ADDISON DING EMANG Universiti Malaysia Sarawak (UNIMAS)
This study presents the development of a Predictive Model for Environmental Risk Index (PMERi) using machine learning for Small and Medium Enterprise (SME) welding workshops. Welding environments are inherently hazardous, exposing workers to multiple simultaneous risks such as excessive heat, high noise levels, poor air quality, elevated humidity, and inadequate lighting. These factors do not act independently; instead, they interact in complex, non-linear ways that significantly increase the likelihood of occupational health issues. However, conventional Occupational Safety and Health (OSH) practices remain largely reactive and often assess environmental factors in isolation, limiting their effectiveness in preventing long-term health risks.
To address this limitation, PMERi introduces a proactive, data-driven framework that integrates five key environmental parameters into a single, actionable risk index. The model is powered by a Random Forest algorithm, chosen for its robustness in handling non-linear relationships and multidimensional data. Environmental data were systematically collected at 10-minute intervals using calibrated sensors in a controlled workshop environment. Through advanced feature engineering, the original five variables were expanded into 39 derived features, enabling the model to capture complex environmental interactions more accurately. The model’s reliability and generalizability were validated using both 5-fold cross-validation and time-series cross-validation techniques.
The developed PMERi system enables early identification of hazardous conditions, supporting timely interventions such as optimizing local exhaust ventilation systems or enforcing appropriate personal protective equipment usage. By transforming traditional safety practices into predictive and preventive strategies, PMERi enhances workplace safety and operational efficiency. Furthermore, this research aligns with the United Nations Sustainable Development Goals, particularly SDG 3 (Good Health and Well-being), SDG 8 (Decent Work and Economic Growth), and SDG 9 (Industry, Innovation, and Infrastructure). Ultimately, PMERi offers a scalable, cost-effective solution that empowers SMEs to adopt intelligent safety systems in line with Industry 5.0 advancements.