An ontology for Hazard Identification in Risk Management
Rebekah Gilmmour · The University of Queensland · 2004
This aim of this research project was to create a knowledge base for the area of hazard identification in risk management. Hazard identification is the focus as it is an important part of the process of risk management. Such a knowledge base is created for intelligent software agents who need a formal structure and language to be able to share and ‘reason’ with information. This knowledge base is represented in ontology form. This project involved research in the area of hazard identification and then the construction of an ontology to represent this knowledge. The various methods of hazard identification are the classes of this ontology. The ontology was constructed in a program called Protege 2000, which was developed by Stanford University. The ontology produced is an excellent result. The ontology allows for easy access to the hazard identification knowledge base. Querying examples show how intelligent agents would manipulate knowledge. The major limitation is the absence of explanation of method procedures.The following recommendations have been made for future work in this area-1. Extension of ontology across all areas of risk management.2. The publication of the HAZID ontology on the World Wide Web.3. Link with OntoCAPE.4. HAZID ontology could be extended to further detail classes.5. Development of some simple risk management agents to use the ontology6. Test the multi-agent system on a number of case-studiesIn conclusion, the hazard identification ontology was constructed successfully, and has established a knowledge base which can be published and used by intelligent software agents.