Estimation Method to Determine Shipment Size of Iron Ore Trade in Maritime Transport

Maritime transport data released by governments, consultants, and other organizations are essential in preparing various logistics plans and policies. In international cargo flow data pertaining to ports, aggregate data, such as annual and monthly statistics of imports and exports are commonly publi...

Full description

Saved in:
Bibliographic Details
Published inJournal of the Eastern Asia Society for Transportation Studies Vol. 14; pp. 975 - 986
Main Authors KOSAKA, Hiroyuki, TEZUKA, Takenori, ARATANI, Taro
Format Journal Article
LanguageEnglish
Published Eastern Asia Society for Transportation Studies 11.03.2022
Subjects
AIS data
Deep learning
Matrix balancing
Shipment size
Trade statistics
Online AccessGet full text

Cover

More Information
Summary:Maritime transport data released by governments, consultants, and other organizations are essential in preparing various logistics plans and policies. In international cargo flow data pertaining to ports, aggregate data, such as annual and monthly statistics of imports and exports are commonly published. However, certain port authorities and data distribution agencies publish detailed data, such as shipment sizes (payload) of ships calling at ports. In this study, we estimated the shipment sizes using machine learning based on detailed iron ore trade data and automatic identification system (AIS) data. Additionally, we developed a method to adjust the estimates by applying a matrix-balancing technique to improve the accuracy of the obtained estimates.
ISSN:1881-1124
DOI:10.11175/easts.14.975