SCOPE: a blackboard model-based decision support system for crude-oil trading

Abstract

A complex decision problem dealing with crude-oil trading is hierarchically decomposed into multiple subproblems. A blackboard architecture for a Decision Support System (DSS) called SCOPE is developed in which the subproblems generated during the initial decision analysis can be mapped to relevant knowledge sources incorporating the most appropriate techniques for each. A prototype implementation of SCOPE-0, an expert system application using Knowledge Engineering Environment (KEE) for supporting the most critical subproblem—crude-oil evaluation—is presented. Some of the lessons learned from this research project, methodological issues, and promising future directions for DSS design are outlined. Copyright © 1999 John Wiley & Sons, Ltd.