Introduction:

Artificial intelligence is a basic component of the so-called Industry 4.0, so its introduction into technological processes constitutes a significant contribution to its approach to this paradigm. The objective of this work is to develop and validate a group of technologies based on soft computing tools applied to mechanical manufacturing processes, with the purpose of bringing them closer to the Industry 4.0 paradigm.

Methods:

Among the tools developed, fine-tuned, and applied, both machine learning techniques were included, such as artificial neural networks, fuzzy systems, and deep learning for modeling and pattern detection tasks, as well as metaheuristics inspired by nature. for optimization. The applications have been aimed at manufacturing processes, such as welding, machining and micromachining, solving problems such as failure detection, quality control and process optimization based on their sustainability.

Results:

In all the case studies considered, the applied techniques proved to be fully effective for solving the problems posed, both those focused on modeling and pattern recognition, as well as those aimed at optimizing processes. In addition to being widely validated in laboratory conditions similar to industrial ones, the technologies developed were introduced into the productive practice of liquefied gas containers, demonstrating favorable impacts both economically (decreased production costs) and environmental (lower energy consumption in the production process). Conclusions: The effectiveness shown by the artificial intelligence tools applied to the solution of practical problems of manufacturing processes, constitute a valuable set of know-how, with wide possibilities of being applied to various sectors of the industry, allowing their incorporation into solutions. of Industry 4.0, such as digital twins and cyber-physical systems.

        · resumen en Español     · texto en Español     · Español ( pdf )