Last summer, three project proposals received €10,000 DigiShape seed money to further develop and test their idea. With DigiShape Seedmoney, promising ideas are given an impulse to be further developed and tested, so that it becomes clear what is needed for the next step towards application, collaboration and upscaling. The projects focus on data-driven spatial development of the North Sea, hybrid modelling of salt intrusion and groundwater levels.
SeaScape: an AI-queryable marine use map
Information House Marien, MARIS and Nelen & Schuurmans will work on an AI-enabled marine use map. In this map, various marine data sources are combined into one integral and understandable spatial image of the North Sea.
The sea use map is a demonstration case to show how large datasets can be better queried by AI Agents. With this use case, the project partners show what valuable data the Marine Information House has available and how it can be made more accessible to and by AI.
Contact person: Jermo Janmaat
More information on the project page
Hybrid Modeling of Salt Ingress
HKV, TU Delft, Rijkswaterstaat and the Port of Rotterdam Authority are investigating how physical models and AI can be combined to predict salt intrusion faster and more reliably.
The project is working towards a hybrid model: known physical processes form the basis, while AI is used to fill in the missing knowledge in the model. The intended result is an open-source model for a Dutch estuary (e.g. the Rhine-Meuse delta).
New project partners are welcome to join.
Contact person: Gijs Hendrickx
More information on the project page
Tailor-made groundwater
BZ Ingenieurs & Managers, Deltares and the University of Amsterdam are working on a methodology that can be used to generate reliable groundwater levels at any location that are suitable for geohydrological and dike strength analyses.
Groundwater analyses depend on monitoring well data. These groundwater monitoring networks can have limited coverage and expanding them is expensive and sometimes not feasible. AI and Machine Learning can offer a solution, but this requires a high quality of the underlying methodology and training data.
Within the seed money project, a solid methodology is being developed, which can serve as the basis for a follow-up project in which a Proof of Concept is actually developed.
Contact person: Karel Hagemans
More information on the project page
Continuation
The three project teams will start working on their seed money project in the coming period. We will soon speak to the project leaders individually about their plans: what are they going to do now, what do they want to achieve during this phase and where can the DigiShape community contribute?
