<![CDATA[Raw material inventory management in catering businesses required an accurate planning system to prevent stock shortages and overstock conditions that could disrupt operational activities. This study aimed to develop a web-based inventory prediction system by implementing the Weighted Moving Average (WMA) and Single Exponential Smoothing (SES) methods and to compare the accuracy of both methods using the Mean Absolute Percentage Error (MAPE). The data used in this study were historical raw material usage records from Bubin Catering covering the period from March 12, 2025, to May 8, 2025. The system was developed using the Laravel framework and included inventory management, incoming goods, outgoing goods, prediction, and detailed calculation features. The results of Black Box Testing showed that all system functions operated as expected. Based on the evaluation of five raw material samples, the WMA method produced MAPE values of 33%, 36%, 24%, 55%, and 41%, while the SES method produced the best MAPE values of 32.53%, 36.80%, 56.70%, 61.00%, and 46.60%. Overall, the WMA method produced lower prediction errors for four of the five raw material samples, whereas the SES method performed better for only one sample. Therefore, the WMA method was more suitable for predicting inventory requirements using the historical data of Bubin Catering in this study.]]>
Copyrights © 2026
