Determining Purchase Orders Using the Weighted Moving Average Method in a Web-Based System (Case Study: CV. Gilang Mandiri Tasik)

Abstract

CV. Gilang Mandiri Tasik is a company that sells various kinds of products such as basic necessities, biscuits, snacks and others. The problem that is often faced in the inventory management process is the riskof overstock. So we need a system to predict stock in the future period by referring to data from the previous period. This study uses the Weighted Moving Average method. The Weighted Moving Average method is a moving average method that is widely used to determine the trend of a time series.

In testing the system, the author examines the data collected, namely transaction data on CV. Gilang Mandiri Tasik from November 2021 to April 2022. Based on the results of the analysis, the data for the last 6 months (sales transaction period) under the name of Resto Cooking Oil product, was analyzed using a 3-month moving period and the number of periods predicted for the next 2 months. Processed according to the WMA formula produces: MSE (Mean Squared Error) = 20.136.574.07, RMSE (Root Mean Squared Error) = 4.487.38, MAE (Mean Absolute Error) = 2.861.11, MAPE (Mean Absolute Percentage Error) = 38, 33%. Then the prediction results are known to meet the stock for the next 2 months (May 2022 = 6,083 Stock) and (June 2022 = 6,708 Stock).