Using the compromise programming model for optimal decision making in flood- management

Flood protection must be taken into account in almost all development projects. Owing to the fact that hazards associated with flooding (such as disruption of services, health impacts, famine and disease), the new approaches in flood risk management are needed. Therefore, it is essential that the selection of flood management option by cost-benefit viewpoint to be shifted to the selection of flood management option by considering the economic, technical, social, and environmental aspects. Due to the complexity of these challenges, water resources planners require a holistic, adaptive, incremental, and sustainable decision-making process, where multiple non-commensurate, competing, and often conflicting objectives (criteria) must be addressed and reconciled. During the past several decades, multi criteria decision making (MCDM), by its philosophical underpinning and the theory, methodology, and practice that have been developed on the basis of its holistic philosophy, has served as a harmonizing agent in technology, society, and policy. Thus, formidability of this method in addressing the flood management alternatives challenges is undeniable fact. In this paper, there are seven flood management measures in Gorganrood Watershed flood management project, including: conservation of natural condition, Golestan Reservoir management, levee construction, diversions-canal construction, flood forecasting and warning system and flood insurance. Prioritization of the alternatives is in demand of evaluation criteria. Each alternative would be prioritized based upon proposed MCDMs to investigate the most conclusive alternative. The flood management project is ranked based on eleven criteria, including: expected average number of casualties, recovery rate, gradual rate, expected annual damage, safety feeling, employment rate, public participation, landscape protection, wildlife habitat conservation, water quality conservation and technical feasibility and performance. These criteria have been classified into four main groups as social, economic, environmental and technical features. Compromise Programming model (CP) is employed to rank these seven alternatives which are compared with simple additive weighted model (SAW). SAW is known to be common method, which is accepted in many researches due to its simplicity. In this method, alternatives are assessed with respect to each criterion, whilst CP defines the best solution as the one in the set of efficient solutions whose point is at the least distance from an ideal point and the aim is to obtain a solution that is as close as possible to some ideal alternative. The results show that by increasing the value of P parameter, the CP model emphasizes on the importance of recovery rate, expected average number of casualties and safety feeling, which categorized in economic, social and technical feasibility, respectively. This result is obtained by assessing the changes in criteria weights while the P parameter increase from P=1 to P=∞. In the proposed model, integration of a flood warning system and flood insurance was chosen as the most conclusive alternative for flood hazard mitigation. This alternative is a combination of a pre and post disaster action. In contrast, SAW model is unable to highlight or discriminate between the criteria. One of the criticisms over MCDM states that different techniques may yield different results when applied to the same problem. An analyst looks for a solution that is closest to the ideal, in which alternatives are evaluated according to all established criteria. Therefore, it is necessary to compare the MCDMs plus assessment of subjectivity with sensitivity analysis of input data. In this study, sensitivity analysis was performed to examine the response of alternatives when the criteria weights changed to its minimum and maximum values. The results show that the CP model has less sensitivity to changes in criteria weights. Due to considering the ideal solution distance, CP model has more sensitivity to economic and technical criteria in comparison with environmental and social criteria. Although SAW model has less sensitivity to changes in criteria weights, it makes no priority between criteria even in sensitivity analysis. It is observed that the change of the MCDM methods produces differences in the final ranking of the alternatives. The discrepancy that appears between the rankings obtained by different MCDM methods, highly depended on the difference in their mathematical modeling while solving a decision problem. Thus, the main focus must lie on the selection of the most appropriate MCDM method to be adopted, not to mention that proper structuring of the decision problem, considering the relevant criteria and decision alternatives are imperative facts on this issue.