(182t) An Ontology-Based Automated Generation of Training Scenarios: Development of Process Safety Rule Engine

The preconditions of the usefulness of an operator safety training model in large plants are the versatility and accuracy of operational procedures, obtained by detailed analysis of the various types of risks associated with the operation, and the systematic representation of knowledge.

Existing scenario-based training system were not flexible, it can be focused on implantable memories due to they are limited, and the efficiency of knowledge acquisition is low owing to uniformity process and no differences trainees. Therefore, automatic generation of training scenarios provides a variety of training situations and opportunities when mounted on a VR-based training system, and can provide optimal step-by-step intelligent training tailored needs and abilities of individual trainees. This requires the development of events in the training domain, simulation and the representation of knowledge about goals and behaviors. Ontologies enable coherent processes, devices, operations, reasoning and modeling, sharing and reuse of knowledge.

In this study, we use the ISO 15926 an international standard for plant data integration, management and the process ontology of OntoCAPE to generate a knowledge node based on the HAZOP analysis of the RDS process of the petrochemical plant. Ontology on the results of causation for 22 items of action items due to process safety and operational impact. And automatic generation of safety training scenarios is implemented by using inference modeling of SWRL (Semantic Web Rule Language) based on OWL-DL. It is expected that it will be possible to develop a swift automatic generation system by increasing the recyclability of scenario inference engine compared with the automatic generation based on XML extraction + HTN planning of the proposed ontology.