An Experimental Water Consumption Regression Model for Typical Administrative Buildings in the Czech Republic

Abstract

Pressure management is the basic step of reducing water losses from water supply systems (WSSs). The reduction of direct water losses is reliably achieved by reducing pressure in the WSSs. There is also a slight decrease in water consumption in connected properties. Nevertheless, consumption is also affected by other factors, the quantification of which is not trivial. However, there is still a lack of much relevant information to enter into this analysis and subsequent decision making. This article focuses on water consumption and its prediction, using regression models designed for an experiment regarding an administrative building in the Czech Republic (CZ). The variables considered are pressure and climatological factors (temperature and humidity). The effects of these variables on the consumption are separately evaluated, subsequently multidimensional models are discussed with the common inclusion of selected combinations of predictors. Separate evaluation results in a value of the N 3 coefficient, according to the FAVAD concept used for prediction of changes in water consumption related to pressure. The statistical inference is based on the maximum likelihood method. The proposed regression models are tested to evaluate their suitability, particularly, the models are compared using a cross-validation procedure. The significance tests for parameters and model reduction are based on asymptotic properties of the likelihood ratio statistics. Pressure is confirmed in each regression model as a significant variable.

Objectives, Keys and Results in the Water Networks to Reach the Sustainable Development Goals

“…Consideration of water scarcity and climate change is key to optimizing water and energy resources in the community and in order to guarantee meeting the demand for water [16]. This challenge should be solved considering the use of information and communications technology (ICT) in the supply and irrigation networks.…”

A response surface model to predict and experimentally tune the chemical, magnetic and optoelectronic properties of oxygen-doped boron nitride

“…Upper and lower control limits (UCL and LCL) are then calculated, which are defined as the residual mean plus or minus three standard deviations (3σ). 54 Data points lying above or below the UCL and LCL, respectively, are deemed as outliers, which could impose bias on the data set and skew the response surface design and/or model predictions. The results for the studentised residuals tests for the oxygen content, specific paramagnetic OB3 content and band gap are presented in Figure S36.…”

Water Networks Management: New Perspectives

“…The effect of water pressure and climatological variables (temperature and humidity) on water consumption in an administrative building in Czech Republic was examined [2]. For the identification of water consumption and its prediction, regression models were used.…”