An Intelligent Hybrid Decision Support System for the Management of Water Resources

During the last two decades the scarcity of water resources is exponentially growing worldwide and especially in semiarid countries like coastal and island communities of the Mediterranean region. Cyprus is one of the countries in the Mediterranean where the water shortage problem is evident and very severe. This paper describes an intelligent and flexible, hybrid decision support system, for the management of water resources in the regions along the South Conveyor Project, the largest water development project in Cyprus. In particular, the paper presents the integrated system and its architecture, together with detailed descriptions of the hybrid components synthesizing it. Introduction During the last twenty-five years the scarcity of water resources is exponentially growing worldwide. As a result, there has been a serious and growing concern about the water shortage problem, resulting in substantial progress in different aspects of water resource development and management in different parts of the world (Lintner 1996, Rahman 1998). The problem of the scarcity of water is especially severe in the semi-arid countries and in particular, in the coastal and island communities of the Mediterranean region, where water demand satisfaction is not possible (World Commission on Water 1999). Cyprus, situated in the northeast part of the Mediterranean region, is among the countries facing a severe water shortage problem (Hendawi 1998) and focusing in finding efficient and systematic ways for effective water management (Tsiourtis and Kindler 1995, Christodoulou and Socratous 1995). Considering the constant reduction in the water supply available, extraordinary efforts are needed to satisfy the demand for water, which increases proportionally with development. This task becomes even harder, due to the lack of a global view of the water situation in Cyprus, the nonexistent long term planning for its management and the utilization of ad-hoc approaches based solely on the Copyright © 2002, American Association for Artificial Intelligence (www.aaai.org). All rights reserved. expertise of engineers and not on scientific facts. Such approaches result in unduly conservative or over-optimistic utilization of water, which may either lead to underutilization or to large deficits of water. Consequently, there is a need for a methodological approach for supporting decision-making. Such a methodological approach has been realized in terms of an interdisciplinary, intelligent and flexible, decision support system for the management of water resources in the regions along the South Conveyor Project (SCP), the largest water development project in Cyprus (Water Development Department 1983). The proposed system aims to support water managers face effectively the drought, and hence to minimize the water restrictions imposed on consumers and generally to use the water supply infrastructure effectively. The final system will be installed at the Water Development Department (WDD) in Cyprus. Its benefits are expected to be highly significant, not only for Cyprus but also for other countries with similar problems. The architecture and the associated methodology can be applied not only to water management problems but also to other supply and demand problems of similar content. The development of the proposed decision support system started under the INCO-DC project MEDWATER, funded by the European Union and is now continuing under project NHREYS, funded by the Cyprus Research Foundation. These projects are examples of a number of research projects in the area of decision making for water management. Other noteworthing examples, which in some aspects are closely related to the decision support system described in this paper, are the EUREKA project (Eureka Project EU487 1990) that generated the WaterWare system (Environmental Software and Services 1998), providing an integrated framework that provides easy access to advanced tools of data analysis, simulation modeling, rulebased assessment and multi-criteria decision support and a FAO project for the river Nile (Georgakakos 1997) developed in USA. The rest of the paper describes the system and the architecture underlying it, elaborates on the individual system components, describes the state of the evaluation of the system and outlines future work. 272 FLAIRS 2002 From: FLAIRS-02 Proceedings. Copyright © 2002, AAAI (www.aaai.org). All rights reserved.