Sales forecasting for Chemical Products by Using SARIMA Model

Sales forecasting is widely used in enterprise resource management, which provides valuable information for efficient management. Sales forecasting facilitates the company to produce and stock products on demand. Based on the analysis of time series, future sales can be predicted through historical sales. This research presents and analyses the use of the Seasonal Auto-Regressive Integrated Moving Average (SARIMA) method for forecasting the monthly sales of a chemical company. The dataset for the SARIMA model is based on the company's sales data. The Difference Operation is used to stabilize the time series and the Box-Jenkins method is used to select the most suitable model among five SARIMA models. The model is used to forecast the company's sales. The forecasting precision of the model is evaluated by Mean Square Error. The results of the forecast are similar to actual sales data, which verifies that SARIMA has good results in the chemical company's sales forecasting. Based on the sales data, we also use the SARIMA model to forecast the company's sales in the future. Sales increase in comparison with last year, indicating that the SARIMA model is useful in Sales forecasting for the chemical industry.