DEVELOPING METHODOLOGY FOR FORECASTING SPARE PARTS CONSUMPTION IN AN AUTO SERVICE ENTERPRISE

Abstract

Efficient forecasting of spare parts consumption is essential for the smooth operation and profitability of auto service enterprises. Inaccurate forecasting often leads to inventory imbalances—either overstocking, which ties up capital, or stockouts, which cause service delays and customer dissatisfaction. This article presents a comprehensive methodology that combines historical data analysis, statistical forecasting models, seasonality adjustments, and integration with enterprise systems. The use of models such as ARIMA and linear regression, alongside classification techniques like ABC analysis, enables more precise demand prediction. Furthermore, incorporating external variables and emerging technologies such as artificial intelligence significantly enhances forecast accuracy. The methodology is validated through a comparative analysis of forecasting techniques using error metrics like MAPE and RMSE. The findings demonstrate that data-driven and adaptive forecasting approaches can improve inventory control, reduce costs, and optimize service performance in the automotive sector.