Multiobjective Optimization Approach for Preference-Disaggregation Analysis Under Effects of Intensity
A widely used approach in Multicriteria Decision Aid is the Preference Disaggregation Analysis (or PDA). This is an indirect approach used to characterize the decision process of a Decision Maker (or DM). By means of a limited set of examples (called a reference set) provided by the DM, the PDA approach estimates the parameter values of a preference model that is characterized by the DM. This paper proposes a new optimization model for PDA, and its solution through an evolutionary algorithm. The novel features in the definition of the model include the use of the effect of the intensity (i.e. the variations among the criteria values used to evaluate decision alternatives), and new ways to combine the number of consistencies and inconsistencies with respect to the reference set. Through an experimental design performed to evaluate the fitness of the new model, it was corroborated its effectiveness to fit the DM preferences, and also it showed comparable results with that provided by an state-of-the-art strategy.
KeywordsDecision Maker Optimization Model Pareto Front Strict Preference Fuzzy Preference Relation
This research was partially funded by the following projects: the project 3058-Optimización de Problemas Complejos of the Programa de Cátedras CONACyT.
- 6.Fernandez, E., Navarro, J., Mazcorro, G.: Evolutionary multi-objective optimization for inferring outranking model’s parameters under scarce reference information and effects of reinforced preference. Found. Comput. Decis. Sci. 37, 163–197 (2012)Google Scholar
- 7.Covantes, E., Fernández, E., Navarro, J.: Robustness analysis of a moea-based elicitation method for outranking model parameters. In: CCE, pp. 209–214 (2013)Google Scholar
- 11.Cruz, L., Fernandez, E., Gomez, C., Rivera, G., Perez, F.: Many-objective portfolio optimization of interdependent projects with a priori incorporation of decision-maker preferences. Appl. Math, (8)Google Scholar
- 12.Bastiani Medina, S., Cruz Reyes, L., Fernandez, E., Gómez Santillán, C., Rivera Zarate, G.: An ant colony algorithm for solving the selection portfolio problem, using a quality-assessment model for portfolios of projects expressed by a priority ranking. In: Design of Intelligent Systems Based on Fuzzy Logic, Neural Networks and Nature-Inspired Optimization, pp. 357–373 (2015)Google Scholar