Abstract
•A new Genetic Algorithm based Flexibility Optimization has been developed.•The algorithm was applied to the design of utility distribution systems.•The new GAFO was tested and showed capital cost savings of 14 to 72%.
Urban sprawl, climate change, and resource scarcity will impact how sustainable cities approach future challenges surrounding the management of water resources. A major component of all urban water systems is distribution, which constitutes approximately 80-85% of the total cost of a water-supply system. Traditionally, water distribution systems (WDS) are designed using the ‘worst scenario,’ or ‘robustness’ to improve system reliability. Deterministic assumptions are historically inaccurate and require a new design approach that recognizes uncertainties and offers more adaptability. A Genetic Algorithm based Flexibility Optimization (GAFO) model is developed in Visual C++ and linked with EPANET for the design of WDS that are more adaptable. Unlike traditional GA optimization, GAFO involves a dynamic decision-making process that recognizes a range of possible future conditions and maximizes the flexibility of a WDS at the lowest cost. The outcome is a WDS that can follow different future trajectories (changing conditions) and generate a staged implementation strategy that allows a stepwise evolution of the WDS over time. The GAFO model was tested on several hypothetical cases and was found to perform well in terms of convergence and flexibility. Compared with conventional, non-flexible designs, cost savings in the range of 35% to 72% were realized.
