Current Volume 9
The petrochemical industry involves complex operations that pose significant occupational hazards, particularly during maintenance and shutdown activities. These operations expose workers to heightened risks due to the involvement of heavy machinery, hazardous chemicals, confined spaces, and dynamic work environments. Traditional safety management approaches often rely on reactive measures and historical incident analysis, which may not effectively anticipate emerging hazards. This study presents the development of a predictive assessment model aimed at identifying and mitigating occupational hazards in petrochemical maintenance and shutdown operations proactively. The model integrates multi-source data, including historical incident reports, operational parameters, environmental conditions, and workforce factors, to forecast the likelihood and severity of potential hazards. Utilizing advanced machine learning techniques, the model processes both qualitative and quantitative data to classify risk levels and predict hazard occurrences with improved accuracy. Feature selection highlights critical factors such as equipment type, duration of maintenance tasks, environmental variables (e.g., temperature, toxic gas presence), and human factors influencing hazard manifestation. Model validation was conducted using cross-validation techniques and real-world shutdown case studies, demonstrating significant predictive performance improvements over conventional risk assessment methods. The results identify key predictive factors and provide actionable insights to safety managers, enabling timely interventions and resource allocation to high-risk operations. This predictive framework offers a proactive tool for petrochemical facilities to enhance occupational safety, reduce incident rates, and minimize operational downtime during maintenance shutdowns. By transitioning from reactive to predictive hazard management, the industry can better safeguard worker health and optimize safety protocols. The study concludes with recommendations for integrating real-time sensor data and continuous learning algorithms to further refine predictive capabilities, ensuring adaptive and resilient occupational hazard management in petrochemical maintenance environments.
Predictive, Assessment model, Occupational hazards, Petrochemical maintenance, Shutdown operations
IRE Journals:
Cynthia Obianuju Ozobu
"A Predictive Assessment Model for Occupational Hazards in Petrochemical Maintenance and Shutdown Operations" Iconic Research And Engineering Journals Volume 3 Issue 10 2020 Page 391-402
IEEE:
Cynthia Obianuju Ozobu
"A Predictive Assessment Model for Occupational Hazards in Petrochemical Maintenance and Shutdown Operations" Iconic Research And Engineering Journals, 3(10)