Computational Timber Structures for Zurich Airport
Interview with Preety Anand of BIG: BJARKE INGELS GROUP
Can you begin by describing your current role at BIG and how its focus on computational design, combined with your background in this area, attracted you to join the company?
I am working as Senior Architect and Computational Designer for the Zurich Airport project. My main task is to coordinate and develop the load bearing structure of the project, which gives me the opportunity to utilise my computational design skillset for developing the complex geometry of a timber structure.
The geometric playfulness and complexity of BIG’s projects together with their progressive attitude towards design technology are key reasons for my interest in the office, as there will be many opportunities for implementing my skillset.
How does the computational design team collaborate with the broader architectural and engineering teams at BIG?
The Computational Design team at BIG operates as a group of full-time specialists under the wider Design Technology umbrella.
In their respective branch office, each specialist assists project teams with training and day-to-day computational design support, which can range from a quick script to help a designer speed up a repetitive manual task, to extensive collaborations with the project team and external consultants on complex geometry management or masterplan development.
As a global unit, the Computational Design team collaborates on the development of relevant in-house automation and analysis tools, and keeps up a knowledge base and script library.
For the Timber Zurich Airport project you’ll present at CDFAM, could you share the thought process behind adopting computational design methods in timber construction?
During the competition and schematic design phase, Computational workflows were implemented for parametrically modeling the complex geometry of the Dock, Root and Skylight. These included controlling the overall massing, and creating the base surface geometry of the root skylight and the overall load bearing timber frame structure, amongst others.
Having computational workflows has allowed for the creation of different design iterations as well as more overall geometric precision and control, which allows us to precisely develop the project’s geometric design intent – such as the varying inclination of the timber members as well as the alignment of the timber members at the joints.
Having developed the timber frame structure parametrically has also allowed us to create stick models for structural analysis. As 3D models can be machine-read, utilising this geometric data seamlessly for processes of fabrication and assembly is an essential added benefit.
Given the extensive use of timber, optimized through automated fabrication processes like CNC cutting. How does this emphasis on timber and automation reflect BIG’s approach to sustainability and construction efficiency?
The use of timber, a renewable resource, helps to reduce the carbon footprint of construction projects. Automated fabrication processes like CNC allow for greater precision and less waste, further improving the sustainability of the project. This approach aligns with BIG’s philosophy of combining innovative and ambitious design with pragmatic efficiency and environmentally conscious practices.
The project utilized Grasshopper for its development, from the slabs to the roof panels. How does the integration of software tools, particularly transitioning from Rhino to Revit, impact the workflow and management of data and what other software tools were employed in the project?
The building volume has been the base for developing the geometry and articulation of the envelope and the timber structure. Similarly, interior building elements are largely dependent on the articulation of the timber structure with the beam and column inclinations as driving factors. An example of this dependency is the slab outline on each floor, as it is related to the v-column angle it spans between.
Similarly, the roof panel geometry is not only a result of the varying roof beam angles, but also of their grid spacing. There are a number of examples in the interior fit out as well, such as the vertical angle of the glazing panels along the v-columns, which is dependent on the varying column angle along the dock.
For implementing this design intent, a literacy of computational design tools within the team is necessary.
As these building elements are parametrically modelled using Rhino Grasshopper, the question of how to transfer these elements to a documentation and collaboration platform obviously comes up.
Throughout this process, data management plays a key role, especially if the connection between Rhino and Revit is seen as a live-link and not as a one-off connection. This live-link is then not only useful for transferring geometric data, but also for gradually adding data for element identification.
In an ideal scenario, we aim to place and control native Revit elements using Rhino.Inside Revit by extracting key geometric data such as insertion and work point coordinates from our Rhino model.
Updating element data instead of replacing entire instances keeps the GuID intact and thus all hosted elements are kept. From a modeling point of view, both platforms demand developing different approaches. While each member of the timber v-column structure is parametrically modelled individually, the family in Revit is set up as a complete unit consisting of columns and beams, which can be placed along each gridline. The varying geometric articulation is driven by the instance values of the work point coordinates.
In terms of software and tools, what gaps have you identified in current offerings, and how do you envision future tools addressing these needs to facilitate better data exchange and project execution?
Parametrically modelling complex geometry works best in Rhino using Grasshopper. The challenge is transferring the geometry to a documentation and collaboration platform.
The aspect of interoperability between both platforms has increasingly improved and expanded in terms of the tools and possibilities, reflecting the needs of the AEC industry. This facilitates documenting projects based on complex geometry with ease, whereas the tools ensure a wide accessibility.
Despite these developments, running Rhino in Revit is still occasionally unstable and requires frequent plugin updates.
In an ideal scenario, users should be enabled to parametrically model complex geometry intuitively within the documentation platform itself. This could potentially also help in overcoming Interoperability issues while exchanging data with external consultants. It would allow users to work more seamlessly and efficiently without having to use workarounds.
Finally, what do you hope attendees will learn from the project you will be presenting at the CDFAM symposium, and what are you looking to gain from participating in the event?
Computational thinking is the common language of all participants regardless of their professional background. I am hoping to contribute to this symposium by showcasing a large scale infrastructure project, which is benefitting from a meaningful application of computational design tools throughout different project phases.
On the other hand, I am also keen to see the breadth of computational thinking and application for design and fabrication in other fields. This event will bring the various members of the Computational Design community closer together and will enable a productive exchange.
