A stochastic early warning detection system model for drinking water distribution systems security

Following the events of 9/11 2001 in the US, a deliberate contamination intrusion into a drinking water distribution system is considered a major terrorist threat. Chemical or biological injected agents can spread throughout the system causing sickness or death among the people consuming the water. A methodology is developed and demonstrated in this paper to enhance water distribution system security, linking EPANET and a genetic algorithm in an overall framework for optimally allocating monitoring stations, aimed at capturing deliberate external terrorist hazards intrusions through water distribution system nodes: sources, tanks, treatment plant intakes, consumers - subject to extended period unsteady hydraulics and water quality conditions, for a given defending level of service to public - a maximum volume of polluted water exposure at a concentration higher than a minimum hazard level. The methodology developed and demonstrated extends previous work of the authors on this topic by treating the demands and the injected pollution rates quantities as random variables, and by explicitly taking into account a delay between the pollution event and the monitoring equipment response capability.