Artificial Intelligence Methods for Process Automation:A Systematic Literature Review

Villegas Chavez R.A.; Dios Castillo C.A. · Proceedings of the LACCEI international Multi-conference for Engineering, Education and Technology (2025) · DOI: 10.18687/leird2025.1.1.420

Resumen

Process automation through artificial intelligence improves accuracy and operational efficiency. In this systematic review of publications indexed in Web of Science and Scopus, using the PRISMA method, 66 articles were selected. Automation increases accuracy, strengthens robustness as models adapt to new data, and improves efficiency. In the literature from 2024 to 2025, machine learning predominates for structured data and deep learning for unstructured data; the YOLO family is the most frequent in deep learning and decision trees in machine learning. Deep learning models Darknet 19, ResNet 18, ResNet 50 and ResNet 10 reached 100%, and the Extra Trees algorithm in machine learning reached 99.7% in health studies, which account for the majority of experiments. © LEIRD 2025.All rights reserved.

Instituciones: Universidad Tecnológica del Perú