and application ontologies with the aim of enabling semantic interoperability among applications in supply chains. The mapping rules are represented by SWRL, a rule description language that is independent of the rule languages specific to inference engines. In addition, a case study of a printer supply chain is illustrated to demonstrate the proposed methodology. Finally, a prototype system is developed using the Java platform to implement the SCO-based approach to semantic integration of SCM.