Comparing risk-reduction measures to reach water safety targets using an integrated fault tree model

Abstract

In the third edition of the Guidelines for Drinking-water Quality, the World Health Organization concludes that an integrated management of risks in source waters, treatment systems and distribution networks is the most effective way to guarantee safe drinking water to consumers. The integrated approach is fundamental to avoid sub-optimisation of risk-reduction measures. This paper presents an application of an integrated and quantitative risk model for comparing risk-reduction measures to support decisions for reaching specified water safety targets. A fault tree approach is used for structuring the risk analysis and for estimating the risk, expressed as Customer Minutes Lost (CML). Input information is a combination of hard data and expert judgements. Uncertainties in input information are considerable and modelled by a Bayesian statistical approach. A drinking water system in Sweden is used to exemplify model application. Quantitative safety targets have been confirmed at the political level as a basis for long-term planning of investments and reinvestments. One target defines an acceptable risk level of 144 annual CML for the average consumer. For the current system structure an estimated risk of 612 CML was obtained. Four risk-reduction alternatives were compared and they reduce the risk to between 50 and 81 CML, i.e. below the acceptable level. The paper describes how a structured and thorough analysis of risk-reduction measures can facilitate transparency and long-term planning of drinking water systems.