Sensitivity Analysis of an Agent-Based Model Used to Simulate the Spread of Low-Flow Fixtures for Residential Water Conservation and Evaluate Energy Savings in a Canadian Water Distribution System

Abstract

The results of a sensitivity analysis performed to examine the impact of uncertain parameters in an agent-based model (ABM) on the prediction of water use and energy use for pumping and water treatment in a distribution system are presented. The ABM was coupled with a water end-use model and the EPANET2 network solver to simulate word-of-mouth communication between water users and the adoption of water-efficient fixtures that lead to water use and energy savings in distribution systems. Three key ABM parameters (adoption probability, initial penetration rates, and connections per agent) were found to have an important impact on the adoption rate of low-flow fixtures. The parameter values were chosen from relevant literature and the sensitivity analyses determined that the adoption of four specific low-flow fixtures—toilets, showers, washing machines and dishwashers—were estimated to have the potential to reduce water use (9%) and energy use (9%) for pumping and treatment within a mid-sized Canadian water distribution system. The work provides an innovative framework that simulates human interactions and evaluates how these interactions can affect water and, subsequently, energy use in distribution systems.