<![CDATA[Accurate spare part inventory control remains a challenge for service centers, as fluctuating demand often results in stock-outs or excess inventory that increases operational costs and reduces customer satisfaction. The research involves creating a web-based inventory for AGH Center, incorporating ensemble-based demand forecasting with the Economic Order Quantity (EOQ) approach. The developed system, using six months of historical transactional and spare parts consumption data, produces demand forecasts for the next 7, 14, and 30 days, using a weighted combination ofARIMA, Exponential Smoothing, Linear Regression, and Moving Average forecasting estimation methods. Accuracy of the demand forecasts is measured with MAPE, RMSE, MAE, R², bootstrap confidence intervals, and statistical tests for thresholds. Ensemble forecasting has proven to be superior to other benchmarks as depicted with MAPE metric scores at annual averages of 12.8%, 16.4%, and 19.7%, greatly improving the estimation of annual demand forecasts. Integrating EOQ further enables total cost savings of 26.11%. The system has automated forecasting and EOQ calculations, and advanced computerized frameworks to solve for demand in real-time which though predictable, remains a remarkable adaptation to spare parts inventory control]]>
Copyrights © 2025
