PREDICTION SEAFARER TURNOVER BY USING THE RANDOM FOREST ALGORITHM.

The maritime industry is crucial for global trade and relies significantly on the commitment and expertise of seafarers, who encounter many difficulties throughout their journeys. This study examines the influence of environmental workplace characteristics on the rates at which seafarers leave their jobs, acknowledging the significance of comprehending how these aspects affect the ability to keep seafarers in their positions. The main objective of this study is to determine the crucial factors that influence seafarers' choices to exit the maritime business, specifically by applying the Random Forest algorithm to forecast turnover. This study is based on a quantitative examination of data obtained from 230 seafarers. Out of these, 200 valid responses were assessed. The study investigates the environmental factors that influence the rates at which seafarers leave their profession. A correlation analysis was performed to investigate the association between these factors and turnover. The Random Forest method was utilized to construct a predictive model, discerning the most influential factors contributing to seafarer turnover. The results indicate a robust and statistically significant association between environmental conditions and the rate at which seafarers leave their jobs. The Random Forest model found job satisfaction, autonomy, and work-life balance as significant indicators that strongly influence turnover. These observations provide the foundation for creating focused interventions aimed at decreasing turnover rates in the maritime industry. Resolving the issue of seafarer turnover necessitates a collaborative endeavour including many entities in the maritime industry, such as shipowners, operators, and regulatory authorities. The study suggests the development of policies and programs that give priority to the well-being of seafarers, decrease stress levels, and improve working conditions in order to encourage seafarers to stay at sea for extended periods of time. This research provides significant suggestions for enhancing staff management and retention tactics in the marine industry by utilizing the predictive capabilities of the Random Forest algorithm. [ABSTRACT FROM AUTHOR]