Application of Time Series Forecasting Method to Estimate National Salt Demands

Abstract

Salt is one of the most important commodities for domestic use and as a raw material for industry. It is essential to make an estimate salt requirement to meet them appropriately. The purpose of the study was to estimate salt needs using the time series forecasting method and to identify the most effective technique for salt needs forecasting. Forecasting analysis uses Naive, Moving Average, Weighted Moving Average, Exponential Smoothing, Exponential Smoothing with Trend, and Trend Projection methods. Forecasting accuracy is tested using MAD, MSE, and MAPE. Based on the results, the Trend Projection is the most effective time series forecasting technique for predicting salt requirements. This method was selected due to its lowest error rate value (MAD of 0.16, MSE of 0.04, and MAPE of 4.28%) compared to other methods. According to projected estimates, the amount of salt required in 2024 would be 4.86 million tons.