Work-Life Conditions as the Primary Determinant of Seafarer Mental Health: An Explainable Machine Learning Analysis

Area of Science:

• Maritime Safety and Occupational Health
• Applied Machine Learning
• Psychological Determinants

Background:

• Seafarer mental health is critical for transportation safety and operational efficiency.
• Traditional methods struggle with complex, non-linear factors influencing seafarer well-being.
• There is a need for advanced analytical tools to understand and predict psychological states in maritime professionals.

Purpose of the Study:

• To develop and validate an explainable machine learning (ML) framework for predicting seafarer psychological states.
• To identify and quantify the key determinants impacting seafarer mental health.
• To provide interpretable insights for targeted mental health interventions in the maritime industry.

Main Methods:

• A survey of 500 seafarers was conducted, collecting data on 12 risk factors.
• A Random Forest classifier was employed for prediction, optimized using Bayesian hyperparameter tuning.
• SHapley Additive exPlanations (SHAP) were utilized for feature importance and interaction analysis.

Main Results:

• Current work-life conditions were identified as the most significant factor influencing seafarer psychological states.
• Favorable environmental conditions enhance the positive impact of career development and social recognition.
• High onboard service pressure negatively affects mental health, irrespective of other positive factors.

Conclusions:

• Systemic improvements in onboard living and working conditions are foundational for effective psychological interventions.
• Tailored support strategies should follow environmental enhancements for optimal impact.
• The study offers a data-driven, interpretable framework for precision mental health management in maritime operations.