Analysis of Financial Stability in the Shipping and Logistics Industries

Purpose: This study evaluates the financial resilience of Korea’s shipping and logistics industries using machine learning models and SHAP analysis to identify key financial variables. Research design, data, and methodology: Financial data from 100 companies (2001-2023) were analyzed using LightGBM, XGBoost, Random Forest, Linear Regression, and LSTM. SHAP analysis highlighted the importance of variables affecting Tobin's Q. Results: According to the result, LightGBM achieved the best performance for both industries. Equity ratio emerged as the most significant variable, with debt management metrics prioritized in shipping and operational efficiency in logistics. Conclusions: This paper revealed that tailored strategies focusing on debt management for shipping and efficiency improvement for logistics are essential. This study highlights the utility of machine learning and SHAP for financial resilience analysis.