See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/304658467

Combinatorial Equilibrium Modelling

Conference Paper · September 2015

CITATIONS READS

4 353

2 authors, including:

Patrick Ole Ohlbrock

ETH Zurich
14 PUBLICATIONS   23 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Creative Statics View project

3D vector-based graphic statics View project

All content following this page was uploaded by Patrick Ole Ohlbrock on 01 July 2016.

The user has requested enhancement of the downloaded file.

 Proceedings of the International Association for Shell and Spatial Structures (IASS)
Symposium 2015, Amsterdam
Future Visions
17 - 20 August 2015, Amsterdam, The Netherlands

Combinatorial Equilibrium Modelling

Patrick Ole OHLBROCK*, Joseph SCHWARTZa

* ETH Zurich, Institute of Technology in Architecture, Chair of Structural Design

Stefano-Franscini-Platz 5, 8093 Zurich, Switzerland
ohlbrock@arch.ethz.ch

a
ETH Zurich, Institute of Technology in Architecture, Chair of Structural Design

Abstract
The present research develops a theoretical framework (based on Graphic Statics and the Graph Theory)
to incorporate the control of qualitative structural behavior in the conceptual design phase in a novel
way. This paper presents an innovative computer-aided modelling approach for equilibrated spatial
structures with any combination of compression and tension forces.
Keywords: structural design, combinatorial equilibrium modelling, real-time structural design tools,
strut-and-tie models, topological graphic statics

1. Introduction
In structural design, models are used to predict the physical behavior of structures. Since the rise of the
digital culture during the late 20th century, computers have played an increasingly essential role for
these predictive behavior models (Au [1]). However, most of the currently used computational models
are not appropriate for the conceptual structural design phase (Mueller [12]). Either they require a large
set of yet unknown information (such as topology, geometry and material properties) or they are
restricted to specific structural typologies like compression-only surface structures (e.g. RhinoVault).
Whereas with the first models the designer doesn’t have the active control over the structural behavior,
with the latter approach the designer loses control over the architectural form (Otto [18]). Furthermore
both approaches nowadays have in common that they focus very early on the metric and quantifiable
behavior of the structure and on the relationship between form and force only. The observation by
Billington [3] that there can be “no (quantifiable) optimum in structures, but only many reasonable
choices, allowing the designer the freedom to express his own ideas” underpins this shortcoming. Cecil
Balmond [2] with his informal/sequential understanding by way of illustration uses the words “rhythm”,
“fluctuations” or “episodes” in space to describe what structures can provide. In order to be useful in
the conceptual design phase an approach should consequently be able to describe visually the influence
of “lateral shifts” which are transformations in between topologically fixed design spaces (Goel [6]) in
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions

order to make sure that several radically different solutions from different design spaces are known.
According to Marples [10], these radical variations are indispensable to get a clear picture of “the real
nature” of the problem during the co-evolutionary matching process of problem and solution in the
conceptual design phase. However, at present, neither a satisfactory theoretical design framework nor a
digital tool exists that makes this explicitly available to the structural designers.
The research presented here tries to overcome the described shortages and develop a theoretical
framework to incorporate the control of qualitative structural behavior in the conceptual design phase
in a novel way. This paper presents an innovative computer-aided modelling approach for equilibrated
spatial structures with any combination of compression and tension forces including the ability to
interactively modify the qualitative structural behavior by means of topology and states of each force
member (compression or tension) interactively in order to make “lateral shifts” in between design spaces
possible. The structure of this papers is as follows: first, the basic concepts of the theoretical framework
of the Combinatorial Equilibrium Modelling (CEM) are discussed; the middle part describes the main
properties and technical peculiarities of this Combinatorial Modelling approach; finally, the paper
presents a first prototypical implementation within the CAD software package Rhinoceros and a playful
application of the approach for a “single-cell” design scenario independent of scale, material and
program.

