A time-dependent decision support system for multi-attribute decision-making

Article type: Research Article

Authors: Ríos-Insua, Sixto | Jiménez, Antonio | Mateos, Alfonso

Affiliations: Department of Artificial Intelligence, School of Computer Science, Madrid Technical University, Campus de Montegancedo s/n, 28660 Madrid, Spain. Tel.: +34 91 3367437; Fax: +34 91 3524819; E-mail: [email protected]

Abstract: We describe a decision support system including multiple conflicting objectives and with consequences spanning several time periods. The system is supported by an imprecise multiattribute additive utility model, which includes the possibility of different discounting methods, aimed at identifying the optimal strategy by means of interactive multiobjective simulated annealing. The system can be useful for solving problems, including, if necessary, decisions with imprecise consequences over different time periods, which are modelled by means of an objective hierarchy. This objective hierarchy considers appropriate preferences based on judgemental inputs for assessing imprecise individual utility functions and imprecise scaling factors on objectives and attributes, which are aggregated into a scalar or vectorial additive utility function to evaluate the available alternative decisions. The system also includes different sensitivity analyses to check the robustness of the conclusions. Finally, an application to the evaluation of remedial strategies for restoring contaminated aquatic ecosystems illustrates the usefulness and flexibility of this decision support tool.

DOI: 10.3233/ICA-2004-11105

Journal: Integrated Computer-Aided Engineering, vol. 11, no. 1, pp. 63-75, 2004