<![CDATA[Fluctuating demand for school accessories, such as caps and neckties, presents a significant challenge for the garment manufacturing industry in determining optimal production quantities. Inaccurate production planning can lead to overstocking or stock shortages, resulting in increased operational costs and reduced customer service quality. This study aims to implement a Linear Regression model to predict school accessory orders using historical sales data from 2021 to 2025. A quantitative research approach was adopted, consisting of data preprocessing, feature engineering, dataset partitioning into 80% training and 20% testing sets, and the development of a prediction model using Linear Regression. The input variables included year, month, education level, product type, time index, semester, and new academic year period. The experimental results demonstrate that the proposed model achieved a Mean Absolute Percentage Error (MAPE) of 12.91% and a Mean Absolute Error (MAE) of 149.61 units, indicating good predictive performance for forecasting school accessory orders. Furthermore, semester, product type, and the new academic year period were identified as key factors influencing demand fluctuations. The proposed model can serve as a decision-support tool for production planning and inventory management in the garment manufacturing industry, enabling more efficient resource allocation and demand-driven production planning.]]>
Copyrights © 2026