2. Concepts

2.1 Theory of Plasticity & Graphic Statics

Structural behavior is usually described in terms of forces. Generally the calculation of the inner forces
are based on the Theory of Elasticity and carried out with numerical methods such as FEM. By using
FEM the designer gets an overview of the structural behavior, but can hardly use it for the design,
especially in the conceptual design phase (Kotnik and D’Acunto [8]). Furthermore the results are highly
influenced by the designer’s assumptions on material properties, boundary conditions, the load history
and the future behavior of the structure during the use. The fact that this theory is not able to describe
the behavior of many construction materials in an appropriate manner underpins the need for a change
(Muttoni et al. [14]). The main characteristic of the Theory of Plasticity, which represents an alternative
to the Theory of Elasticity, is that it concentrates on the ultimate rather than on the serviceability state
of design. The Theory of Plasticity is therefore based on the assumption of rigid-plastic material
behavior. While neglecting the material stiffness and the deflections the Lower Bound Theorem as a
part of the Theory of Plasticity allows to focus on equilibrium solutions. Through the application of the
Lower Bound Theorem it is furthermore possible to use Graphic Statics to visualize and construct these
equilibrium states of a structure. Graphic Statics directly illustrates the reciprocal relationship between
form and force and make it thereby intuitive and understandable (Maurer and Ramm [11]). This
argumentation is also supported by successful practitioners such as Nervi [15] as well as by the scientific
community in this research area, exemplarily by Caitlin Mueller [12, p.41]:

“Human designers are highly visual and can process and evaluate information much more quickly and
fully when it is presented graphically.”
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions

2.2 Multiplicity of solutions & topological thinking

Another major benefit of the Theory of Plasticity and its Lower Bound Theorem is that it allows for a
multiplicity of qualitatively differing structural behaviors. That is, for of a given structure (described
only with geometric boundaries) and considering the same load scenario, infinite equilibrated solutions
are possible as long as the forces fulfill equilibrium conditions, are lying within the geometric
boundaries and satisfy the yield conditions, defined by the material (Muttoni [14], p.9) As already
briefly described in the introduction the work with a certain amount of differing structural alternatives
appears to be the key for the conceptual design phase and supports the application of Graphic Statics
and its theoretical backbone. A well-known way of working with different alternatives is based on
structural typologies. A system-inherent problem of this rather conventional approach is pointed out by
Laurent Ney [16]:

“A typology has a name, and the form and the relationship between the elements is described. The
advantage of this is that it is easy to talk about structure, but the disadvantage is that how the structure
looks is predetermined… This approach has a perverse effect: the vocabulary freezes the object, and
the objects thus frozen assume a sort of inviolable legitimacy. In order to arrive at new forms and
concepts we have to free ourselves form such pre-defined typologies.”

In the first instance, one of the aims of the present work is to overcome those pre-defined typologies by
extracting and analysing important equilibrium properties (force location, force direction, force
magnitude and force connectivity) and their basic relations, respectively their topological structure in a
meaningful way. An analogy of this approach can be found in the guiding principle of Le Ricolais work
in which the topological arrangement of the structure plays the key role, “since the essential aspects of
form are not expressed in figures or measure units, but rather belong to the realm of number and the
non-measurable” (Le Ricolais [9]). Furthermore, it is the goal of this research to show that based on
combinations of very basic equilibrium elements new structures can emerge which are not associated
with pre-defined typologies. This thinking has a reference to Herman Hertzbergers [7] structuralistic
manifesto, where he states that each design no matter at which time, no matter where it is designed is
nothing else then an interpretation (a combinatory and a variation) of the archetype. This guiding
principle can be easily implemented with the use of Graphic Statics which is not fixed to any type of
structure, but is able to describe the geometrical relationship between force and form for any
combination of tension and compression forces.

One reasonable simplification in this research is to work with discrete resultant forces and their
combinations instead of their underlying stress fields. This makes the handling especially for the
conceptual design phase much easier but is nevertheless accurate to represent the structural behavior
(Schlaich [19]). Connecting the presented concepts leads to the observation that the multiplicity based
on the Lower Bound Theorem and a combinatorial approach expressed through Graphic Statics gives
potential for the design of non-typological structures at an early design stage.
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions

