The tracking of ships' movements around the globe has become substantially easier over the past decade through the use of data from the Automated Identification System (AIS) - position data that are sent out by all oceangoing ships that can be picked up by satellites. However, AIS data has several shortcomings: they are incomplete in space and time, they can be manipulated, the AIS transponder can be turned off, and there is no information about the type of cargo onboard. The objective of the SHIPTRACK project is to investigate and demonstrate how other types of satelite data can be used to supplement and complement data to increase the accuracy and information that can be extracted about global ship and cargo movements. We are particularly interested in optical satellite data (i.e. pictures of the earth's surface) and radar satellite data that measure the height of objects and landscape formations. As an example, we can envisage that optical data can be used to identify the type of commodity stored in a particular port terminal, which can then be used as input to identify the cargo that is loaded or discharged from a ship calling at the terminal. Radar satellite data, on the other hand, can potentially be used to measure the height of port stockpiles of commodities such as coal over time, facilitating the tracking of supply and demand dynamics of the commodity. The SHIPTRACK project is a multidiscplinary co-operation between researchers at SNF AS at the Norwegian School of Economics in Bergen, Norway and UiT Arctic University of Norway, Narvik, as well as startup vake.ai, shipowners Utkilen and Western Bulk, and the Bergen Shipowners' Association.
Ship tracking data from the Automated Identification System (AIS) has several known quality issues: incomplete coverage in space and time, often erroneous data, and ease of manipulation. Moreover, AIS contains no information about cargo type, the true cargo volume/weight onboard, or the storage situation in the port. These challenges create the need for new knowledge in three main areas: Firstly, to identify and utilize alternative data sources that can be fused with AIS ship position data in order to fill in the gaps. Secondly, to develop novel methods to remotely identify the cargo types stored or loaded/discharged in a particular terminal. Thirdly, to track changes in stored volumes in wet and dry bulk ports using remote sensing technology. The SHIPTRACK project is a co-operation between shipping researchers at NHH/SNF, satellite data experts at the UiT Arctic University at Narvik, remote sensing startup VAKE.ai, shipping companies Utkilen and Western Bulk, and the Bergen Shipowners' Association. The project is highly interdisciplinary and aims to develop new knowledge from fusing satellite (optical and radar) and AIS data. The project intends to develop novel methods for the automated tracking of ships (during gaps or faulty AIS transmissions) and the extraction of metadata such as speed and loading condition from optical satellite data. Combined with new methods for the classification of commodity types and monitoring of stored volumes (both liquid and dry bulk) using remote sensing, this creates a very powerful tool for increasing visibility in the commodity supply chains and bulk freight markets. In terms of impacts, the project results contribute to better spatial resource allocation for bulk commodities, for instance by reducing uncertainty in restocking decisions, demand estimates and transportation planning. Improved ship tracking also facilitates safer and more efficient transportation and the prevention of potential of illegal and harmful activities.