The value of expert judgments in Decision Support Systems
Date
Advisors
Journal Title
Journal ISSN
ISSN
Volume Title
Publisher
Type
Peer reviewed
Abstract
It is a challenge to improve a decision support system (DSS) based on expert judgments; the literature proposes to improve accuracy and performance by increasing the sophistication and complexity of the DSS, but at what cost? This study presents a model for encoding a DSS based on expert judgments and evaluating its efficiency, establishing a three-part analysis structure: information requirements (number of judgments), quality requirements (quality assurance mechanisms), and algorithmic complexity. With a focus on the cost of judgments, a systematic and quantitative coding of the performance and cost in each part of the DSS is established. A “break-even point” efficiency measure, defined as the maximum percentage of the optimal performance that can be paid per unit of resources, is proposed to ensure that the use of the DSS remains profitable. Counterintuitively, the results of a case study show that the efficiency of DSSs does not necessarily increase with respect to the informativeness level of DSSs. Overall, this study provides a new method for evaluating the efficiency of DSSs.