Forecasting demand trends in automotive industry: Comparative analysis of exponential smoothing and regression analysis

Abstract

This study analyzes demand trends using statistical methods, specifically exponential smoothing and regression analysis, applied to data from an automotive supply chain company. The analysis of order records for the first 28 weeks of the year reveals that exponential smoothing, with a smoothing parameter of α=0.5, provides more accurate forecasts compared to regression analysis. This conclusion is supported by lower forecast error values (MAPE, MSE, and MAD) for the exponential smoothing method. The findings suggest that exponential smoothing is a more reliable tool for demand forecasting in this context.