THE JOURNAL OF KMUTNB

Volume 32, No. 04, Month OCTOBER, Year 2022, Pages 1004 - 1013

Sale forecasting for cosmetics appropriate for manufacturers in asean

Pheeraya Taothong

Abstract Download PDF

This research aimed to study the forecasting model for sales of three new suitable cosmetics products by comparing the information from the past 12 periods of sales volume using time series to forecast 24-month sales in advance. Data analysis was performed by using the Winter’s technique, decomposition technique, and trend analysis techniques. To find the most suitable forecasting method, criteria of the lowest Mean Absolute Deviation (MAD), Mean Square Errors (MSE), and Mean Absolute Percent Error (MAPE) for the lowest value were employed. The results revealed that Winter’s techniques was the most suitable forecasting model for brand T and brand S and Mean Absolute Percent Error (MAPE) was equal to 27 and 35 respectively. The decomposition technique was appropriate for brand B and Mean Absolute Percent Error (MAPE) was equal to 17. The overall assessment of the application for assisting the decision of the manufacturer was in good level with approximately 4.16 scores. The solution can be used as a system for sale forecasting in order to prepare the manufacturer both on the quantity and the quality of the products and services.

Keywords

Forecasting; Product Sales; Cosmetics Product

---