Access to adequate water is a battle the City of uMhlathuze faces daily, with the disadvantaged and impoverished areas being most affected as a result of a rapidly increasing growth in human population. The future water condition will be impossible to manage unless the municipality is able to address the increased demand for water which often leads to illegal water connections. The goal of this study is twofold: first to predict human population growth over a 50-year period and secondly, to formulate a linear programming model to determine the number of additional reservoir tanks which need to be built, in order to satisfy the human water demand for the next half century. This study compares the application of automated time series forecasting models like ARIMA, ETS, and BATS model to predict the growth in human population. The best model to forecast human population growth was then selected using the forecasting assessment criterion of RMSE, MAE, MASE and MAPE. Based on the forecast results obtained in this study, ARIMA (1,1,0) with drift model has shown a better prediction accuracy in terms of the RMSE = 3,725, MAE = 2,265, MASE = 0,211, and MAPE = 0,757, than the ETS(A,Ad,N), and BATS(1,{0,0}, 1,-) model to explain the observed values of a time series. The forecast results derived from ARIMA (1,1,0) with drift model have indicated that human population growth will triple by 2067, albeit at a higher pace than at any period since 1996. On the basis of human population growth forecasts, it is suggested that 687 additional reservoir tanks with a capacity of 47,500 kilolitres are built within the next 50 years in order to supplement the increased demand for water that comes with an increase in human population. An implication of these findings is the possibility that, should anticipated human population growth materialise, the current reservoir tanks are likely to come under stress due to the increased demand for water.

  • Water provision planning.

  • Application of automated time series forecasting models to predict the growth in human population.

  • Formulation of a new linear programming model along with its solution to determine the number of additional reservoir tanks to be built for half a century in advance.

Graphical Abstract

Graphical Abstract
Graphical Abstract
This content is only available as a PDF.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY-NC-ND 4.0), which permits copying and redistribution for non-commercial purposes with no derivatives, provided the original work is properly cited (http://creativecommons.org/licenses/by-nc-nd/4.0/).