Knowledge management in chemical process industry.

Information and knowledge are among the major resources in chemical process enterprise. Effective knowledge sharing and decision coordination are important to collaborative product development and integrated manufacturing. The integration of knowledge management in chemical process industry can provide the enterprise an environment for knowledge sharing and coordinate decision-marking, it can also help the enterprise to realize the best value of its knowledge assets and make businesses more competitive and profitable. In this work, an Ontology-based knowledge management system is proposed for knowledge integration and decision support in chemical process industry. Information technology, artificial intelligence and chemical engineering domain technology are integrated into a unified system to support knowledge integration, cooperate manufacturing, enterprise management and information service in chemical process industry. The system infrastructure includes Ontologies, knowledge repository, information retrieving agent, knowledge discovery tools and user interface. Ontology plays an important role in the knowledge management system for knowledge integration, knowledge sharing and reuse. Ontology classifies the knowledge base, integrates sources of knowledge into the knowledge repository, supervise database and user interface construction, and severs as a backbone of the knowledge management system development. A flexible and systematic approach for ontology development and implementation is established in this work to support ontology creation and application in the knowledge management system. Knowledge retrieving services are developed in the knowledge management system to extract information and knowledge from various data sources. Information retrieving agents retrieve information from the knowledge repository according to the user's requirement, and provide cleaned information through information filtering. Ontology-based information retrieving approach is utilized in this work. Data mining technique is applied to extract the implicit and potentially useful information, and also predict trends by mining the historic data. Knowledge management in chemical process industry consists of a set of practices aimed at monitoring the process operation and providing decision support for the engineers and managers. However, currently available computer-aided systems for chemical process engineering are normally isolated, which make it difficult for data and information exchange and decision support. Multi-agent system is utilized in this work to coordinate these tasks and incorporate the disparate information resources. Process simulation, rule- base decision support, artificial intelligence such as artificial neural network (ANN) are integrated in this system for process analysis, data processing, process monitoring and diagnosis, process performance prediction and operation suggestion. A multi-agent system developed on the basis of JADE (Java Agent Development Framework) is integrated in the knowledge management system, in which software agents are designed to perform the tasks of process monitoring, process performance prediction, manufacturing management and information service. With a common communication language and shared ontologies, agents can communicate and cooperate with each other to exchange and share information, and achieve timely decisions in dealing with various enterprise scenarios. The implementation of knowledge management system will provide well-organized information for technical monitoring in chemical process industry, and enable the knowledge integration and sharing among researchers, engineers and managers. The application of the knowledge management system in chemical process industry can also help the engineers to coordinate in manufacturing execution, and provide decision support based on up-to-date information and knowledge