Aplicações de modelos multicriteriais hierárquicos e multi decisores para alocação de água no Sistema Curema-Açu.

Abstract

The water demands increase everyday and may generate situations of conflicts regarding to the allocation of this precious and scarce liquid among its users. In cases where the water demand is superior its availability, like the Nort heast region of Brazil, there is the need of a careful planning of water allocation, through the water rights instrument, to minimize potential water users’ conflicts. In order to achieve that, social, economic and environmental aspects, structured in a h ierarchical and systemic way, as well as multi decision makers, should be taken into account. On the other hand, there are no softwares that perform a multicriteria analysis taking into account a hierarchical structure, multi decision makers and the possi bility of including uncertainties in their judgment with respect to the relative importance of criteria. Therefore three methods for transforming a hierarchic problem into a non hierarchic problem and three methods for multi decision makers analysis are presented and analyzed in this work. These methodologies were applied at two study cases: a hierarchically structured fictitious one with 4 users, 9 criteria and 7 decision makers, and to a study of potential water rights grants priorities to be provided to users of river reach 2 of Curema -Açu system, in Paraiba state. The water supplied by the Curema-Mãe D’Água reservoir, in this system, is source of conflicts among water users of Paraiba and Rio Grande do Norte states. The VIP Analysis, a multicriteria decision support system that has characteristics such as additive property and possibility of considering uncertainties in criteria weights, was used. The analyses were performed for scenarios where criteria weights were well defined or only their ran king was considered. The results have shown that different hierarchical methods produced different results, while different multi decision makers methods produced equal results. The indicated hierarchic analysis, given the additive property of VIP Analys is, is the one where the decision makers attributed weights to sub criteria were generated from their multiplication bottom up.