Selection of the Best Forecasting Method at PT. Indaco Warna Dunia
Maharani Sari Mutiara Mentari, Irwan Iftadi
Abstract
Open-access reader
Maharani Sari Mutiara Mentari, Irwan Iftadi
Abstract
Open-access reader
PT. Indaco Warna Dunia is a decorative paint company in Indonesia that produces products under the brands Envi, Belazo, and Top Seal. Preliminary observations revealed that the forecasting method used by the company is ineffective and inaccurate. This inaccurate forecast result company’s problem in fulfilling the demand. This study aims to select the best forecasting method to improve forecast effectiveness and accuracy. The research was conducted at the Tarakan depot, and the products understudy were a fast-moving product category, specifically the Envi brand. Several forecasting methods such as Moving Average, Weighted Moving Average, Single Exponential Smoothing, Exponential Smoothing with Trend, and Holt's Double Exponential Smoothing. The accuracy of forecasting is the most important and it can be measured with MAPE (Mean Absolute Percentage Error). The results showed that Holt's Double Exponential Smoothing method is the best for three products, while the Exponential Smoothing with Trend method, and Single Exponential Smoothing method are the best for one of the products, respectively.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
PT. Indaco Warna Dunia is a decorative paint company in Indonesia that produces products under the brands Envi, Belazo, and Top Seal. Preliminary observations revealed that the forecasting method used by the company is ineffective and inaccurate. This inaccurate forecast result company’s problem in fulfilling the demand. This study aims to select the best forecasting method to improve forecast effectiveness and accuracy. The research was conducted at the Tarakan depot, and the products understudy were a fast-moving product category, specifically the Envi brand. Several forecasting methods such as Moving Average, Weighted Moving Average, Single Exponential Smoothing, Exponential Smoothing with Trend, and Holt's Double Exponential Smoothing. The accuracy of forecasting is the most important and it can be measured with MAPE (Mean Absolute Percentage Error). The results showed that Holt's Double Exponential Smoothing method is the best for three products, while the Exponential Smoothing with Trend method, and Single Exponential Smoothing method are the best for one of the products, respectively.
Key concepts: Exponential smoothing, Smoothing, Exponential function, Selection (genetic algorithm), Product (mathematics), Mean absolute percentage error, Computer science, Double exponential function