Predictive System for Heat Stress in High-Risk Occupational Environments

Bonifacio R.F.A.; Rojas J.J.H. · Journal of Robotics and Control (JRC) (2025) · DOI: 10.18196/jrc.v6i6.28088

Resumen

Thermal stress from extreme temperatures combined with high relative humidity and minimal ventilation is an occupational concern for employees across a variety of industries. This paper presents an AI-based predictive analytics system for real-time thermal risk calculation, utilizing temperature, humidity, wind speed, and selected environmental factors, as well as physiological parameters, based on FIALA and ISO 7933. The model was synthesized through stratified sampling with 15,000 synthetic data points and validated through simulations covering a wide range of thermal scenarios. The Ensemble Boosted Trees (EBT) algorithm produced a correlation coefficient greater than 0.95 and a mean absolute error of less than 0.2 °C in the prediction of the core temperature. Similarly, the risk classification was based on exceeded physiological limits and being able to issue preventive alerts 7 to 8 minutes before exposure to critical levels, demonstrating early warning capacity. The contribution of this study is the formulation of a new hybrid system combining AI and the Internet of Things, which incorporates modeling of synthetic physiological systems in the field of proactive occupational health and safety. In the future, the focus of the work will shift to pilot tests in real work environments to adjust the system to sensor noise, variations in physiology, and environmental parameters.

Instituciones: Universidad Continental