Knowledge-based product service configuration in servitization

To realize knowledge-intensive product service configuration in servitization, based on the extraction technique of ontology and neural networks, a complete product service configuration method was proposed from the view of knowledge representation, acquisition and reasoning. In this method, structural knowledge and rule knowledge were represented by Web ontology language and semantic Web rule language respectively. An improved skeleton method was proposed to acquire structural knowledge from expert experience, and an approach combining Local Cluster (LC) neural network and rule extraction algorithm was presented to extract rule knowledge from historical data. Configuration reasoning processes were performed with the support of Java Expert System Shell (JESS). The effectiveness of proposed method was validated by services of building control product.