Integrating evolution strategies and genetic algorithms with agent-based modeling for flushing a contaminated water distribution system

Abstract

Water utilities can prepare for water distribution hazards, such as the presence of contaminants in the pipe network and failure of physical components. In contamination events, the complex interactions among managers' operational decisions, consumers' water consumption choices, and the hydraulics and contaminant transport in the water distribution system may influence the contaminant plume so that a typical engineering model may not properly predict public health consequences. A complex adaptive system (CAS) approach couples engineering models of a water distribution system with agent-based models of consumers and public officials. Development of threat management strategies, which prescribe a set of actions to mitigate public health consequences, is enabled through a simulation–optimization framework that couples evolutionary algorithms with the CAS model. Evolution strategies and genetic algorithm-based approaches are developed and compared for an illustrative case study to identify a flushing strategy for opening hydrants to minimize the number of exposed consumers and maintain acceptable levels of service in the network.