چكيده به لاتين
Abstract:
Population growth, water pollution, shortage of fresh water resources, and the vulnerability of the water distribution network are the causes that many regions of the world are currently facing moderate to severe water crises in different parts of the world. Undoubtedly, water crises is one of the main potential threats and major concerns of managers in each country. Integrated water resources management approach can be a resolution to this problem as it addresses simultaneously economic, social and environmental benefits. In this paper, maintenance and repairs of resources and facilities against potential disruption including mechanical failures (failure of facilities) and hydraulic failures (for example unstable rainfall), and also the definition of new water resource development projects for the management of residual water resources are proposed. Due to possible disruptions, the network faces uncertainties in demand, capacity of dams and pipelines. Strategic decisions such as the construction of new refineries and pipelines to increase available water and tactical decisions including water allocation, resistance facilities before disruption or restoration lost capacities after that to enhance the resilience of the network are among the decisions are determined by the model. Two objective functions are proposed for this model in which cost of water allocation is defined as an economic function and the effect of large dams on employment is defined as a social function while environmental preferences is considered as a firm constraint. The ε-constraint method has been used to convert the multi-objective into single-objective problem. Then concept of two-stage stochastic programming approaches based on scenario is being used to deal with uncertainty in proposed model. The model has been coded in GAMS software by using CPLEX as an exact solution. The Verification of the model has been done by compromising the value of the stochastic solution (VSS) and the expected value of perfect information (EVPI) versus proposed robust optimization. The analyzes indicate that robust approach causes 4 miliard rial cost reduction in minimum condition and 37 miliard rial in maximum condition. Also by changing some parameters, the model's behavior have been examined, then the model accuracy have been proved. To measure the efficiency of the optimization model, a case study with the real data for the water distribution network of Tehran, Iran in the upstream level has been implemented. In the end, numerical results are presented for optimal decision making.
Keywords: Disruption, Resilience, Sustainable Development, Two-stage Stochastic Programming, Water Network.
