Risk factors identification and injury severity classification in Alaska’s mining industry using statistical and machine learning approaches
Document Type
Article
Publication Date
1-2025
Department
Department of Geological and Mining Engineering and Sciences
Abstract
This study examines safety-related factors influencing injury severity in Alaska’s mining industry using workers’ compensation data. The injury severity was characterized by claim type (medical only and lost time). Statistical analyses, including chi-square tests and logistic regression, identified significant associations between claim type and factors such as age group and cause of injury, while mine type, gender, body part injured, and nature of injury were found to be non-significant. Logistic regression revealed that older adults (OR = 2.76) and young adults (OR = 1.78) had higher odds of severe injuries, with strain injuries as the most frequent cause. Machine learning models were developed to classify injury severity using these factors, with logistic regression demonstrating the most consistent performance on test data (average score: 0.62). The findings emphasize the importance of targeted safety interventions for high-risk groups and prevalent injury causes to enhance workplace safety. While this study provides actionable insights for safety management, it is constrained by the limited scope of six safety-related factors and a relatively small dataset. The approach adopted in this study offers a framework for developing safety management models applicable to other mining operations and industrial contexts.
Publication Title
International Journal of Mining, Reclamation and Environment
Recommended Citation
Chatterjee, S.,
Kadrolli, P.,
Kaunda, R.,
Miller, H.,
&
Majdara, A.
(2025).
Risk factors identification and injury severity classification in Alaska’s mining industry using statistical and machine learning approaches.
International Journal of Mining, Reclamation and Environment.
http://doi.org/10.1080/17480930.2025.2459238
Retrieved from: https://digitalcommons.mtu.edu/michigantech-p2/1462
