Abstract

Water distribution systems (WDS) are vulnerable to accidental contamination events and intentional attacks that may cause dire effects on public health. In the event of water contamination, consumers may complain about unusual color, smell, or taste of their drinking water to their local water utility. Utility managers use this information to implement response actions that address the water quality problem. To maximize the effect of the response actions, the location of the source of contamination should be known. Consumer complaints can be used to identify the source. Since consumers behave in a complex manner depending on their characteristics (i.e. age, gender, mobility, and water consumption habits), complexity is added to the existing dynamics and complexity of WDS modeling. An Agentbased Model (ABM) is used to simulate complex consumer actions within a WDS during a contamination event. ABM and a WDS simulation model are coupled with an evolutionary algorithm to solve for the contamination source characteristics. However, since consumer behavior is less predictable, the use of their complaints to identify the source of contamination may not always lead to one unique source. Nonuniqueness within a WDS makes it difficult for utility managers to implement optimal response actions. To address the problem of non-uniqueness, alternative sources of contamination are generated.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call