3. Combinatorial Equilibrium Modelling (CEM)

3.1 Qualitative behavior and quantitative behavior

A key contribution of this research is the way forces in equilibrium are represented. On top of the
quantitative reciprocal behavior between the form diagram and the force diagram (Cullmann [5]) (right
side of figure 1) a visual representation of the qualitative behavior is proposed (on the left side of fig.1).
This qualitative behavior can be subdivided into two diagrammatic representations. The first one is
named force flow diagram. This representation, which is also used by Fivet, visualizes topological
dependencies of the nodes in a force network (strut-tie) as a directed graph. With this understanding,
borrowed from the Graph Theory, structural behavior can be represented as a qualitative two-
dimensional force flow network G with force connection nodes V and directed force edges e. Within
this research the term “force flow” is not related to a dynamic effect respectively a change over time.
On the contrary, the force flow diagram represents the topological connectivity between the force edges
and thereby defines the sequence of force edges from external applied loads Li (source) to the reactions
at the supports Si (sink). For a given force flow network the shortest path from each source to its closest
sink can be identified. Those edges which lie on these paths are called es and their direction follows the
force flow; in the given example these edges are e1, e2, e4, e5, e7 and e8; and they belong to the set Ms.
The other two edges e3 and e6, which are not lying on a shortest path from a source to its closest sink,
belong to the set of Mns. The second diagram, representing the qualitative behavior, is named force state
diagram. This diagram visualizes the combinatory state of the forces; whether the force edges are in
compression or in tension. A compressive force edge is coloured blue, a tensile force red and an external
force green. In the example the force state vector B contains the following combination of compression
and tension; {c, c, c, c, c, t, c, c}.

Figure 1: Visual representation of the qualitative and the quantitative structural behavior.
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions

The quantitative behavior is represented with a form diagram, showing the geometry of the structure
and the locations of the forces within this structure, and a force diagram, representing the magnitudes
of the forces by the absolute length of the corresponding line. Corresponding forces in the form and in
the force diagram are parallel. A load element Li in the force flow network corresponds to load vector
���⃗
Q ı in the form and in the force diagram. A support element Si corresponds to a support reaction force
vector ���⃗
R ı and an inner force element in the flow network ei to an inner force vector s��⃗.
ı

3.2 Assumptions
In order to attain a reciprocal relationship between the form and the force diagrams and respect the
equilibrium conditions, the following two rules apply for the construction of a generic force flow
network:

1) In order to ensure that any node is in static equilibrium, the output force of this node must
counterbalance the connected input forces. This principle is illustrated for node V1 in figure 2
and corresponds to the rule that all the forces that meet in the same node in the form diagram
must form a closed polygon in the force diagram (Bow [4]).

Figure 2: Visualization of rule 1 for node V1 and corresponding P1

2) In order to make sure that this modelling approach works for arbitrary external skew loads Qi,
a constraint should be considered: at each node exactly one of the input forces is allowed to lie
on the shortest path from its source to its sink and as such be part of the set Ms. This constraint
makes sure that assigned forces do really meet in space and corresponds to the rule that all the
forces that form a closed polygon in the force diagram must meet in the same node in the form
diagram to ensure the sum of the moments of each force in relation to the same point to be zero
(Bow [4]).
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions

Figure 3: Visualization of rule 2 for node V1 and corresponding P1

If those conditions (1 & 2) are satisfied for all the nodes in the force flow network, equilibrium can be
guaranteed and constructed by means of geometrical operations without the need for any numerical
procedure to be implemented into the process in any phase. Modifications in the force flow network
affect both form- and force diagram simultaneously. The non-reciprocal relationship from these
qualitative diagrams to their initialized form and force diagram is described in the next paragraph.

Apart from the additional force flow representation, another aspect introduced by this research is to
consider the support locations not compulsory as fixed but rather as a variable parameter. This
conception allows to generate, control and compare different qualitative and quantitative properties for
the same structure. It will be shown that this approach represents a useful alternative to other
typological-based approaches where the rigour formulation of the supports location often prevents the
direct exploration of a lateral multiplicity.

3.3 Process
The parameters that connect the qualitative and the quantitative behavior in this research are the line
parameter λs, for those force edges that lie on a shortest path (set Ms) and the force magnitude factor νns,
for those force edges that are not part of any shortest path (set Mns) in the force flow network. The line
parameter λs (0 < λs <1.0) describes the location of the consecutive node along the line of action of the
input force, from the current node (λs = 0) to the intersection point of the line of action and the boundary
geometry of the structure (λs = 1.0). The force magnitude factor νns (0 < νns < ∞) describes the factor
that is multiplied with the input force with the lowest index. The force state (B) defines whether the
forces in the form diagram are tensile or compressive forces.
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions

Figure 4: Influence of line parameter λ1 and λ2 for edges e1 and e2 on a shortest route

For example, developing the previous case with two independent external loads, a rectangular boundary
geometry and the qualitative behavior described at the left side, two metric parameters (λ1 for e1 and λ2
for e2) are initialized. The directions and the magnitudes of the corresponding inner forces ���⃗ s1 and s���⃗2
are given by the directions of the external loads ����⃗ ����⃗2 . The curve parameters λ1 and λ2 describe the
Q1 & Q
location of the next nodes P3 and P4 along the line of action. Here the values set, for example, to 0.2 and
0.4. With this initial flow network it is not possible to change the directions of any inner force.
Furthermore the location of the supports (P5 and P6) as well as the directions and magnitudes of the
support forces (R ����⃗1 and ����⃗
R 2 ) are pre-defined.

