Comparative Study to Determine Optimal Models for Forecasting the Number of Tap Water Users of the Provincial Waterworks Authority Pathum Thani Branch

Abstract

The objective of this study was to determine optimal models for forecasting the number of tap water users of the provincial waterworks authority (PWA) Pathum Thani branch. In this study, we applied 5 forecasting techniques to analyze the data. These include linear trend analysis method, quadratic trend analysis method, cubic trend analysis method, exponential trend analysis method, and power trend analysis method. Our raw data were secondary data, which were taken from the data center and information technology planning division, the PWA Pathum Thani branch. All 91 values were separated on a monthly basis between from January 2012 and July 2019, which were used and separated into 2 sets. The first set with 84 values were separated on a monthly basis between from January 2012 and December 2018 for comparing and finding the most suitable forecasting method via criteria of the lowest mean absolute deviation (MAD) and mean absolute percent error (MAPE). Then the selected suitable method was used to determine the most suitable forecasting period by the second group, which contained 7 values and were separated on a monthly basis between from January 2019 and July 2019. The lowest MAPE was used as the criteria of each period. The result indicated that cubic trend analysis method was the best method, with the forecasting model: . When  represents time from the method, it was implemented for forecasting 3, 5 and 7 months, and showed that the method was suitable for advance 7 months.