AI for Resource Mapping
The construction industry needs to make better use of the existing building stock. That requires an overview of what buildings are made of and the condition of their materials.
In this pioneer project, we investigated whether mobile scanning technology, point cloud generation, clustering, and AI-based segmentation can deliver exactly that — providing the data foundation needed to work more resource-efficiently with existing buildings.
Project period: October 2025 – June 2026
In collaboration with:
Digitalisation as a Key Enabler of the Green Transition
The construction industry faces a necessary transition in which transformation and renovation should be prioritized over new construction. This requires a much greater focus on preserving and reusing building components and materials rather than buying new ones. To succeed in this transition, it is essential to have structured data and an accessible overview of the existing building stock. This data holds the answers to questions such as: What materials does the building contain? Where are the materials located in the building? What are their dimensions? And what condition are they in? At the same time, it's important to bring this information into the decision-making process early. With this information, we will be able to make the right decisions, realize the reuse potential, and become more resource-efficient.
With this as a starting point, ConTech Lab joined forces with LINK Arkitektur, Christian Kongsgaard ApS, GreenDozer, Erik, Geopartner, dj-mg, COWI, Circue, the City of Copenhagen, OK, Dansk MiljøAnalyse, Konvika, AFRY, and Next Dimension to explore how new technology can make resource mapping more efficient and data-driven.
Mandatory Requirements for Selective Demolition Increase Demand for Digital Solutions
As of July 1, 2025, selective demolition has become a legal requirement for all buildings with a removed floor area of 250 m² or more. The requirement means that materials must be identified, separated, and sorted so they can be reused or recycled, and that environmentally hazardous substances such as asbestos, PCB, and PFAS are handled correctly. The purpose of the law on selective demolition is to reduce resource waste and support a circular economy in construction.
The construction industry is currently in the process of adapting to these new legal requirements and finding effective ways to comply with them. This creates a need for new working methods and forms of collaboration across stakeholders — and it triggers a significant demand for tools that can ease this transition. In particular, there is a need for mobile solutions that can quickly and intuitively create an overview of a building's materials and deliver structured data for use in planning, design, and documentation.
What is "comply or explain"?
– It's a political principle, and the starting point is that buildings should be preserved rather than demolished.
What does it mean in practice?
– If you want to demolish, you must be able to document that it's the right choice.
Why is it relevant for ConTech Lab and this project?
– Through initiatives such as AI for resource mapping, ConTech Lab is working to give the construction industry concrete methods to map, assess, and reuse building materials, so that "comply or explain" can be translated into action.
The Pioneer Project Builds on Important Industry Experience and a Prototype
The pioneer project builds on important industry experience and a prototype. Expert knowledge of environmental and resource mapping, in-depth understanding of user needs, and existing prototype solutions helped form the foundation for this joint pioneer project. The pioneer project built further on LINK Arkitektur's work with ReUseX. ReUseX laid the groundwork and shaped a back-end solution that allows users to scan and analyze the reuse potential of building components and materials via a mobile phone.
"ReUseX is the result of several years of dedicated work using AI to promote sustainability and circular economy in the construction industry. By applying advanced technologies such as LiDAR scanning and AI-based segmentation, ReUseX makes it quick and easy to map reusable building components in a property, optimize the utilization potential, and thereby reduce the amount of construction waste."- Jan Buthke, LINK Arkitektur.
"ReUseX is the result of several years of dedicated work using AI to advance sustainability and circular economy in the construction industry. By applying advanced technologies such as LiDAR scanning and AI-based segmentation, ReUseX makes it quick and easy to map reusable building components in a property, optimize the utilization potential, and thereby reduce the amount of construction waste," - Jan Buthke.
Video: Christian Pauli Bro
AI for Resource Mapping
In this pioneer project, ConTech Lab, together with partners including LINK Arkitektur, is exploring how mobile scanning technology, point cloud generation, clustering, and AI-based segmentation of building components and materials can support a more efficient and data-driven approach to resource mapping. The project's objective is to identify the specific needs of environmental and resource mapping specialists as well as project coordinators, and to explore how these needs can be supported through a digital tool that streamlines workflows, reduces errors, and enables greater scalability.
