-

This paper presents a new approach for reliability-based optimization of water distribution networks. The approach includes three sub-models (a steady state simulation model, a reliability model and an optimization model). In this method, nodal and system reliabilities in critical conditions (such as excessive of demands, shortfall in minimum desirable nodal pressures, pipes failure, etc.), are considered as a constraint. Reliability is defined as the ratio of the available outflow to the total demand considering probability of any mechanical and hydraulic failures. The innovation of the algorithm is in the use of a pure head driven simulation model to simulate pressure deficient conditions, properly. This model relates the available outflows to the available nodal pressures, which produces available nodal outflows efficiently. These values are the best input for the reliability model. The optimization model uses external penalty and gradient functions to obtain the least-cost configuration of the network. Finally by presenting two test networks, the efficiency of the method is illustrated.