Enhancing Service Levels in a Peruvian Beverage SME: An Innovative Model Integrating Machine Learning and 5S Methodology

• Autores: Loyola Ferrer, Gabriela, Veldi Díaz, Wiebke Fernanda, Quiroz Flores, Juan Carlos
• Formato: artículo
• Fecha de Publicación: 2025
• Institución: Universidad de Lima
• Repositorio: ULIMA-Institucional
• Lenguaje: inglés
• OAI Identifier: oai:repositorio.ulima.edu.pe:20.500.12724/23244
• Enlace del recurso: https://hdl.handle.net/20.500.12724/23244; https://doi.org/10.14445/22315381/IJETT-V73I3P140
• Nivel de acceso: acceso abierto
• Materia: Pendiente

In Peru, the food and beverage services sector shows constant annual growth and a demand experiencing significant monthly variations. However, SMEs in this sector face recurring problems due to their low levels of preparedness, making demand fluctuations a latent threat. Food and beverage service SMEs suffer from inadequate inventory management, inaccurate supply planning and a lack of process optimization, which leads to a low level of On Time in Full (OTIF) deliveries. This is detrimental to an SME, as customer satisfaction is crucial in this sector. Due to this, the present study proposes an innovative Lean Management model that integrates 5S tools and Machine Learning to increase OTIF in a Peruvian beverage service SME. The research focuses on the Milk Tea and Smoothie product lines, which exhibited stockouts and high preparation and stock review times through analysis and diagnosis. The proposed model results in a reduction of 75.98% in the total cycle time of an order. Additionally, the implementation of Machine Learning helped reduce stockouts by providing a more accurate supply forecast, improving forecast error by 19.72% and 38.71% for tapioca and milk, respectively. These indicators led to a 51.42% increase in OTIF. Thus, this management model effectively innovates by adapting tools often used for manufacturing and production to the services sector, thereby achieving outstanding results in both efficiency and customer satisfaction.