By connecting the nodes V3 and V4 with an additional force edge e3, the qualitative behavior of the
structure can be transformed from a column-like into an arch-like behavior. Since this member e3
doesn’t lie on a shortest route a force magnitude parameter (ν3) is initialized. The direction of this
compression force ���⃗ s3 is defined by the locations of the nodes V3 and V4 in the form diagram. They are
in turn dependent on λ1 and λ2, as discussed above. If the magnitude factor of ν3 equals 0, then the
direction of s���⃗4 corresponds with s���⃗1 and ���⃗ s2 . A magnitude factor of ν3 = 1.0 defines the magnitude
s5 with ���⃗
of force ���⃗
s3 as 1.0 times the force magnitude of the incoming force ���⃗ s1 (because e1 has a lower index
than e2) at node V3. The same magnitude factor respectively the same force ���⃗ s3 is then taking into
account at node V4. The influence of different values (0.35, 0.7 and 2.1) for ν3 is shown in the form and
the force diagram in fig. 5. The support locations (P5 and P6), as well as the support reaction forces (R ����⃗1
����⃗
& R 2 ) are now dependent on three parameters (λ1, λ2, ν3).
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions

Figure 5: Influence of line parameter ν 3 for values set to 0.3, 0.7 and 2.1

To transform the arch into a beam-like behavior (illustrated in the beginning on figure 1), two nodes
and one tension force can be added to the described flow network of figure 5. This beam-like force flow
network, specifically the two last forces s7 and s8, which correspond to the support forces R ����⃗1 and R
����⃗2
are then depending on four line parameters (λ1, λ2, λ4, λ5) and two force factor parameters (ν3, ν6).
Based on this modelling approach it is possible to build up and laterally change force flow networks
which correspond to certain qualitative behaviors interactively. Based on each flow network a variation
of initialized parameters within a design space opens up. These parameters can be either actively
changed so that the form of the force flow becomes directly shaped by the designer or passively
computed, when objectives are defined. A straight-forward objective could be the distance between the
actual support forces and the eventually desired and pre-defined support locations. A minimization of
this residual objective can be carried out with the help of an evolutionary optimization algorithm
(Ohlbrock [17]).

4. Designing with CEM

4.1 Implementation of the Interactive Tool
To test the approach of CEM in 3D, an interactive computer-aided design prototype has been developed
within the CAD software Rhinoceros, which is a common used tool in 3D design. The required code
was written in the scripting environment RhinoPython which can be accessed via Grasshopper, the
associative modelling plug-in for Rhinoceros. The qualitative behavior can be controlled and changed
by moving the sliders from compression to tension of the given flow network in the screenshots upper
left window. The geometrical boundaries of the structure and the vectors of the loads are modelled
directly in the 3-d window in the middle. The corresponding metric degrees of freedom (λs & νns) can
be controlled within the Grasshopper panel on the lower left side of the screenshot. The prototype shows
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions

very robust characteristics and allows quickly to explore qualitatively differing structural concepts for
arbitrary geometrical boundaries and load situations in real-time.

Figure 6: Screenshot of the Interactive Tool

4.2 Design Examples

With the help of a small design example the prototype is illustrated. The process starts with the
modelling of a simple cube, representing the geometrical boundaries of the single-cell design space and
the definition of the governing load case. Figure 7 shows a small selection of six exemplarily
qualitatively differing force flows which are quickly generated with this new approach.

a b c

d e f
Figure 7: Six exemplarily generated force flows
Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions

Typological structure (Hanging roof)

The typology of the first option corresponds with one of a hanging roof known from arenas such as the
Madison Square Garden in New York City. On the right side a possible form diagram is shown for
identical metric values (λs & νns) for each layer in the flow network. Here an active “shaping” of the
form and force diagram is carried out.

