Research Projects

HyProTwin

Infrastructure projects require access to large volumes of technical inventory data. However, this data is often unstructured, scattered across various IT systems, or available as digitized paper documents. Manual research is time-consuming and ties up valuable specialist expertise. 

HyProTwin is developing a hybrid, AI-based digital twin that intelligently links information from diverse data sources and provides it in a process-oriented manner – for more efficient planning, operation, and maintenance of infrastructure. 

  • Project start: February 2025
  • Runtime: 3 Years
  • Funding amount: > €2 million
  • Number of partners: 3
Objectives

Digital twins and semantic data integration
The goal of HyProTwin is to create digital twins of infrastructure structures such as locks, weirs, and other water-related facilities. Existing data – such as plans, maintenance records, or historical documents – is linked using AI, clearly organized, and made easily searchable. 

Innovation approach: Hybrid AI for more efficient information processing
In this project, the partners are combining two AI methods within an infrastructure context:  

◾ ️Symbolic AI (ontologies, semantic data models) for structuring and standardizing domain knowledge 

◾ Subsymbolic AI (machine learning, NLP, OCR, large language models) for automatic analysis and extraction of information from documents 

This combination enables intelligent analysis and provision of relevant data to stakeholders in infrastructure management. 

Reducing the effort required to search for information 
A central goal of the project is to reduce the significant amount of time spent searching for inventory data. The AI-powered platform is designed to help users find the information they need faster and significantly streamline work processes. 

Objectives

Digital twins and semantic data integration
The goal of HyProTwin is to create digital twins of infrastructure structures such as locks, weirs, and other water-related facilities. Existing data – such as plans, maintenance records, or historical documents – is linked using AI, clearly organized, and made easily searchable. 

Innovation approach: Hybrid AI for more efficient information processing
In this project, the partners are combining two AI methods within an infrastructure context:  

◾ ️Symbolic AI (ontologies, semantic data models) for structuring and standardizing domain knowledge 

◾ Subsymbolic AI (machine learning, NLP, OCR, large language models) for automatic analysis and extraction of information from documents 

This combination enables intelligent analysis and provision of relevant data to stakeholders in infrastructure management. 

Reducing the effort required to search for information 
A central goal of the project is to reduce the significant amount of time spent searching for inventory data. The AI-powered platform is designed to help users find the information they need faster and significantly streamline work processes. 

Objectives

Digital twins and semantic data integration
The goal of HyProTwin is to create digital twins of infrastructure structures such as locks, weirs, and other water-related facilities. Existing data – such as plans, maintenance records, or historical documents – is linked using AI, clearly organized, and made easily searchable. 

Innovation approach: Hybrid AI for more efficient information processing
In this project, the partners are combining two AI methods within an infrastructure context:  

◾ ️Symbolic AI (ontologies, semantic data models) for structuring and standardizing domain knowledge 

◾ Subsymbolic AI (machine learning, NLP, OCR, large language models) for automatic analysis and extraction of information from documents 

This combination enables intelligent analysis and provision of relevant data to stakeholders in infrastructure management. 

Reducing the effort required to search for information 
A central goal of the project is to reduce the significant amount of time spent searching for inventory data. The AI-powered platform is designed to help users find the information they need faster and significantly streamline work processes. 

Viereth weir facility (Viereth weir, dismantling of the roller gate by BAW, licensed under CC BY 2.0 Attribution 2.0 Generic Deed)
Project Approach

AI integration into existing data infrastructures 
HyProTwin uses state-of-the-art AI technologies to automatically capture data from existing systems, link it, and consolidate it into a digital twin. A structured database ensures that the solution can be flexibly expanded. 

Application & Demonstrator 
The project is developing a HyProTwin demonstrator that serves as a digital information tool. Using an intelligent search function, users receive targeted answers to technical questions – based on a structure’s digital twin. 

In this process, structured and unstructured data are analyzed together and provided in context. 

Case Study: German Waterways 
Using the example of German waterways, where many facilities have been in operation for decades, HyProTwin demonstrates how AI can support maintenance, analysis, and planning. Through data that is better linked, measures become easier to plan and decisions more substantiated. 

Benefits & Added Value 

▪️Faster workflows 
Less time spent on research, faster planning, and better decisions.

▪️Better overview 
All information is consolidated and easy to find.

▪️Wide range of applications 
The solutions developed can also be applied to other areas. 

Project Approach

AI integration into existing data infrastructures 
HyProTwin uses state-of-the-art AI technologies to automatically capture data from existing systems, link it, and consolidate it into a digital twin. A structured database ensures that the solution can be flexibly expanded. 

Application & Demonstrator 
The project is developing a HyProTwin demonstrator that serves as a digital information tool. Using an intelligent search function, users receive targeted answers to technical questions – based on a structure’s digital twin. 

In this process, structured and unstructured data are analyzed together and provided in context. 

Case Study: German Waterways 
Using the example of German waterways, where many facilities have been in operation for decades, HyProTwin demonstrates how AI can support maintenance, analysis, and planning. Through data that is better linked, measures become easier to plan and decisions more substantiated. 

Benefits & Added Value 

▪️Faster workflows 
Less time spent on research, faster planning, and better decisions.

▪️Better overview 
All information is consolidated and easy to find.

▪️Wide range of applications 
The solutions developed can also be applied to other areas. 

Viereth weir facility (Viereth weir, dismantling of the roller gate by BAW, licensed under CC BY 2.0 Attribution 2.0 Generic Deed)
Project Approach

AI integration into existing data infrastructures 
HyProTwin uses state-of-the-art AI technologies to automatically capture data from existing systems, link it, and consolidate it into a digital twin. A structured database ensures that the solution can be flexibly expanded. 

Application & Demonstrator 
The project is developing a HyProTwin demonstrator that serves as a digital information tool. Using an intelligent search function, users receive targeted answers to technical questions – based on a structure’s digital twin. 

In this process, structured and unstructured data are analyzed together and provided in context. 

Case Study: German Waterways 
Using the example of German waterways, where many facilities have been in operation for decades, HyProTwin demonstrates how AI can support maintenance, analysis, and planning. Through data that is better linked, measures become easier to plan and decisions more substantiated. 

Benefits & Added Value 

▪️Faster workflows 
Less time spent on research, faster planning, and better decisions.

▪️Better overview 
All information is consolidated and easy to find.

▪️Wide range of applications 
The solutions developed can also be applied to other areas. 

Project Partners

Contact for the Project
Jan Luca Siewert
Research Scientist

HyProTwin is being developed by CONTACT Software in collaboration with the German Federal Waterways Engineering and Research Institute (BAW) and the Institute for Numerical Methods and Informatics in Civil Engineering (IIB) at the Technical University of Darmstadt. The project is supported as part of a nationwide funding program for digital innovations in the infrastructure sector.