A Methodology for Water Quality Regulatory Compliance Using Bayesian Belief Networks

Abstract

Water quality failures in the developed world occur primarily in drinking water distribution systems in rural or small communities. Increased water system monitoring through the use of information technologies, such as sensors and SCADA systems, can reduce the risk of undetected water quality failures and support small utilities in keeping abreast of increasing regulatory stringency. However, because of financial constraints, rural utilities may be particularly susceptible to some of the shortcomings associated with these technologies, such as false readings and data corruption. Bayesian Belief Networks (BBNs) are proposed as a means to integrate sensor information with other operational information to mitigate technological shortcomings and increase certainty about the state of a given drinking water system. The methodology used to develop the BBN for a case study system is discussed in detail.