Breakthrough Algorithm Promises to Reshape Architecture with Faster, More Efficient Gridshell Design
Table of Contents
A new computational method developed by researchers at the University of Tokyo and a U.S.-based structural engineering firm is poised to revolutionize the design of large, lightweight structures, particularly gridshells. The innovation dramatically reduces processing time, making complex designs more accessible to architects and engineers.
A growing demand for sustainable and visually striking architecture is driving interest in gridshells – curved, thin surfaces constructed from an interconnected grid of structural members. Traditionally built from reinforced concrete, modern architects are seeking alternatives due to the material’s cost, environmental impact, and limited transparency. Gridshells, utilizing materials like metal, glass, or timber, offer a compelling solution, capable of spanning vast areas without internal supports.
The Rise of Gridshells in Modern Architecture
Gridshells are ideally suited for expansive public spaces, offering column-free environments in locations such as train station entrances, historic courtyard restorations, and public squares. Iconic examples include the Great Court of the British Museum, the glass roof of the Dutch Maritime Museum, and New York City’s Moynihan Train Hall. Despite these successes, the complexity of designing and constructing gridshells has historically limited their widespread adoption due to a lack of efficient computational tools.
To address this challenge, Masaaki Miki of the University of Tokyo and Toby Mitchell from Thornton Tomasetti collaborated to create a new algorithm capable of identifying optimal gridshell shapes that balance structural integrity with complex geometries.
Overcoming Computational Hurdles in Gridshell Design
Previous attempts at computational gridshell design faced significant limitations. While earlier systems utilizing NURBS (Non-Uniform Rational B-Splines) surfaces – a standard in computer-aided design (CAD) – showed promise, they struggled with highly irregular shapes and required excessive computing power. A task that once demanded 90 hours on a high-end graphics processing unit (GPU) now takes approximately 90 minutes on a standard central processing unit (CPU) thanks to the updated method.
“The project began in 2020 with an interest in shell structures, often made of concrete,” one researcher explained. “Traditional designs aim for shapes that carry their own weight entirely through the force of compression, but this limits how expressive or sculptural they can be. We set out to find new ways to design shells that consider forces of compression as well as tension, allowing greater design freedom.”
NURBS: The Key to Speed and Accuracy
The new method’s strength lies in its direct integration with NURBS surfaces. Unlike mesh-based approaches that rely on thousands of triangular pieces, NURBS provide smooth, continuous, and mathematically precise representations of curved forms. This compatibility with existing architectural design software streamlines the workflow. The research team even developed a plug-in for Rhinoceros, a popular NURBS-focused CAD program, enabling architects to utilize the approach within a familiar environment.
The algorithm represents stress distribution on a NURBS surface and employs newly developed algorithms that achieve a 98% increase in processing speed. This advancement eliminates the need for expensive GPUs, making advanced gridshell form-finding accessible to a broader range of professionals. The resulting gridshells are demonstrably stable and suitable for construction using metal and glass.
“Because we are addressing a real-world problem, we have been rigorously validating our solutions by several test methods we also developed,” a senior official stated. “When the tests revealed failures in the method, it was stressful. However, we are now totally happy because all solutions pass the tests.”
Future Outlook: Expanding to Timber Gridshells
While the current research concentrates on metal-and-glass gridshells, the team intends to extend the technique to encompass composite timber gridshells in the future. This expansion would further broaden the application of the technology and contribute to more sustainable building practices.
This research received partial funding from the Nomura Foundation, the JSPS Grants-in-Aid for Scientific Research (KAKENHI; grant number 23K17784), and JST ASPIRE (grant number JPMJAP2401).
