Resumen
La prevención de accidentes laborales en entornos industriales de alto riesgo requiere soluciones capaces de anticipar condiciones inseguras en tiempo real. El objetivo de esta investigación fue desarrollar un modelo predictivo basado en inteligencia artificial y sensores inteligentes para la prevención dinámica de accidentes laborales. La metodología integró tecnologías del Internet de las Cosas Industrial (IIoT), algoritmos de aprendizaje automático, simulación Monte Carlo e inteligencia artificial explicable para analizar variables ambientales, operacionales y humanas. Los resultados mostraron que el modelo XGBoost alcanzó una exactitud del 94 % y un AUC de 0,97, mientras que el análisis SHAP identificó la fatiga operacional, las vibraciones anormales y la temperatura ambiental como los principales factores asociados al riesgo. Se concluye que la integración de sensores inteligentes, inteligencia artificial y simulación avanzada fortalece la seguridad industrial mediante sistemas predictivos orientados a la detección temprana de riesgos y la toma de decisiones preventivas.
Citas
P. Armenteros-Cosme, M. Arias-Gonzalez, S. Alonso-Rollan, S. Marquez-Sanchez, and A. Carrera, “Advancements in artificial intelligence and machine learning for occupational risk prevention: A systematic review on predictive risk modeling and prevention strategies,” Sensors, vol. 25, no. 17, Art. no. 5419, 2025, doi: 10.3390/s25175419.
J. Park and D. Kang, “Artificial intelligence and smart technologies in safety management: A comprehensive analysis across multiple industries,” Applied Sciences, vol. 14, no. 24, Art. no. 11934, 2024, doi: 10.3390/app142411934.
M. A. S. Akash, “Artificial intelligence-enabled predictive safety analytics for proactive hazard identification in industrial manufacturing environments,” International Journal of Scientific Interdisciplinary Research, vol. 7, no. 1, pp. 416–452, 2026, doi: 10.63125/vp0mpx51.
J. Shekh, “A systematic review of artificial intelligence-based predictive safety models for reducing workplace injuries in manufacturing and construction,” American Journal of Advanced Technology and Engineering Solutions, vol. 6, no. 1, pp. 180–227, 2026, doi: 10.63125/jfpn5t74.
B. K. Nakka and K. Dilip, “Intelligent deep learning system for monitoring workers in high-risk areas,” Journal of Recent Innovation in Science and Technology, vol. 1, no. 2, pp. 1–12, 2025, doi: 10.70454/JRIST.2025.10201.
A. Yurtcu, “Wearable technologies and Internet of Things for real-time hazard monitoring in occupational health and safety: A systematic content analysis,” Innovative Approaches to Engineering Problems, vol. 1, no. 2, pp. 45–54, 2025.
M. Khurram et al., “Artificial intelligence in manufacturing industry worker safety: A new paradigm for hazard prevention and mitigation,” Processes, vol. 13, no. 5, Art. no. 1312, 2025, doi: 10.3390/pr13051312.
D. R. Moura et al., “Artificial intelligence platform for monitoring and risk prevention in confined spaces,” presented at the Offshore Technology Conference, Houston, TX, USA, May 2025, doi: 10.4043/35923-MS.
R. Pires et al., “AI-powered safety: Transforming occupational risk management through predictive intelligence,” Procedia Computer Science, vol. 280, pp. 929–934, 2026, doi: 10.1016/j.procs.2026.04.116.
M. Ramirez-Pena, A. Cerezo-Narvaez, M. Otero-Mateo, A. Pastor-Fernandez, and M. Batista, “AI-enhanced safety in project-based production: A new era in workplace risk,” in Artificial Intelligence and Digital Transformation, ser. Lecture Notes in Information Systems and Organisation, vol. 78. Cham, Switzerland: Springer, 2025, doi: 10.1007/978-3-032-00118-4_4.
B. R. Polo-Escobar, R. E. Polo-Moreano, L. M. Estrada Torres, and R. M. Cornejo Urbina, “Interfaces hombre-máquina adaptativas basadas en inteligencia artificial y neuroergonomía para la reducción de errores humanos en sistemas industriales complejos,” Athenea, vol. 7, no. 23, pp. 54–65, 2026, doi: 10.47460/athenea.v7i23.133.
