On the Dutch waterways, data and models provide more and more insight into shipping movements, waiting times, energy consumption and incidents. Within the DigiShape Open Source Platform (D-OSP) project, Deltares, Witteveen+Bos, Van Oord, TU Delft and Rijkswaterstaat worked on open source building blocks that make these insights more usable. We spoke to Fedor Baart, senior advisor shipping models at Rijkswaterstaat, about the benefits of D-OSP.
“D-OSP was a broad project, with different work packages, each tackling its own issue,” says Fedor. “What connects them is that they are all taking a technological step to work better together with data, models and open source code. My involvement was mainly with Captain Hindsight, Water and Logistics and Digital Twin Waterways.”
Fedor knows D-OSP from two sides: first, he worked on the development of solutions from Deltares. In his new position at Rijkswaterstaat, he is mainly looking at how the developed products can be applied in the practice of waterway management. “That makes my role in this interview a bit special, but it is very nice to see that what I have helped develop myself really offers added value in practice.”
Captain Hindsight: Rewinding What Happened on the Water
Fedor was Captain Hindsight’s work package leader. “In it, we worked on better tools to reconstruct shipping accidents and other events on the water,” he says. “You want to be able to look back with data: what happened, where did the problem start and what can you learn from it as a waterway manager?”
For Rijkswaterstaat, this helps to make waterways ‘forgiving’. “Just like with road design, you want someone to be able to make a mistake without it having immediate major consequences.” As an example, he cites the ship that ran aground against the weir in Borgharen in 2024. “When we rewound the situation, it turned out that the ship had been in trouble at the bridge before. So then you don’t just look at the place where it eventually went wrong, but also at the moment when the problems started. In this case, that meant inspecting the bridge as well.”
Time is no longer a neglected issue
Technically, Captain Hindsight is mainly about the combination of time and geographical data. “In standard GIS, time has always been a bit of a neglected child,” Fedor explains. “Spatial aspects are well worked out and time often came after that. Within Captain Hindsight, we looked at how you can make better use of time-dependent, moving objects. A ship is first here and then there. You want to be able to capture that well and look back.”
To do this, the project worked with moving features, a standard for moving objects that change position over time. Within D-OSP, this approach has been tested with large AIS datasets. “That standard was already under development and we put it to the test with difficult shipping data. That turned out to work well. In QGIS you can now make standard animations with time.”
This will also make this way of working more widely usable than just for shipping. Fedor also used the same approach, for example, to map out how the Dutch lock network has come into being over time, and how the maintenance task is also growing as a result.
D-OSP built on existing open source work, including by Anita Graser on moving features in QGIS. “A lot was already ready. We tested the approach with specific AIS data and added extra functionality for shipping, such as calculating the collision angle between two ships.”
During the DigiShape workshop in October 2025, participants tested for themselves whether they could work with it. They reconstructed an accident in which a ship collided with the Baltimore Bridge. Fedor: “That actually went pretty well. Then you see that the approach not only works for developers, but is also transferable to a wider group of users.” A nice side effect was that the architecture developed in the Infrastructure and Mobility work package could also be used in part.
Water and logistics: from model to usable calculation
Within the Water and Logistics work package, work was done on open source models for shipping and logistics, such as OpenTNSim and OpenCLSim. “Those models already existed, but we have made them more applicable in parts,” says Fedor.
An example is the energy module. Previously, you could only use it as part of a simulation: you modelled shipping behaviour and calculated energy consumption within it. “Now you can also apply the same energy calculation to data you already have, for example about ships that have actually sailed through a waterway. Without having to build the entire model first.”
Components have also been taken further around locks and port logistics, such as the lock tool, links with the sea lock formulation and applications around the port accessibility tool. “This allows us to better investigate why ships wait: for the tide, for a place to moor, for diesel prices or for something else.”
Digital Twin Waterways: calculating the network as a whole
While Water and Logistics helps to better analyse specific situations, the Digital Twin Waterways work package is in line with the question of what this means for the waterway network as a whole. “With many issues, you have to be able to switch between the level of detail and the network picture,” says Fedor. “You want to be able to look at a lock or queue, but also at the route behind it. What happens if a connection fails? What detour options are there? And what does that mean for travel time, costs or energy consumption?”
“In this work package, we worked with the EURIS excavation, a digital network structure that can be used to calculate European waterway connections. European countries have made their waterways available digitally, but that does not automatically mean that you can calculate with them. Sometimes two countries supply the same stretch of river because that river is on the border. Sometimes small connections are missing. Within D-OSP, work has been done to tie these pieces together better.”
This makes it possible to better analyze routes and connections. “Now you can sail all the way from the Netherlands to the Black Sea digitally, taking into account factors such as depth, energy consumption, waiting times and costs at the touch of a button.”
Building on open source
Rijkswaterstaat will continue the results of D-OSP in practice. “We support OpenTNSim and OpenCLSim as an experimental model and actively participate in it ourselves. For example, by further improving schematizations, putting data in order or testing parts in projects. For example, in the lock module at the Houtrib locks, we briefly rotated that module to test whether it works properly for that purpose. This way you apply the model for additional calculations and you immediately learn what is still needed.”
For Fedor, D-OSP mainly shows what happens when parties tackle small technical barriers together. “They were often challenges that were already on the shelf,” he says. “Too small for a large project, and at the same time very useful to tackle properly. I think the great thing about the DigiShape approach is that as experts you solve things together that make you think: why wasn’t this arranged yet? As a sector, we can build on that for years to come.”
“D-OSP touches on three barriers that recur in many digital collaborations: trust, interoperability and shared infrastructure. Because we work with various DigiShape partners on concrete work packages, we prevent each organization from solving the same technical puzzles separately. This creates open source code, data and models that the sector can build on more broadly.”
– Gert-Jan Schotmeijer, project leader D-OSP
The results of D-OSP will be made more accessible via DigiShape in the coming period. The project page lists the work packages, public sources and first links to tools and datasets. The lessons from D-OSP are also shared with the community during DigiShape meetings.
