Improvement of a restaurant’s Service Level through Process Standardization and Machine Learning Techniques

Abstract

This research aimed to demonstrate how combining engineering tools with Machine Learning models can help reduce the percentage of unserved customers in a restaurant. To achieve this, a methodology was designed that integrated process standardization through BPMN diagrams, the development of a digital reservation registration system, and the implementation of a predictive model based on linear regression. This approach made it possible to connect process improvement with data analytics, resulting in a practical solution adapted to the company’s real operational conditions. Additionally, a custom code has been developed to make the most of the available information, despite the initial limitations related to the lack of digitalization. The results show significant improvements. The service level increased from 86.6% to 95.8% using the program Arena simulator, while demand planning accuracy improved from approximately 85% to 95%. Furthermore, a system guaranteeing 100% traceability of reservations was implemented, something the company did not have before. These findings demonstrate that integrating BPMN, process standardization and Machine Learning is a viable, efficient and replicable alternative for demand planning in environments with limited resources. Finally, the study concludes that digitalization plays a key role in enhancing operational efficiency within the food service sector. It is recommended to strengthen a data-driven organizational culture and progressively advance toward the adoption of digital predictive systems.