Cutting Edge Real Estate Analytics

Leading British universities like the Royal College of Arts or University College London are using the unique API of a Swiss PropTech startup to enter new fields with their research. As do leading Swiss financial institutions. Data sharing and strategic partnerships like these support scientific development and at the same time help to accelerate the advancement of real estate technologies.

Data for Behavioral Analysis of Architecture

(Source: Archilyse)

How to optimally divide office space so that users feel comfortable? Is a specific hospital floor plan design supporting the healing process of the patients?

Dr Kerstin Sailer, Reader in Social and Spatial Networks at the Bartlett School of Architecture at University College London, devotes her research to investigate such questions. Sailer analyses, for example, how hospital floor plans play a role in “shaping and channeling, distributing movement, bringing people together, allowing people to talk and communicate”.

Supported by Marie Müller, psychologist, neuroscientist and currently PhD student at the Faculty of Brain Sciences at University College London, Sailer’s work helps to understand the correlations of spatial configurations and human behavior and even allows to predict certain experiences. More and more researchers globally focus on theses topics, and their progress is enormous.

Sailer and Müller need on the one hand 3D spatial analysis, and, on the other hand, data about behavior and experience. In collaboration with the Swiss based startup Archilyse, Sailer and Müller now have access to 3D spatial analysis data, that they did not have before. The third dimension, which was not possible to analyse previously, allows them to push the boundaries of their field. Sailer and Müller now investigate how much their models gain with this new perspective. Sailer is building up this study on her paper on the British Library in London, in which she correlated two dimensional spatial configuration with user behavior already.

To answer the question at hand, the unique algorithms of Archilyse are helpful as they deliver data for all three dimensions for the first time. Therefore Archilyse calculates exactly what, and how much of it, one can see at every point in a building or how the specific daylight conditions are throughout the year. Also, if spaces are rather exposed or enclosed in general.

Data for Gaining Urban Knowledge

(Source: Archilyse)

“Data obtained through computational and statistical tools promises to revolutionise the way we understand our domestic and work environments as well as the design and policy decisions that emerge from them.” Dr Sam Jacoby.

Dr. Adrian Lahoud, Dean of the School of Architecture at the Royal College of Arts, and Dr Sam Jacoby, Royal College of Art School of Architecture Research Leader, were initiators of the Laboratory for Design and Machine Learning which is currently working on a pilot study to analyse housing interiors and their spatial organisation with the aid of machine learning. Seyithan Özer, is a PhD student working in the lab and dedicates his research to methodologies and methods of production of architectural and urban knowledge, technology, and governance.

He uses machine learning algorithms (e.g. also applied in asset management) to study the designs of existing housing stock in London in order to establish an empirical basis upon which discussions on space standards and design manuals, can be grounded.

The analysis of the existing housing stock allows an in-depth understanding of the differences, repetitions, and the range of spaces in use in comparison to the existing policy definitions and the underlying assumptions on plan configurations, distribution of uses, and room shapes. Özer’s study focuses on the relationships of dimensions and areas to related design decisions. This entails testing and evaluating different methods of classification, and ultimately contributes to an inclusive housing design guidelines.

For his analysis Özer uses the API provided by Archilyse to generate new categories of forms, layouts, and floor plan typologies and tests them. Archilyse digitised a large set of London apartments and analysed them. The analytical data that Archilyse provided automatically included: total floor area and basic footprint dimensions (minimum bounding rectangle) of dwelling unit, floor area, basic dimensions (minimum bounding rectangle), circumference, centroid, number of windows with total window width, number of kitchen and bathroom elements and the number of staircases for each room, number of doors together with their centroids, and the number of doors between each room.

Data for Floor Plan Quality Assessment

But Archilyse doesn’t only deliver data for academia: “We do hundreds of analyses at the click of a mouse”, explains Matthias Standfest, CEO of Archilyse.

Credit Suisse used Archilyse’s analytical information for its annual study on the Swiss Real Estate Market which, this year, was dedicated to the quality of floor plans. The economists of Credit Suisse teamed up with the Swiss startup which used its algorithms to quantify and compare aspects like brightness or furnishability. The hyperdimensional features that Archilyse calculates are used to make more precise statements about the quality of a property and thus to achieve greater accuracy and precision with regard to price estimates.

The FUTURE PropTech in London in May 2019 which Archilyse is going to join is the perfect ground to find about cooperation potentials.

Margarete Sotier is responsible for marketing and PR at and editor of the blog

Related Events