Impact of Fireworks Industry Safety Measures and Prevention Management System on Human Error Mitigation Using a...

Area of Science:

• Industrial Safety and Health
• Accident Prevention
• Machine Learning Applications

Background:

• Human error (HE) is a primary cause of accidents and explosions in the fireworks industry (FI).
• Identifying and mitigating human factors (HFs) is crucial for preventing workplace incidents in hazardous environments.
• Past disasters in the fireworks and match industries highlight the urgent need for improved safety measures.

Purpose of the Study:

• To investigate the impact of human factors contributing to human error in fireworks industry disasters.
• To analyze the correlation between unsafe behaviors and influential factors in hazardous chemical warehouse accidents.
• To develop a machine learning-based prediction model and an expert system for mitigating human error risks.

Main Methods:

• Utilized machine learning (ML) techniques, including Support Vector Machine (SVM), Random Forest (RF), Naïve Bayes (NB), and Extreme Gradient Boosting (XGBoost).
• Collected and analyzed questionnaire data from 500 participants in the fireworks and match industries in Tamil Nadu, India.
• Compared the accuracy of various ML models to identify the optimal predictor for human error.

Main Results:

• Extreme Gradient Boosting (XGBoost) achieved the highest classification accuracy at 94.41%.
• Significant correlations were found between individual behaviors and the likelihood of human error.
• The study identified physical, psychological, and organizational factors influencing human error in the fireworks industry.

Conclusions:

• The proposed expert system (ES) effectively mitigates human error risks in the fireworks and match industries.
• Implementing robust safety management systems (SMS) is essential for preventing fatalities and injuries.
• The developed ML model provides a valuable tool for enhancing workplace safety in high-risk environments.