Interface agent for process plant operation-toward the next generation interface

Agent technology has been illuminated these days in many fields including manufacturing. We describe the notion of an interface agent for process plant operation starting from task analysis of plant operation and generic characteristics of agency, then introduce an idea of case-based realization for an agent's perception and memory. Sensation of anomaly was built on top of a memory organization package that was developed by Schank et al. (1994) Clustering of plant states by version space enabled us to discriminate between a plant situation and another, and allowed operators to label them through teaching. A feasibility study using real data of a plant showed the competence of sensation for anomaly detection, and the demonstration on connecting to a plant simulator showed the usefulness of this cognitive interface agent in a malfunction scenario.