Monte-carlo Simulation Techniques in a Multi-attribute Decision Support System

This paper presents a Multi-Attribute Decision Support System aimed at aiding decision-makers in identifying optimal alternatives in complex decision-making problems. The system is based on a multi-attribute additive value model and admits imprecise assignments concerning weights and utilities and uncertainty about the multi-attribute alternatives. Different sensitivity analyses are possible over the inputs permitting the users to test the robustness of the alternative ranking to gain insight on the final solution. Specifically, Monte-Carlo simulation techniques are performed, which allows simultaneous changes on weights and whose results can be statistically analyzed. An application to the restoration of a radionuclide contaminated aquatic ecosystem is illustrated throughout the paper.