A rule-based decision support system for sensor deployment in small drinking water networks

The physical layout of drinking water utilities makes them inherently vulnerable to contamination incidents caused by routine operations. These contaminations present environmental health concerns including but not limited to total trihalomethanes, lead, and chlorine residual issues. To achieve the goal of cleaner production, sensor placement in municipal drinking water networks in response to possible public health threats has become one of the most significant challenges currently facing drinking water utilities, especially in small-scale communities. Long-term monitoring is needed to develop modern concepts and approaches to risk management for these utilities. We developed a Rule-based Decision Support System (RBDSS), a methodology to generate near-optimal sensor deployment strategies with low computational burden, such as those we often encountered in large-scale optimization analyses. Three rules were derived to address the efficacy and efficiency characteristics of such a sensor deployment process: (1) intensity, (2) accessibility, and (3) complexity rules. Implementation potential of this RBDSS was assessed for a small-scale drinking water network in rural Kentucky, United States. Our case study showed that RBDSS is able to generate the near-optimal sensor deployment strategies for small-scale drinking water distribution networks relatively quickly. The RBDSS is transformative and transferable to drinking water distribution networks elsewhere with any scale.