Modelo para o combate a crises de abastecimento urbano de água

Abstract

The objective of this research is to contribute to the planning for facing crisis in urban water supply (CUWS). To achieving this goal, a combination of artificial intelligence techniques and Multiple Criteria Decision Aid (MCDA) was proposed. The development of computational tools (software) to assist the process of decision making and their application to a case study are parts of the scope of the research proposal. The following results were obtained in this research: (i) the knowledge of possible design guidelines for solving CUWS cases; (ii) the knowledge of factors influencing CUWS; (iii) the development of a hybrid model of decision analysis, called CUWS-SAD, based on the principles of vector similarity, knowledge-based systems (expert systems) and on the use of MCDA. The quality assessment of the developed models indicated concordance index of 56%, and that improvements in the models can be obtained through enhancement in the quality and quantity of the input data. The conclusions of the research indicate that the proposed approach for the problem treatment can be considered adequate. However, problems to attain knowledge of human experts suggest further investigations.