Knowledge organization in intelligent tutoring systems for diagnostic problem solving in complex dynamic domains

Turbinia-Vyasa is a computer-based instructional system that trains operators to troubleshoot and diagnose faults in marine power plants. The simulator, Turbinia, is based on a hierarchical representation of subsystems, components, and primitives. Vyasa is the computer-based tutor that teaches the troubleshooting task using Turbinia. The simulator, an interactive, direct manipulation interface, and the tutor (with its expert, student, and instructional modules) comprise the instructional system. To be effective, knowledge about the system and the troubleshooting task, together with the knowledge to infer student’s misconceptions from observed actions and pedagogical knowledge, must be suitably organized and presented to the student. System knowledge is organized in terms of schematics, subsystems, and fluid paths, which are further decomposed into structure, function, and behavior. Information about failures and failure schemas complement the system knowledge. In this paper, we discuss the details of knowledge organization and show how they support the functions of the intelligent tutoring system.