AI Summary & Key Takeaways
AI InsightThe study results showed that the regression method gave the best predictions for the indicators in the water distribution dimension, while the time-series exponential smoothing method gave the best predictions for the indicators in the financial and agricultural activities dimension.
📌 3 Key Takeaways
Abstract
Decreasing availability of water and increasing consumption of it in recent years shows water management plans increase its value. Depending on water consumption in the current and previous years, it is important to predict water consumption in the coming years and make plans accordingly. Evaluation of the performance of water use in agriculture, identifying water resources that are most intensively used and prediction of the potential performance in the coming years have become increasingly important. Irrigation water management is of crucial importance for sustainable food security and needs to plan for saving water due to global warming and climate change in future. Some statistical methods such as regression and time-series to make accurate predictions are used to predict future irrigation management. However which methods are most suitable in this area is a gap in previous studies. This study aimed to determine the most accurate prediction method based on a comparison of the methods used in irrigation performance such as regression, time-series exponential smoothing and time-series ARIMA (autoregressive integrated moving average) model. In the study, Kahramanmaraş region was randomly selected and the irrigation data of 2006–2018 were used and the data of 2006–2017 were analysed to predict the data of 2018. Then, the values predicted using the methods were evaluated based on the actual values of 2018, and the method that projected values similar to the actual values was determined. The study results showed that the regression method gave the best predictions for the indicators in the water distribution dimension, while the time-series exponential smoothing method gave the best predictions for the indicators in the financial and agricultural activities dimension.