Reliability Assessment for Water Supply Systems under Uncertainties

Abstract

Reliability assessment of water supply systems (WSSs) is an important aspect of WSS planning and operations. Traditionally, WSS reliability involves the comparison of hydraulics (e.g., pressure or available water volume) and water quality (e.g., residual chlorine) parameters with their desired minimum level of service under various emergency loading conditions. To compute hydraulic dependent parameters (e.g., pressures, flow), different algorithms solve continuity and energy equations expressed in terms of certain independent parameters (e.g., roughness parameters and nodal demands) with certainty. Similarly, transport equations expressed in terms of different quality parameters are solved deterministically to compute water quality dependent parameters (e.g., residual chlorine). However, it is extremely challenging, even impossible, to estimate network-independent parameters with certainty. Therefore, estimated dependent parameters based on semiquantitative information bear uncertainty, which leads to a question of how reliable current reliability assessment results are. To address this issue, this paper proposes a new means of reliability assessment in which the reliability of a WSS is expressed in terms of utility and associated beliefs. The methodology developed provides an estimate of the uncertainties inherent in traditional reliability assessment results. It is expected that the proposed methodology will help municipalities to make informed decisions in order to increase the safety and security of public health.