ETA prediction

Predicting the ETA of a container vessel based on route identification using AIS data

Master thesis (2017)

Authors

R. Meijer Technology, Policy and Management

Contributors

Y. Tan (supervisor 1)
A.M. Herdeiro Teixeira (supervisor 2)
V.E. Scholten (supervisor 2)
F. Fanitabasi (coach)
Faculty
Technology, Policy and Management
Copyright: © 2017 Ricardo Meijer
Machine Learning Prediction AIS ETA
More Info
expand_more

Published Date

22-08-2017

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Technology, Policy and Management

Abstract

Container vessels arriving in a port before or after their scheduled time can cause problems in the container terminal planning and planning of hinterland transportation. This in turn leads to an increase of the costs in the supply chain. Vessels communicate their Estimated Time of Arrival via Automatic Idetification System(AIS) data to the port. This arrival time is estimated by the crew of the vessel and manually inputted into the AIS. In this research a proof of concept is shown that the Estimated Time of Arrival (ETA) prediction of container vessels can be improved. Vessels en route to the Port of Rotterdam are used as a case study. Different frameworks and algorithms are introduced to improve the data quality of AIS messages, to identify a set of possible routes and to do predictions based on the set of possible routes. It is possible to do predictions based on pre-processed AIS messages and a set of possible routes that perform at the same level as the best guess of a vessel’s crew.