Advancing the use of MACBETH in health settings: A new decision support system to assist moving from qualitative judgments towards numerical value functions

Miguel Joaquim Roque Fernandes

Key information

Authors:

Miguel Joaquim Roque Fernandes (Miguel Joaquim Roque Fernandes)

Supervisors:

João Carlos Nunes Bana e Costa; Mónica Duarte Correia de Oliveira (Mónica Duarte Correia de Oliveira)

Published in

06/18/2024

Abstract

Standing for Measuring Attractiveness by a Categorical Based Evaluation Technique, MACBETH has been widely used in complex evaluation contexts to construct quantitative value models based on qualitative judgements of difference in value elicited from a panel of evaluators. When the qualitative judgements are expressed by each evaluator separately, with no group discussion to reconcile their judgements, decision analysts face the problem of how to derive a collective value function from the evaluators’ individual judgements. This work proposes an original technique, implemented in Python, to address this issue under a MACBETH multicriteria framework. For each evaluation criterion, the proposed collective value function respects the majority rule and assumes a collective constant trade-off attitude. A second research topic focuses on the reconciliation of value functions from different decision makers. Thirdly, this work also examines the impact of variability in value functions created from incomplete matrices of judgments. For each challenge, a module was developed in a Web-based Decision Support System and tested for a case.