The synthesis of the AHP as a well-posed mathematical problem and matrix norms appropriate for sensitivity analysis via condition number

Gustavo Benitez Alvarez, Rafael Guimarães de Almeida, Cecília Toledo Hernández and Patrícia Alves Pereira de Sousa

International Journal of Mathematics in Operational Research, 2025, vol. 30, issue 1, 111-134

Abstract: The analytic hierarchy process (AHP) is a decision-making method, which has as its greatest criticism the rank reversal effect. This paper formulates the fourth step of the AHP (synthesis) as a 'well-posed' mathematical problem. A theorem guarantees the existence of the square condensed original formulation for the AHP. This means that any decision problem modelled by AHP with a different number of alternatives and criteria can be condensed into a model with an equal number of alternatives and criteria without loss of condensed information. This condensed formulation can be better conditioned than the original rectangular formulation of the AHP. The square condensed equivalent formulation is also a 'well-posed' mathematical problem. The concepts are applied to two practical cases from the literature, and sensitivity analysis is performed. Four classical matrix norms are reformulated to obtain theoretical bounds for the error estimate closer to actual error.

Keywords: multiple criteria analysis; rank reversal; linear systems of equations; sensitivity analysis. (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.inderscience.com/link.php?id=144549 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ids:ijmore:v:30:y:2025:i:1:p:111-134

Access Statistics for this article

More articles in International Journal of Mathematics in Operational Research from Inderscience Enterprises Ltd
Bibliographic data for series maintained by Sarah Parker ().