Urban water supply system optimization and planning: Bi-objective optimization and system dynamics methods

Abstract

Sustainable, low-cost, environmentally conscious development is a major challenge for urban water supply systems. This study considers household and industrial water demand and establishes a bi-objective model to balance the economic and environmental cost of a three-stage urban water supply system including water production, distribution, and treatment processes. Considering different attitudes toward the increasing imbalance between water demand and supply, three hypothetical scenarios are designed based on which three sets of data on water demand and wastewater production volumes are defined. Then, a two-step method is applied to obtain solutions to decision variables (potable water use rate and wastewater untreated rate). After optimization, a system dynamics model is applied to test water supply system performance under short-, medium-, and long-term planning scenarios. Finally, the model is applied to Singapore to test its feasibility. The proposed bi-objective model provides managerial insights into the use of alternative water resources. For example, in 2020, water demand is simulated as 5.63×108m3. After optimization using the proposed model, the quantity of alternative water is 2.5×108m3 (water from the local catchment), 2.5×108m3 (imported water), 5.31×107m3 (desalinated water), 4.05×107m3 (NEWater for indirect potable use), and 8.10×107m3 (NEWater for direct non-potable use). Total water supply is 6.75×108m3, which means there is some redundancy in water supply. Further, three indexes—supply-demand ratio, economic cost, and environmental cost—help to identify the suitable planning period: the medium-term focus is the most appropriate option as it promises water adequacy for over 15 years and has lower energy costs and less negative environmental impact during the evaluation years. In addition, this study proves the influence of decision makers’ preferences on contradictory objective functions (economic cost minimization and environmental negative impact minimization). Overall, the proposed model balances a trade-off between the risk of an increasing future supply-demand imbalance and current water adequacy. Lastly, the proposed model offers a paradigm to identify a sustainable planning period.