AI can turn floor plans into 3D housing models. What does it take?
10
October
2025
5/5 (1)
I read “Automatic Reconstruction of Semantic 3D Models from 2D Floor Plans” by Cambeiro Barreiro et al. (2023). The goal they had was to turn scanned plans into vector 3D models that you can export as IFC, a building data format. It lets different BIM (Building Information Modeling) tools share the same data, like walls, rooms, and doors, etc. This matters especially for big real estate portfolio holders, like housing associations. These parties would be able to gather insights from the vast amounts of data, which are normally only available on paper.
The core problem
Symbols and techniques differ, and old drawings are clumsy. Windows and doors break through walls. The scans are biased. Redrawing by hand is labour-intensive, expensive, and error-prone. The stakes are higher for large portfolio holders, such as housing associations. They oversee thousands of residences. They are unable to compare layouts, energy features, or risks across the stock in the absence of structured data.
How it actually works
You begin with floor plan scans or PDFs. Like a human, the system reads the plan. It recognises windows, doors, and walls. It corrects gaps and crooked lines. After that, it creates clean vectors from these shapes. Lastly, it exports IFC and creates a basic 3D model. Common BIM tools can use that file. You can then measure areas, tag units, and count rooms. Additionally, you can link maintenance notes or energy data. Structure is crucial. You can perform the same checks on buildings in the portfolio at large scale once the data is in a standardised format.
source: Cambeiro Barreiro et al., 2023
Why portfolio owners should care
Data is locked in drawers by paper plans. They become structured BIM thanks to this pipeline. This opens up the portfolio’s unit mix, wall types, window counts, and layout features. It facilitates large-scale renovation planning, accessibility assessments, and energy audits. You receive a cleaner handoff to designers and contractors and expedited due diligence.
Takeaway
Digitising plans is the first step if you own a sizeable housing portfolio and would like to digitize. Segment walls, identify symbols, and carefully translate everything using floor-plan AI. To prevent lock-in and maintain the data’s utility across tools, export to a common data type like IFC. The improvements are easy to understand in terms of speed, consistency, and insight at the portfolio level.
Reference: Cambeiro Barreiro, A., Trzeciakiewicz, M., Hilsmann, A., & Eisert, P. (2023). Automatic Reconstruction of Semantic 3D Models from 2D Floor Plans.
AI in Housing Valuations: Make it explainable
19
September
2025
No ratings yet.
AI can now help to set property values. What are the risks involved?
I came across this paper called Enhancing Explainable AI Land Valuations Reporting for Consistency, Objectivity, and Transparency. By Y. Yim and C. Shing (2025). The paper goes into detail on how explainable AI can ethically support the valuation of properties. This is a sensitive topic, since the valuation of properties affects many parties, like banks, buyers, sellers and the cities. There are major efficiency gains to be made by implementing artificial intelligence and machine learning in this sector. However, their integration also raises legal and ethical questions.
The Core problem
Many models act like a black box; this undermines the duty of the valuer to deliver transparent and consistent valuations. The legal system requires the properties to be thoroughly inspected. and the process to be well-documented.
A possible solution: Making AI explain itself
There are 3 pillars to ethically implement these innovations while integrating these technologies. Consistency: The model should provide repeatable results, following the same process. Objectivity: There should be a clear separation between the developers and validators of a model. Transparency: the model, data and limits should be well documented and easy to understand.
XAI tools like SHAP can be used to make the AI explain itself. These tools show how each feature/variable pushes the price up or down. The chart shown below ranks drivers such as zoning, building age, and floor area. This turns a score into a story that a client can follow.
“The visualisation sets a foundational baseline value (E[f(X)]) of 13.98 on the x-axis, representing the model’s average prediction when no specific feature information is provided. This is the expected value if we were to make a prediction without any additional information. The Output Value (f(x)), which in this instance is 14.07, reflects the actual prediction after accounting for the cumulative effect of the individual features. The colour-coded bars represent the push and pull of each feature on the prediction: the red bars show the features that contribute to an increase in the predicted land value, while the blue bars indicate a decrease.“
Takeaway
AI can increase efficiency and scale valuations. But it must earn trust. Build on three pillars: consistency, objectivity, and transparency. Use SHAP for explanations. Ship reports with a clear checklist and keep human judgement in charge.
References: Yiu CY, Cheung KS. Enhancing explainable AI land valuations reporting for consistency, objectivity, and transparency. Land. 2025;14(5):927. https://www.proquest.com/scholarly-journals/enhancing-explainable-ai-land-valuations/docview/3212060417/se-2. doi: https://doi.org/10.3390/land14050927.
