Forecasting Roof Tile Production Using SMA and SES for Alternatives to Production Management

This research aims to predict roof tile production trends at one of the roof tile companies in Kebumen in order to assist the company management in determining and providing management recommendations for the tile production that occurs so that production management always runs well and is directed. Forecasting is carried out by implementing the Single Moving Average (SMA) and Single Exponential Smoothing (SES) methods on roof tile production transaction data over a 60 month period, namely January-December 2019 to January-December 2023 to produce monthly forecasts for roof tile production predictions. The total sample of training data processed is 1, 415, 987 records, which are roof tile production transaction data, as well as data in January 2024 as test data (to test the accuracy of predictions). The results of forecast testing based on MAPE (Mean Absolute Percentage Error) for SMA forecasting with n-th ordo 9 is the best n-th ordo, amounting to 9.16 %. This means that an accuracy of 90.84 % is obtained, and SES with an alpha value of 0.1 is the best alpha, with a MAPE value of 9.49 % with an accuracy of 90.51 %, which means the accuracy is very good, namely above 90 %. Based on the MAPE results obtained, the SMA method is the right method for carrying out periodic forecasting for roof tile companies in carrying out the production process in order to maintain stability and avoid unexpected things.