Predicting demand of water in Baghdad city: A comparison between ARIMA and Curve Estimation Methods

Abstract

This paper aim aims to predict the quantities of water needed in the city of Baghdad for the next 10 months. This paper focuses on potable water, based on the time series data of the water consumption phenomenon in the city, which was obtained from the Ministry of Water Resources, specifically the Baghdad Water Department. Statistical forecasting techniques were used on the monthly water consumption data for the city of January 2014 until May (2024, a total of 125 months, and that is Baghdad in the period from to reach an estimate of the quantities needed by the city of Baghdad in the future. Curve Estimation and Linear Regression forecasting techniques were used, such as linear regression analysis and the Box - Jenkins (ARIMA) methodology, to obtain the best water consumption model in the city of Baghdad and the most accurate. In This paper we concluded that it is the best model suitable for predicting monthly water consumption in Baghdad city is (3,1,1) ARIMA among the models proposed in the Box-Jenkins methodology in terms of accuracy measures and (Mean Absolute Percentage Error) which reached (2.44-MAPE).While the (Mean Absolute Percentage Error) for the Simple Linear (MAPE=8) Quadratic Regression model and the Quadratic Regression model were also found, the research concluded that monthly consumption will increase in the city of Baghdad, when compared to Between the actual values ​​and the predictive values ​​of the methods used in the paper to predict the future. Finally, it is recommended to take the necessary measures to limit water consumption in the city, through pricing, awareness, education, intermittent supplies and other measures that preserve water resources and achieve sustainability.