Abstract. Rapid formation and optimization of manufacturing production networks (MPN) requires manufacturing service capability (MSC) information of each party be accessible, understandable, and processible by all others in the network. However, at the present time, MSC information is typically encoded according to local proprietary models, and thus is not interoperable. Related existing works are primarily for integration in "isolated automation" of pair-wise or small size networks and thus are not adequate to deal with the high degree of diversity, dynamics, and scales typical for a MPN. In this paper, we propose a model development framework which enables to evolve a reference model for MSC information based on the inputs from proprietary models. The developed reference model can serve as a unified semantic basis supporting interoperability of MSC information across these local proprietary models. Methodology for resolving structural and other semantic conflicts between deferent models in model development is also presented.Keywords: manufacturing service capability, ontology development, patternbased ontology transformation, canonicalization.
IntroductionToday, service capability information of manufacturers is typically represented according to some models developed by individual enterprises or communities. These local proprietary MSC information models are not interoperable because of their differences in service category, capability structure and values. As a result, manufacturers often have difficulty in quickly discover suppliers with required capabilities without a significant level of human involvement. A MSC information reference model that is semantically rich can help reconcile semantic difference among local proprietary models and increase access and precision to capability information. However, related existing works [1,2,3] are primarily for integration in isolated automation of pair-wise or small size networks with less semantic diversity and thus are not adequate to deal with the high degree of diversity, dynamics, and scales typical for a manufacturing production networks (MPN).