Zustandsprognose von Ingenieurbauwerken auf Basis von digitalen Zwillingen und Bestandsdaten

Research output: Contribution to specialist publicationFeature article/contribution (Feuilleton)Contributedpeer-review


  • Hubert Naraniecki - , MKP GmbH (Author)
  • Robert Hartung - , RWTH Aachen University (Author)
  • Steffen Marx - , Institute of Concrete Structures, TUD Dresden University of Technology (Author)
  • Katharina Klemt-Albert - , RWTH Aachen University (Author)


Condition state prediction of engineering structures based on digital twins and inspection data. The condition state of engineering structures is significantly important for the maintenance and safe operation of the transport infrastructure. Knowledge of the current condition state of a structure is therefore crucial to economic maintenance management. In addition to the current condition state, economic maintenance strategies are based on knowledge of the future development of a structure's condition. The future condition development of a structure can be estimated by an intelligent data linkage of bridge data and thus maintenance measures can be derived in advance. The paper presents an approach to derive data driven predictions of the condition states of railway bridges based on Building Information Modeling (BIM) and digital building models, linked with inspection data and Structural Health Monitoring (SHM). For this purpose, Machine Learning (ML) methods are applied to the aggregated data of the bridge structures of the DB Netz AG and evaluated in an asset related manner. The objective and data driven prediction provides a further basis for reliable decisions in maintenance management.


Original languageGerman
Issue numberDOI: 10.1002/bate.202100100
PublisherErnst & Sohn [Berlin]
Publication statusE-pub ahead of print - 10 Dec 2021
No renderer: customAssociatesEventsRenderPortal,dk.atira.pure.api.shared.model.researchoutput.ContributionToPeriodical

External IDs

ORCID /0000-0001-8735-1345/work/142244469



  • Building Information Modeling, Eisenbahnbrücken, Entscheidungsbäume, Machine Learning, prädiktive Instandhaltung, shBIM, Structural Health Monitoring, Zustandsprognose