Abstract
With increasing global change pressures (urbanization, climate change etc.) coupled with existing un-sustainability factors and risks inherent to conventional urban water management, cities of the future will experience difficulties in efficient decision making on the infrastructure development. Projections of future global change pressures are plagued with uncertainties which cause difficulties when developing urban water infrastructures that are insensitive to these global change uncertainties. In this paper a methodology is presented that generates optimal urban water networks that are adaptable and sustainable under future global change pressures. These flexible systems are characterized by their ability to cope with uncertainties and have the capability to adapt to new, different, or changing requirements. The flexible design tool presented in this paper consists of two major components. The first component is a methodology for developing scenario trees that reflect uncertainties associated with future demand for water. These scenario trees represent the uncertainty envelope associated with demand projections over time. The second component is an optimization model that considers the phased design of the water network, taking into account the likeliness of different demand scenarios over time (as expressed by the scenario trees). The GA based optimization model identifies the optimal staged development of the network that gives the optimal expected value of the network both in terms of costs and benefits. The flexible design tool is then applied to the design of an example network with a design horizon of 30 year. The solution is presented as a phased design in 5 year stages and is compared with a design undertaken in the traditional way. This comparison clearly highlights the benefits and the efficacy of applying flexible design approaches for water systems operating under future uncertainties.