Contribution of the ratio scale of expert judgments in the analytic hierarchy process

Sáenz-Royo, Carlos (Universidad de Zaragoza) ; Chiclana, Francisco

Resumen: The pairwise comparison ratio scale that the Analytic Hierarchy Process (AHP) acquires a central role because it allows the incorporation of intangible information through expert judgments. Existing literature has been shown that preference results are sensitive to the scale used in ratio judgments. This paper applies the Intentional Bounded Rationality Methodology (IBRM) to study the performance of the most used AHP ratio scales in the literature. By adapting the AHP ratio scales to the IBRM, several scenarios are proposed based on the distribution of the latent performance of the alternatives for two different problems. In problem 1, it is assumed that the AHP comparison judgements are numerical and known to the expert, while in problem 2, the AHP comparison judgements are described linguistically without the expert being aware of their transformation into numerical values. Problem 1 is used to answer the following research question (1) which among a set of seven different ratio scales used in literature favor AHP expected performance?, while Problem 2 is used to answer the following research question (2) how much the expected performance in AHP deteriorates when the expert is only guided by verbal judgments? The expected performance of each of the considered ratio scales is obtained in each scenario for different levels of expertise of a decision-maker. For the first problem, the balanced scale and the power scale show the best and worst expected performances, respectively. Problem 2 compares these two scales and the results show their performance differences are below 8%, which is interpreted as a stability property of AHP with respect to scale changes. Finally, it is also shown that no matter which ratio scale is used, the requirement of the consistency property in AHP contributes positively to the expected performance.

Idioma: Inglés

DOI: 10.1016/j.eswa.2025.128938

Año: 2026

Publicado en: Expert Systems with Applications 296 (2026), 128938 [11 pp.]
ISSN: 0957-4174

Financiación: info:eu-repo/grantAgreement/ES/AEI/PID2020-113338RB-I00
Financiación: info:eu-repo/grantAgreement/ES/MINECO/ECO2017-86305-C4-3-R

Tipo y forma: Artículo (Versión definitiva)

Área (Departamento): Área Organización de Empresas (Dpto. Direcc.Organiza.Empresas)