Technology of the Week – The Housing Industry
5
October
2017
5/5 (3) The video below describes how online platforms revolutionized the housing industry and the way in which house owners, house buyers, and tenants connect with each other in the Dutch housing market:
Group 45: Rosanne Baars, 406184 ; Roy Ouwerkerk, 459406 ; Yuxin Sun, 406080 ; Pieter Vreke, 372189
1. History
The first real estate brokers in the Netherlands arose in 1284. They acted as the connecting party between trading partners. They earned money from commissions. In the home-rental industry, homeowners initially connected with tenants via physical notes in public spaces. In the 19th century housing corporations and social housing emerged. Consequently, private homeowners struggled to find tenants and started using the intervention of brokers, who received a fee for every contract signed (Van den Elzen, 2013).
2. Current Situation
Decades later, the rise of two-sided online brokerage platforms completely changed the way in which homeowners and tenants communicate. The emergence of these platforms weakened the role of offline brokers, which has several benefits:
For online brokerage platforms, the physical infrastructure and assets that offline brokers use is no longer needed.
Building and scaling networks became cheaper.
Homeowners have access to a larger customer base.
Tenants have access to a larger number of houses and it is easier for them to compare, due to more transparency.
Transaction costs decreased, since most of the physical communication is replaced by online communication.
These eventually led to a decrease in effort and time needed for the rental process. However, for the process of buying/selling a house, offline brokers still coexisted along with online platforms, because buying a house has a large impact on people’s lives, which increases one’s willingness to pay (Bloomberg, 2013).
3. Platform Properties
Current housing platforms have several properties:
A triangular structure, composed of four parties: Demand side users: Tenants or buyers Supply side users: Homeowners Platform providers: Online platforms/communities Platform sponsors: Technology providers Platform providers and platform sponsors are mostly employees the same company (Eisenmann, Parker, & van Alstyne, 2009)
Strong cross-side network effects. A large number of house-owners offering houses on a website attracts tenants, and vice versa.
Subsidies for either the demand or supply side of the platform, while charging the other side. In this way, more users are attracted and network effects increase. The reason why the same part of the platform is not charged consistently, is that different platforms target different niches with different willingness to pay.
Interoperability; many platforms redirect demand side users to related platforms.
Targeting of niches, who have needs for different features.
Low homing costs. Subscription fees are reasonably priced and currently reducing.
4. Future Expectations
Housing platforms are expected to change in the following way:
The number of housing platforms is expected to keep increasing due to existence of niches and low homing costs niches exist and homing costs are low. At the same time, population growth, internationalization and increasing transparency may lead to an increase in housing rental as opposed to buying property (Independent, 2016).
Platform’s profit margins may increase, since increasing demand and decreasing supply for housing rental might lead to increased willingness to pay of demand side users (De Volkskrant, 2014). However, increasing competition among platforms might drive margins down. These two effects can eventually cancel each other out.
Smart-home devices will increase efficiency for both landlords and tenants (Independent, 2016). This will make property management easier, since it might eliminate physical communication and provides more information.
Increasing use of big data analytics. For example, Housing Anywhere is experimenting with this. (Statsbot, 2017).
Convergence of the house rental and real estate industries, because house buyers might get more comfortable with online approaches (Harvard Business Review, 2016).
Companies from adjacent markets may envelop incumbents.
References
Bloomberg. (2013, March 8). Why Redfin, Zillow, and Trulia Haven’t Killed Off Real Estate Brokers. Retrieved from Bloomberg.com: https://www.bloomberg.com/news/articles/2013-03-07/why-redfin-zillow-and-trulia-havent-killed-off-real-estate-brokers
De Volkskrant. (2014, November 2014). De kloof met de Randstad is niet meer te dichten. Retrieved from De Volkskrant: https://www.volkskrant.nl/binnenland/de-kloof-met-de-randstad-is-niet-meer-te-dichten~a3780161
Eisenmann, T., Parker, G., & van Alstyne, M.W. (2009). Opening Platforms: How, When and Why? Platforms, Markets and Innovation, Gawer, A. (ed.), Northampton, MA: Edward Elgar, 131-162.
Harvard Business Review. (2016, November 17). Real (estate) disruption: how technology may change the housing market. Retrieved from Harvard Business Review: https://rctom.hbs.org/submission/real-estate-disruption-how-technology-may-change-the-housing-market/
Independent. (2016, August 10). How technology could revolutionise the future of renting. Retrieved from Independent: http://www.independent.co.uk/money/how-technology-could-revolutionise-the-future-of-renting-smart-meter-landlord-bills-a7182306.html)
Rabobank. (2017). Rental housing: Rising prices in a high-potential market. Retrieved from Rabobank: https://www.rabobank.nl/bedrijven/cijfers-en-trends/vastgoed/real-estate-report-2017/sub-markets/rental-housing
Statsbot. (2017). Housing Anywhere discovered the best way to share data across a team and help them stay on track with key metrics. Retrieved from Statsbot: https://statsbot.co/customers/housinganywhere
Van den Elzen, W. (2013). The future of the Dutch housing corporations.
