3-parameter log-logistic distribution modelling and scenario development to evaluate the United Kingdom’s water demand

A major unresolved problem in water consumption modelling is the identification of a statistical distribution function which best represents the random nature of water consumption patterns. Relatively, few studies can be found in the literature in which the statistical methods have been considered to address the random variations in water demand. Furthermore, it is often assumed in water demand modelling, that variations in water demand for drinking water distribution systems follow the normal distribution with insufficient justification. The purpose of this study is to evaluate the real water demand data from two water companies from United Kingdom (UK) to identify best fit statistical distribution function to simulate water demand. Daily water consumption data for five years ranging from 2009 to 2013 were obtained from two water companies in the UK and were analysed by fitting into the normal, log logistic and 3 parameter log logistic distributions. Statistical modelling was performed using the MINITAB18 statistical package. The Anderson Darling (AD) statistical test was used as the goodness of fit parameter in the analysis