Internship | Graphs Infrastructure Renewal

Position: Internship | Knowledge Graphs for Infrastructure Renewal
About this position

This internship offers the opportunity to contribute to innovative research at the intersection of Knowledge Graphs and emerging technologies, with the goal of improving data-driven decision-making in complex infrastructure environments. You will work on challenges related to integrating and structuring information from diverse data sources, enabling organizations to make better-informed, more transparent, and explainable decisions throughout the asset management lifecycle.

As part of a multidisciplinary team, you will investigate how Knowledge Graphs can be combined with complementary technologies such as machine learning, large language models, computer vision, or geospatial analytics to create intelligent solutions for real-world problems. The position combines scientific research with practical experimentation, including literature studies, data analysis, prototype development, and evaluation. Throughout the internship, you will have the freedom to explore a research direction that matches your interests and expertise while contributing to the development of innovative methods for knowledge integration, decision support, and infrastructure intelligence.

Ideas are welcomed, and you have the space to develop your expertise.

What will be your role?

During this internship, you will work on exploring and developing methods to integrate Knowledge Graphs with technologies to address real-world infrastructure renewal challenges. You will have the flexibility to focus on a specific research direction that aligns with your interests and academic background.

Research directions:

AI Agents and Knowledge Graphs Explore how LLM based agents can use Knowledge Graphs to carry out infrastructure assessment tasks, such as finding and combining relevant information, using existing engineering rules and calculations, identifying missing information, and explaining assessment results.

Advanced Engineering Reasoning and Knowledge Graphs Explore how Knowledge Graphs can support increasingly detailed engineering assessments by incorporating more expressive rules, calculations and degradation models. Investigate how existing domain ontologies, engineering standards and asset specific parameters can be reused to support assessments such as degradation analysis and remaining service life estimation.

Machine Learning and Knowledge Graphs Explore how machine learning can use connected information in Knowledge Graphs to discover hidden relationships, degradation patterns, and similarities between assets that may not be captured by existing engineering rules and assess how these insights can support renewal decisions.

Computer Vision and Knowledge Graphs Explore how information extracted from (inspection) images, such as cracks and deterioration, can be integrated into Knowledge Graphs and combined with engineering knowledge to support infrastructure condition and renewal assessments.

Geospatial Data and Knowledge Graphs Explore how geospatial data and spatial relationships can enrich Knowledge Graphs to understand assets in their network and environmental context and support infrastructure renewal decisions.

This work will involve conducting a literature review to explore existing approaches in the chosen research area, analyzing real-world infrastructure data to identify information gaps and integration challenges, and designing methods that combine Knowledge Graphs with the selected technology. The research will also include developing and testing a prototype using relevant tools, frameworks, and datasets, evaluating its practical applicability, explainability, and value for decision support, and documenting key findings, lessons learned, and recommendations for future research and…