QVAL and GENTREE: Two Approaches to Problem Structuring in Decision Aids
Abstract
Decision analysis is a technology which offers the decision maker a logical structure for solving a specific problem. Through the process of decision analysis, the decision maker may select from several options the one that best suits his/her needs. In practical applications, most decision analytic techniques require a specially trained decision analyst and a sizeable amount of time and expense. This report describes two computer-assisted decision aids which are designed to perform preliminary decision analyses in the absence of a professional decision analyst and for a restricted set of decision problems. The first decision aid, QVAL (Quick Evaluation), has been implemented on an IBM 5110 portable computer and is ready for operation on an experimental basis. QVAL is an interactive program designed to probe the user's memory for a complete set of attributes together with the scores and weights for a consistent evaluation of a set of options. The second decision aid, GenTree (Generic Tree Structuring), is still in the planning stage, but promises to be very valuable in storing experience and knowledge gained while working on problems and applying it to guide the analysis of subsequent ones. By utilizing prior knowledge stored in the GenTree database, this decision aid would approximate the performance of a decision analyst on familiar problem types. In contrast, QVAL would more resemble the performance of the decision analyst on a completely new application area. With the continuing progress being made in the fields of artificial intelligence and interactive computer graphics, more sophisticated computer- assisted decision aids of this nature may be designed.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 01, 1980
- Accession Number
- ADA086059
Entities
People
- Jonathan J. Weiss