Forecasting Motorcycle Spare Parts Demand Using the Least Squares Method

A Case Study of Teguh Jaya Motor

Abstract

Accurate demand forecasting is a critical factor in inventory management, particularly for small and medium-sized automotive workshops where stock shortages may lead to customer loss and operational inefficiencies. This study aims to forecast motorcycle spare parts demand at Teguh Jaya Motor by applying the least squares method to historical sales data from January 2020 to December 2024. A quantitative research approach was adopted, utilizing time-series analysis of ten key spare parts, including clutches, engine oil, brake pads, tires, spark plugs, bearings, pistons, starters, and shock absorbers. The forecasting model was evaluated using the Mean Absolute Percentage Error (MAPE) to assess prediction accuracy. Furthermore, the proposed forecasting approach was implemented in a web-based application developed using PHP, HTML, CSS, and MySQL to support practical decision-making. The results indicate that the least squares method provides high forecasting accuracy, with MAPE values ranging from 3.97% to 10.21%, categorized as “very good” to “good.” These findings demonstrate that the least squares approach is suitable for predicting spare parts demand trends and can effectively support inventory planning in motorcycle workshops.