Authors: P.L. Kunsch
Addresses: CoDE-SMG, Universite Libre de Bruxelles, Boulevard du Triomphe CP 210/01, BE-1050 Brussels, Belguim
Abstract: This article presents a simple and robust multi-criteria procedure providing the set of compromise rankings of alternatives, for one or many decision-makers, given some preference relationships between the criteria. The procedure is derived from the properties of the set of rank, or score vectors of the alternatives. The Pearson correlation serves as a measure of distance between two such vectors. The identification of all possible compromise rank vectors is made by a Monte Carlo analysis, using random weights. The probabilities of alternatives occupying the ranks are provided. The procedure is immune against rank reversals.
Keywords: multicriteria decision making; MCDM; statistical procedures; stochastic preferences; rank reversals; correlation distance; rank vectors; score vectors; compromise rank vectors; preference relationships; random weights; rank matrix; maximum number of criteria.
International Journal of Multicriteria Decision Making, 2010 Vol.1 No.1, pp.49 - 73
Published online: 17 Jun 2010 *Full-text access for editors Access for subscribers Purchase this article Comment on this article