Simulating operator decision processes at Savannah River
- Westinghouse Electric Corp., Pittsburgh, PA (United States). Science and Technology Center
- Seer Systems, Inc., Pittsburgh, PA (United States)
- Westinghouse Savannah River Co., Aiken, SC (United States)
Cognitive Environment Simulation (CES) is both a methodology and an AI tool. As a methodology, it denotes a technique that models human operators' cognitive processes to either (1) aid in designing the interface to a complex system (such as nuclear reactor control room), or (2) assess the cognitive causality that affects the likelihood of human error in specific accident scenarios. As an AI tool, CES is an expert system that models human operators' reasoning and decision processes. In this application, both the methodology and the tool were focused on modeling human intention formation and errors in a problem-solving context. The CES tool consist of an inference engine and knowledge base that are object-oriented at a level of analysis to facilitate the modeling of human decision-making. While descended from the early AI successes of Internist and Caduceus in the arena of medical diagnosis (Pople, 1985), CES has been restructured and enhanced to deal with additional knowledge requirements encountered in real-time control of complex systems. This version of CES receives its input from a virtual display, a file of several hundred plant parameters whose values are sampled every five seconds. Analogously to a crew observing control room displays, CES reads the virtual display file and evaluates what it sees.'' CES' evaluation is based on the changes it observes in relation to its prior knowledge of operational goals, plant structure, event history, and operator procedures that are represented in its knowledge base. Its output is an English-like protocol of observations, explanations, and declarations of recommended action (intent) that it would take if it could. These last also represent actions that human operator(s) could take if they so decide. Through manipulation of its knowledge base, CES can also be caused to make mistakes for human-like reasons.
- Research Organization:
- Westinghouse Savannah River Co., Aiken, SC (United States)
- Sponsoring Organization:
- USDOE; USDOE, Washington, DC (United States)
- DOE Contract Number:
- AC09-89SR18035
- OSTI ID:
- 5780575
- Report Number(s):
- WSRC-MS-91-361; CONF-9111199-1-Extd.Abst.; ON: DE92009950
- Resource Relation:
- Conference: American Nuclear Society (ANS) international winter meeting, Pittsburg, PA (United States), 10-14 Nov 1991
- Country of Publication:
- United States
- Language:
- English
Similar Records
Simulating operator decision processes at SRL
Model-Based Calculations of the Probability of a Country's Nuclear Proliferation Decisions
Related Subjects
21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS
PRODUCTION REACTORS
REACTOR OPERATION
COMPUTERIZED SIMULATION
DECISION MAKING
HUMAN FACTORS
MAN-MACHINE SYSTEMS
REACTOR CONTROL SYSTEMS
REACTOR SAFETY
SAVANNAH RIVER PLANT
CONTROL SYSTEMS
NATIONAL ORGANIZATIONS
OPERATION
REACTORS
SAFETY
SIMULATION
US AEC
US DOE
US ERDA
US ORGANIZATIONS
220400* - Nuclear Reactor Technology- Control Systems
220600 - Nuclear Reactor Technology- Research
Test & Experimental Reactors