In recent years, the global marine logistics industry has changed significantly because of the influence of the global movement of goods; this situation has increased the importance of developing ships that meet market requirements. One such change is the exponential growth in the amount of available data, and attention paid to big data analysis in a variety of fields. It is now possible to obtain vast amounts of marine logistics data, e.g., port, ship, route, international trade, and automatic identification system data. If these data are effectively utilized, great innovation can be achieved in the marine logistics industry.

In this study, we develop a ship allocation model that can predict the demand for bulk carriers and examine the effective principal particulars of ships for cargo transportation. To realize these goals, we develop three distinct models—shipper, shipowner, and operator models—using statistical, hierarchical, and deep learning analysis methods. Moreover, we examine the principal particulars of ships that are expected to be in demand to demonstrate the effectiveness of our proposed model.