The Karamba plug-in developed by Lengfeld & Wilisch Architekten, Skylink, Frankfurt Airport, Germany, 2011

Free parameters define the locations along the bridge where sets of diagonals attach to
the main girders. This parametric arrangement allows specific geometry to be treated
Clemens Preisinger in collaboration as series of numerical vectors that can be fed into any optimisation algorithm.

with Bollinger + Grohmann Engineers

has been developed to predict the
behaviour of structures under external
loads. Intended to be used by architects
rather than being solely confined
to an engineering setting, it enables
a seamless flow of data between
structural and geometric models.
Preisinger here describes the program’s
evolution and application.

Article 16 Clemens Preisinger

Linking
Structure
and Para-
metric
Geometry
110
soma Architecture, White Noise, Salzburg, Austria, 2010 top right The irregular layout of truss diagonals emerged from a genetic
below: Straight aluminium rods of uniform length, arranged in a seemingly random algorithm that optimised for structural performance and efficiency of material.
fashion, form the load-bearing structure of the Salzburg Biennale music pavilion.
bottom right: Parameterisation strategy for the White Noise music pavilion.
Member connectivity and thus structural performance depend only on the
angles of the members with respect to a given reference surface.

Skylink

Karamba is a finite element (FE) program for predicting the The Skylink bridge links Frankfurt Airport with a 300-metre
behaviour of structures under external loads. Though developed (984-feet) distant car park. The work of Lengfeld & Wilisch with
in a structural engineering firm, its main focus does not lie Bollinger + Grohmann Engineers, it is a trussed bridge with the
in this field. It is geared towards use within an interactive, diagonal elements placed by the Karamba design tool.3 The
parametric architectural design environment. One of the goals four longitudinal girders at the corners of the bridge are fixed.
was to create a fast, lightweight tool that facilitates a seamless Two sets of connecting elements join the main girders: one set
flow of data between structural and geometric models. encloses the top, and the other encloses the bottom, resulting
The origins of the Karamba plug-in for Robert McNeel & in the box-like truss. Along the sides of the bridge, the diagonal
Associates’ Grasshopper® date back to a research project elements run in different planes so that complex, costly
entitled Algorithmic Generation of Complex Space Frames,1 connections are avoided. The parameterisation used consisted
carried out at the University of Applied Arts Vienna in of joining predefined poly-lines with sets of diagonals whose
collaboration with Bollinger + Grohmann Engineers. The project position was subject to optimisation.
focused on the viability of applying genetic algorithms (GAs) Maximum displacement and mass of steel served as
to the structural optimisation of real-world structures.2 GAs measures for ranking the solutions in the GA optimisation
rely on a large number of evaluations of the function to be procedure. The number of variables to be optimised amounted
optimised – in the case of structural assessment these are FE to about 400, which made it a very large-scale problem for
calculations. It was found that the main impediment for scaling a GA approach. Roughly 200,000 variants of truss layouts
up the optimisation tasks was the time spent on the recurrent were tested before arriving at the final result. The solution
computation of the statical models, which amounted to days in performs similar to a regular truss with respect to total mass
some instances. This motivated the implementation of a custom and maximum deflection. Although the truss layout arrived at is
FE code which now forms a large part of the calculation core probably not the global optimum, it certainly provides a useful
of Karamba. The parallelisation of time-consuming calculation alternative to conventional, regular geometries.
steps, a fast and extensive interface to scripting languages,
and results calculation on demand, have led to a reduction
of computation time of two to three orders of magnitude
compared with using off-the-shelf FE programs.

111
 soma Architecture, White Noise, Salzburg, Austria, 2010 Grasshopper definition of a structural model
bottom: A close inspection of the music pavilion reveals its underlying Karamba components add physical meaning to a parametric geometry setup.
structural principle: beams in parallel planes connect to each other via
circular studs depending on their relative position.

