Most urban areas are facing an ever increasing demand for fresh water due to population and industrial growth. The paper presents a network programming model to evaluate alternatives for supplying water from various sources-treated fresh water, desalinated sea water and renovated waste water-to satisfy future demands for multiple water use in coastal cities at minimum cost. Water demand projections for municipal, industrial, agricultural and recreational use are specified in terms of their geographical locations, quantity and quality specifications for a future data. Costs of water to match these demands from different sources any by desalination and renovation are assumed known. The network programming computer model can accommodate non-linear costs of water development, treatment and conveyance and uses an iterative technique to obtain a minimum cost solution. The network model described is quite general in the sense that it can be applied to other areas such as waste disposal and water pollution control, solid waste management, air pollution control, pipeline optimization and transportation studies.