Comparison between sensibility analysis methods, used in the decision-making process through the multicriteria evaluation

Multicriteria Evaluation is one of the decision-making processes used in different areas of knowledge. Sensitivity analysis determines robustness of the strategies proposed in the decision-making process, as a consequence of identifying the uncertainties caused by their components: criteria, weights and priorities. In this context, the general purpose of this study was to compare some sensitivity analysis methods, aiming at the validation of the decision-making processes, especially the Ordered Weighted Averaging method, developed in a Geographical Information System (GIS) environment. The following methods were compared: Exploratory, Monte Carlo Simulation, and one ready to use in AHP-OWA module (based on stochastic simulation). In order to perform such comparison we used a decision-making process, with the purpose to defining priority areas for forest conservation. The Exploratory method permitted the development of all decision-making stages in a GIS environment. On the other hand, the AHP-OWA method module allowed a graphic analysis of the criteria influence on the local priority. This analysis contributes to decrease the subjectivity associate with the definition of the criterion order weights. Regarding the Monte Carlo simulation, one can highlight the decrease of the analyst influence on the setting of order weights. The three methods promoted a satisfactory validation of the decision-making process, presenting advantages and disadvantages and, mainly, they contributed to get a pattern of the sensitivity analysis, which can be used in the processes developed with Multicriteria Evaluation.