Exploring the drivers of digital transformation in Chinese port and shipping enterprises: A machine learning approach.
Journal:
PloS one
PMID:
40323911
Abstract
With the transition to a global green low-carbon economy, the urgency for digital transformation in the port and shipping industry has become increasingly prominent in making enterprises more efficient and sustainable. This study focuses on how Chinese port and shipping enterprises, which are key carriers for global containerized trade, can attain digital transformation as a means to tackle environmental challenges and improve competitiveness. Using a representative sample of 83 A-share-listed companies (2008-2023) and employing several modeling techniques, such as Ridge regression, LightGBM, and XGBoost, we investigate a data-driven approach with the support of the Technology-Organization-Environment (TOE) framework. We find that nonlinear models (LightGBM, XGBoost) outperform linear models and emphasize the importance of a supportive environment for green finance. We further perform a number of sensitivity and robustness checks toensure the validity of our findings. These insights provide actionable guidance for policymakers and industry leaders seeking to harmonize digital innovations with green development.