Towards Data-Centric Geospatial Applications: A Knowledge Graph Approach to Road Maintenance in Stockholm, Sweden

Abstract

Background
Digital twins offer great potential for transforming urban management to be more effective and sustainable. By adopting a data-centric approach, cities can become more innovative, moving away from application-centric solutions with data silos toward an interoperable modern data architecture. This thesis designs and develops a data-centric knowledge graph-based tool for road maintenance together with the Stockholm Traffic Administration Office.
Methodology and System Architecture
The research established a workflow to develop a standardized data-centric model to manage the road space. By analyzing the current operational and asset register systems at the Stockholm Traffic Administration Office, architecture was developed for a tool that accessed similar data content. The tool was constructed integrating geospatial road data into a CityGML-3CIM data model as a knowledge-graph, which was linked with the Swedish National Road Database (NVDB). This semantic approach facilitates data description and interoperability, through established standards. The data model was specifically classified to align with existing road maintenance requirements and the open-source software QGIS was utilized for the tool’s spatial visualization and management.
Results
The results demonstrate how a data-centric approach turns data into a central resource in the organization, a "Single Source of Truth" eliminating data silos and facilitating internal and external interoperability. Any update to the knowledge graph is reflected instantly across the organization, maintaining high data quality and removing the need for manual resource-heavy data conversion typical of application-centric solutions. The research offers several recommendations for how the organization can adopt a data-centric approach to road maintenance and concludes that further work on the topic is recommended and essential for the digital transformation of urban management.

Popular Abstract

Breaking Data Silos for Future Smart Road Networks
Every day our cities generate considerable amounts of data about the streets we drive, cycle, and walk on. What if we could use this information to repair potholes before they cause trouble, or to better coordinate winter road maintenance. That is what my thesis set out to achieve. Today urban data is usually trapped inside isolated systems in different city departments. Because these systems cannot talk to each other, maintaining our streets becomes uncoordinated and slower, also more expensive. This project aims to change that by showing how breaking down these digital boundaries can pave the way for smarter, more effective and sustainable city management.
To address this problem, I collaborated with the Stockholm Traffic Administration Office to design a new, shared digital system that puts data right at the heart of the organization. Instead of keeping information locked away in separate silos, as they do today. I built a smart digital model that acts as a single shared intelligent network for the city's roads. To do this, I took complex geographic data from local streets and connected it with national road databases. Instead of treating data as just a rigid list of numbers, this new model understands how things are connected in the real world, how a specific road, sidewalk or intersection relate to one another. To ensure it is practical for city planners, I integrated this entire system into an open-source mapping tool.
The results of this work demonstrate that when data becomes a shared central resource, it can transform how a city operates. Any update made to a road's status is instantly reflected across the entire organization. This reduces the need for manual time-consuming and resource-heavy data conversions typical of latency systems. By removing these digital boundaries we can ensure that every department works with the same information, allowing cities to manage their infrastructure more effectively and deliver better, safer roads for everyone in a connected future.

• author: Wahlberg, Johan
• supervisor: Lars Harrie
• organization: Department of Earth and Environmental Sciences (MGeo); Dept of Physical Geography and Ecosystem Science
• course: GISM01 20261
• year: 2026
• type: H2 - Master's Degree (Two Years)
• subject: Earth and Environmental Sciences
• keywords: Digital twins, data-centric, knowledge graphs, CityGML, 3CIM, road data
• publication/series: Master Thesis in Geographic Information Science
• report number: 213
• language: English
• id: 9233394
• date added to LUP: 2026-06-09 08:56:02
• date last changed: 2026-06-09 08:56:02