A digital tool for resource mapping has the potential to free up valuable time for specialists, who currently experience significant time constraints and often become bottlenecks in the process. This would enable the industry to move away from workflows that are either 1) highly manual and therefore time-consuming, or 2) technology-intensive and heavily dependent on specialist expertise. The ambition is to transition towards a more automated and digital process that supports the principles of the circular economy and contributes to a more sustainable built environment.
Project overview
In the pioneer project, ConTech Lab, together with LINK Arkitektur and others, investigated how mobile scanning technology, point cloud generation, clustering, and AI-based segmentation of building components and materials can create a more efficient and data-driven approach to resource mapping.
The purpose of the project was to identify concrete needs among environmental and resource mapping specialists as well as coordinators, and to find a good way to support them with a digital tool that streamlines the work, minimizes errors, and enables scaling of the work process.
A digital tool to support resource mapping work will free up time for specialists, who currently experience time constraints and become a bottleneck in the process. This will allow the industry to move away from work processes that are:
- manual and therefore time-consuming
or - technology-heavy and dependent on specialists
The aim is to transition to a more automated and digital process that supports the principles of circular economy and more sustainable construction.
Workshops set the direction for the further work
Along the way, we have held several workshops with a broad group of participants from the industry, and several of these threads now live on in other forums. This has had a significant impact on our understanding of how data in ReUseX should be structured so that it becomes as useful as possible for as many users as possible.
In the videos below, you'll get insight into what AI can contribute and an understanding of the challenges that exist today.
Video: ConTech Lab.
Interview with: Dan Skovgaard Jensen, ConTech Lab and Povl-Filip Sonne-Frederiksen, LINK Arkitektur
Video: ConTech Lab.
Interview with: Stine Kolding Pedersen, Københavns Kommune and Jan Buthke, LINK Arkitektur
What results did the project deliver?
The project delivered the following concrete outputs:
- Requirements specification — documentation of user needs, technical requirements, and regulatory requirements for documentation, data structure, and traceability
- Further development of the ReUseX platform — a user-friendly solution that combines mobility with structured data collection and reporting. The platform integrates scanning, point cloud generation, clustering, and AI-based segmentation with a front-end for classification and registration of metadata, including dimensions, material types, and condition. A speech-to-text functionality for quick metadata entry in the field was tested as an innovation element.
- An earlier version of the tool is available on GitHub
Three key insights
The project confirmed a strong and concrete industry need for digital solutions that can increase transparency in resource mapping and shift the work process from manual to automated. Three perspectives in particular stood out clearly:
Standardization and scaling
One of the project's ambitions has been to make data more usable. Jan Buthke from LINK Arkitektur points out that the decisive leap doesn't lie in the scanning technology itself, but in the subsequent structuring:
"What really determines whether data can be used across projects isn't the scan itself, but how the result is structured. When a building scan is converted into a consistent, semantically tagged resource inventory, it can be reused — for the mapping report, for selective demolition, and for further design work." – Jan Buthke, LINK Arkitektur.
The value is broadly distributed
We entered the process with the consultant as the primary user, but along the way we have seen that the interest is broader. Clients request the documentation early, and contractors benefit from an accurate picture of what is actually in the building before work begins. What ties this together is that data becomes accessible and isn't locked to whoever did the scanning. This has sharpened our understanding that the product must be shareable across stakeholders, not just live with one party.
Practice-oriented user involvement
The clearest learning point for us is how much early involvement matters. What looks right on screen doesn't necessarily hold up in the field. Jan Buthke describes what has specifically moved the project forward:
"It's especially the work on user flow and needs analysis that has moved us forward — by talking to those who will use the solution in practice, we've gained a clearer picture of what's actually needed and how the workflow needs to fit together. That's a big part of why we've chosen to keep some choices open and validate them empirically in the upcoming pilot projects rather than locking them in in advance."
Want to know more?
Are you curious about the project, or do you have questions? Contact innovation director at ConTech Lab, Dan Skovgaard Jensen, here. You can also keep an eye on upcoming events on the page here or by signing up for our newsletter.
