How Data Drives the Future of Design

Data has become the currency of modern society. It is the most abundantly generated product of the 21st century. Every action in our lives, from asking for directions using Google Maps to liking a post on social media, produces data that is being mined in a variety of imaginative and profitable ways.

If our daily actions generate an avalanche of information, how much data could the design, construction, and operation of a building produce? Sketches and drawings, simulations and building analyses, BIM models, construction logistics and procurement, post-occupancy data gathered by sensors, and 3D scans all produce an abundance of data. It is, therefore, unfortunate that the adoption of Big Data and Cloud Computing in the building industry is substantially less developed than in other fields.

To review more details about the applications mentioned in this article, please refer to the article “Big Data and Cloud Computing for the Built Environment” by the authors as part of a Springer book Industry 4.0 for the built environment.

Data for performance-oriented design

One of the domains in which the power of these technologies could be leveraged is performance-oriented design.

The requirement to deliver projects that are not only driven by aesthetic criteria but also by hard metrics in relation to their structural, environmental, or social performance has given rise to the question of how time-consuming analyse could productively enhance the design process. Additionally, could the knowledge and data accumulated during a project be used to inform the next one?

These challenges have been a catalyst to investigate how other industries are dealing with similar issues. The widespread adoption of Big Data and Cloud Computing has enabled a truly dramatic increase in their performance over the past few years.

Hydra generates myriads of design options

Foster + Partners’ Hydra is an in-house cloud platform for accelerating computational and performance-driven design. Its ongoing development has been driven by real-world scenarios and large-scale masterplans. Since the latest prototype launched in 2019, Hydra has generated 240,000 design options and run over 1.3 million performance simulations.

One such study was the Guangming Hub, a competition-winning project for a new Transport Oriented Development situated on the high-speed rail link in China. The challenge was to find an optimal combination of buildings based on site-specific typologies while maximizing environmental performance, cutting down average walking times, and increasing views of greenery.

Rather than manually creating and analysing thousands of options to find the best performing, the requirements and design rules were fed to Hydra, which did the heavy lifting. Hydra created 10,000 options and evaluated each one based on simulations while evolving the better-performing solutions.

Design pipelines and workflows will change

Running entirely in the cloud, Hydra reduces computing time from months to hours. It provides the design team with a powerful tool for making data-driven decisions which can influence the project at its most critical stage.

Systems like Foster + Partners’ Hydra also allow for large amounts of data to be collected, processed, structured, and classified in a meaningful manner, ready for further reuse. This is a unique opportunity to rethink current design pipelines and workflows, which in turn could result in the ability to deliver high-quality designs at a much faster pace.

The images are courtesy of Foster + Partners

Authors

Marcin Kosicki

Marcin is an Associate at Foster + Partners in the Applied Research and Development group where he is currently spearheading the development of Hydra – an inhouse cloud platform for computational and performance-driven design. He has also provided expertise in complex geometry, digital fabrication, and machine learning for many high-profile projects all over the world. Marcin graduated both from the Faculty of Architecture of Wroclaw University of Technology, Poland, and The Bartlett School of Architecture, UCL receiving an MSc in Architectural Computation. He is a published author of several research papers as well as a tutor and lecturer at The Bartlett in London.

Marios Tsiliakos 

Marios is a Design Systems Analyst and Associate Partner at the Applied Research and Development Group at Foster + Partners, expanding the boundaries of applied computation and problem solving for the AEC industry in large-scale design projects. Within ARD his interests lie in the domains of complex geometry, data or performance-driven design, and interoperability, focusing on the development of Hermes, F+P’s in-house interop tool. For more than a decade, he’s been teaching extensively in the UK and Europe, namely at the Bartlett-University College London, the Architectural Association, and the Institute for Experimental Architecture in Innsbruck, Austria. His research has been published in peer-reviewed conferences and his generative design work has been exhibited and presented around Europe, Asia, and North America.  

Khaled El-Ashry

Khaled’s research interests combine Architecture and Computer science. Currently works as an Associate Partner at the Applied Research and Development (ARD) group at Foster + Partners in addition to tutoring at the Bartlett school of architecture, University College London. He holds a master’s degree in Adaptive architecture and computation from University College London. His recent work bridges the gap between architecture practice and research with expertise in optimization, fabrication, robotics, virtual reality, and data visualization. He recently worked on multiple state-of-the-art architecture projects like the Mobility Pavilion in the UAE expo, The Lusail stadium in Qatar, and the UAE pavilion in the Milan expo in addition to his research in real-time robotic control.

Martha Tsigkari

Martha Tsigkari is a senior partner at Foster + Partners where she runs the Applied Research and Development Group. Her background spans architecture, engineering, and computer science. She has two decades of experience working on projects of all scales and uses. Martha’s work for the Applied Research and Development group incorporates computational design, human-computer interaction, machine learning, and optimization. She has investigated using deep neural networks and genetic algorithms in the design process, aiming to solve problems ranging from passively actuated micromaterials to performance-driven urban layouts.