Evaluación del algoritmo de cuello de botella móvil utilizando inteligencia artificial en procesos de manufactura

Abstract

This study evaluated the effectiveness of the Mobile Bottleneck Algorithm (CBM) enhanced with Artificial Intelligence (AI) in an oil bottling plant. A comparative analysis was conducted between the traditional method, relying on workers' expertise, and the implementation of CBM with AI. The methodology included data collection from the plant, process simulation using Simulink, and the implementation of a hybrid algorithm. The results showed that in low-complexity plants, worker expertise can be as efficient as automated algorithms. However, in environments with a larger number of machines, CBM with AI improved the accuracy in identifying bottlenecks, optimizing production by 22.73%. The integration of Machine Learning enabled dynamic adaptation to system conditions, enhancing task planning and distribution. It was concluded that implementing CBM with AI is an efficient tool for improving productivity in manufacturing, provided that an adequate database is available. The algorithm's scalability makes it viable for companies of different sizes, facilitating decision-making in dynamic production environments.