Predictive model based on machine learning for raw material purchasing management in the retail sector.

• Autores: Antunez, Julio C., Salazar, Johnny D., Castañeda, Pedro S.
• Formato: artículo
• Fecha de Publicación: 2024
• Institución: Universidad Peruana de Ciencias Aplicadas
• Repositorio: UPC-Institucional
• Lenguaje: inglés
• OAI Identifier: oai:repositorioacademico.upc.edu.pe:10757/675912
• Enlace del recurso: http://hdl.handle.net/10757/675912
• Nivel de acceso: acceso embargado
• Materia: Inventory management; Model interpretation; SMEs

Making raw material purchase forecasts for companies is very difficult and, if inadequately controlled, can affect the company's decision making and profitability. Currently, there are optimized systems or mathematical models to try to predict the demands and solve this problem. In this study, a raw material purchase prediction model is proposed that uses the Elastic Net algorithm to analyze historical sales and inventory data. The model is used to improve prediction accuracy, allowing SMEs to optimize inventories, reduce costs and improve efficiency. Experimental results indicate that the proposed model obtains better results in the MAE, RMSE and R2 indicators.