Determining Influential Household Routines for Domestic Water Consumption Estimation via Genetic Algorithm

Abstract

Domestic water consumption can be affected by many factors. Household routines that involve the use of water appliances such as number of time the occupants of a household took bath, flushing toilet, washing clothes and others ultimately regulate the amount of residential household’s monthly water consumption. Accurately estimating the amount of domestic water consumption is a very challenging task as these household routines differs from one another with one routine may be more influential than the others. This paper therefore proposes the employment of Genetic Algorithm (GA) in order to optimize the coefficient of micro-components of water consumption (coMC) to determine which micro-component of water consumption (household routines) is more influential than the others. This is accomplished by encoding the chromosome data in GA to incorporate the coMC values to minimize the domestic water consumption estimation error and subsequently enabling increased accuracy towards estimating the amount of monthly water consumption. Using household’s characteristics data and monthly water consumption from 80 residential households, it is discovered that there exist micro-components that are more influential towards the water consumption than the others.