White Noise

Karamba has been used at Bollinger + Grohmann Engineers

for a number of projects, either for checking statical feasibility
in the early stages of design, or for performing structural
optimisation. A recent example of the latter is the design of
White Noise by the Austrian firm soma Architecture.4 A mobile
pavilion for cultural events, it consists of a large number of
aluminium rods that interact in a seemingly chaotic manner.
The static system of the pavilion is made up of a number
of arches that span a distance of 12 metres (39 feet). Each of
the arches is made of multiple layers of rods. The members
of one layer connect to elements on the neighbouring layers
via short, circular studs. The number of connections between
neighbouring elements thus depends on their inclination with
respect to each other. The optimisation task here involved
selecting the elements orientation for minimum displacement
under given loads at minimum total structural weight. The
parameterisation of the geometry was done entirely in
Grasshopper, and the solution process handled by its built-in
probabilistic optimisation engine.

112
Shells under horizontal load
Left to right: principal stress lines; colour plot of material utilisation; force flow lines; Grasshopper allows the integration of custom
displaced structure with colour plot of principal stresses.
plug-ins that can then make use of the
program’s graphical user interface elements
and thus interact with other plug-ins in a
well-defined way.

Integration of Karamba Within an

Interactive Modelling Environment

During the research project at the University of Applied Arts a stand-alone project. As development proceeds, the range of
Vienna, the design of a user interface for the program package its applications constantly widens; for example, shell elements
played a very minor role. However, in the day-to-day work can now be included in calculations. But what is perhaps most
of the design office at Bollinger + Grohmann Engineers, the interesting is how the structural properties that are normally
necessary coding of geometric parameterisations as scripts hidden from view – deflections, natural vibrations, force flow
proved cumbersome and reduced the range of possible users lines – can be used creatively in an architectural context. With
to a small circle of scripting experts. To solve this problem, the advent of new digital tools that combine architecture and
the program was ported into existing software that offered an engineering, demand has grown for professionals who have
interactive, generative modelling environment. Grasshopper knowledge in both areas. Karamba is the instrument for this
allows the integration of custom plug-ins that can then make new generation of architectural engineers and engineering
use of the program’s graphical user interface elements and architects to play their new tunes on. 2
thus interact with other plug-ins in a well-defined way. This led to
the first public version of the Karamba plug-in.
Notes
Most aspects of a Karamba model can be made 1. Klaus Bollinger, Arne Hofmann and Clemens Preisinger, Algorithmic Generation of Complex
dependent on parametric input. Such a model consists of an Space Frames, Institute for Architecture, Structural Design,
University of Applied Arts Vienna, 2010. Research project funded by the Austrian Research Fund
assemblage of visual components which look and feel like FWF under the translational research programme, grant L538.
the native software’s building blocks. The fact that Karamba 2. See David E Goldberg, Genetic Algorithms in Search, Optimization, and Machine Learning,
Addison-Wesley (Reading, MA), 1998, and JL Coenders and LAG Wagemans, ‘Engineered by
reacts immediately to any change of input parameters helps Evolution, Two-Dimensional Prototype Application of openStrategy Form Finding’, in Proceedings
to understand structural mechanisms in the design. Using of the IASS 2004 International Symposium Shell and Spatial Structures from Models to
Realization, 20–24 September, 2004, pp 66–7.
Karamba is like watching a film compared to a fixed image; 3. Mark Fahlbusch, Arne Hofmann, Alexander Heise, Klaus Bollinger, Manfred Grohmann
one can easily create a series of images in real time by and Jochen Mahlknecht, ‘Skylink am Flughafen Frankfurt’, Stahlbau, Vol 81, No 8, 2012, pp
638–42.
manipulating the corresponding user interface widget instead 4. Stefan Rutzinger, Kristina Schinegger, Arne Hofmann and Moritz Heimrath, ‘Adaptive Design
of the more traditional approach of having a systems response of a Music Pavilion’, Adaptive Architecture Conference, 3–5 March, 2011.

only at one particular state.

Text © 2013 John Wiley & Sons Ltd. Images: p 110 © Bollinger + Grohmann, photo Enrico
Although its development has been rooted in the context Santifaller; pp 111(tr&br), 113 © Bollinger – Grohmann – Schneider ZT GmbH; p 111(l) ©
of Bollinger + Grohmann Engineers, Karamba has always been Florian Hafele; p 112(t) © Clemens Preisinger; p 112(b) © Peter Eder

113