Figure 8: Force state and form diagram for a hanging roof behavior
Non-typological structure
As discussed earlier, one main motivation for the topological representation and this combinatorial
equilibrium modelling is to overcome typologies and their implied formal limitations. Although the
combinatorial state pattern on the left side seems to be very regular and the metric parameters (λs, νns)
are still identical for each sequence of the flow network, it is not evident from the beginning which
structural typology is the result of this qualitative behavior. It is clear that more complex but also
intriguing forms are hidden in the depth of this CEM-based approach.

Figure 9: Force state- and form diagram for a non-typological behavior

Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions

Comments in a practical context

As a product of this approach the designer possibly generates a spatial force flow which is not a result
of a pure deductive analytical process but rather the result of an inductive, creative and interactive
process based on graphic statics and equilibrium only. The presented approach corresponds quite well
to the structural design philosophy of many successful practitioners such as Frei Otto, Heinz Isler and
Sergio Musmeci [13] who in the 1970s stated that in his design “the stresses are not the unknown”.
Nowadays a tendency towards material-independent force-flows as a departure is detectable among
contemporary structural designers (Sobek [20]). It is obvious that these force flows can only give a first
but strong idea of the structures global behavior.
To get a clearer picture of the behavior of the structure, some further decisions have to be taken. For
example if the forces are discretized as a visible truss-like structure, or if some forces constitute a more
continuous surface structure. Furthermore the materialization and construction detailing obviously plays
an important role in this context. It is clear that further analysis should be based on these material-related
decisions to ensure the structures serviceability and a prevention against buckling of the compressive
members. More to the point, this materialization is the key for the architectural articulation of space but
also crucial for the constructive logic of the structure.

5. Conclusion
The presented Combinatorial Equilibrium Modelling (CEM) approach, implemented within a CAD
environment, allows for the intuitive material-independent exploration and even an active design of
spatial force flows for a defined boundary geometry and arbitrary load scenarios. In such a way, the
global structural behavior can be controlled by the designer, and furthermore be aligned with the
architectural concept quickly, interactively based on geometric operations only. By allowing lateral
shifts between different structural behaviors the presented approach frees the designer from early
determinations and typologies and gives a new perspective to the discussion on the interaction between
structural behavior, architectural design and the use of computational models in the conceptual design
phase.
 Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2015, Amsterdam
Future Visions

References
[1] Au F., A Necessary Resistance within Architect-Engineer Collaboration, in Interdisciplinary
Design – New Lessons from Architecture and Engineering, Kara H & Georgoulias A (ed.),
Harvard Graduate School of Design and Actar, 2012, p.237.
[2] Balmond C., Crossover, Prestel, 2013, p.3.
[3] Billington D., The Tower and the Bridge - The New Art of Structural Engineering, Princeton
University Press, 1985, p.142.
[4] Bow R. H. Economics of construction in relation to framed structures, E & F N Spon, 1873.
[5] Cullmann C. , Die graphische Statik, Meyer & Zeller, Zurich. 1864.
[6] Goel V., Sketches of Thought, MIT-Press, 1995.
[7] Hertzberger H. Vom Bauen – Vorlesungen über Architektur, Aries, 1967, p.31.
[8] Kotnik T. and D’Acunto P., Operative Diagramatology: Structural Folding for Architectural
Design, in Proceedings of Design Modelling Symposium 2013, Berlin, 2013, pp. 193-203.
[9] Le Ricolais R., Matieres, in Structures Implicit and Explicit. Via, no.2, 1973.
[10] Marples D., The Decisions of Engineering Design, Institute of Engineering Designers, London,
1960.
[11] Maurer B. and Ramm E., „Zeichnen als Sprache des Ingenieurs“, in Karl Culmann und die
graphische Statik. Verlag für Geschichte der Naturwissenschaft und der Technik, Berlin, 1998, p.
205.
[12] Mueller C., Computational Exploration of the Structural Design Space, PhD-Thesis, MIT, 2014.
p. 5 &41.
[13] Musmeci, S., Le tensioni non sono incognite, in: Parametro, Nr. 80, October 1979, p. 36-46.
[14] Muttoni A., Schwartz J. and Thürlimann B., Bemessung von Betontragwerken mit
Spannungsfeldern, Birkhäuser, 1997.
[15] Nervi P., Academic training of the designer, in Structures. F.W. Dodge Corp, Chicago, 1956, pp.
11-24.
[16] Ney L., Adriaenssens S., Devoldere S., and Strauven I., Shaping Forces. Brussels: A+ Editions /
Bozarbooks, 2010.
[17] Ohlbrock P.O., Balancing Behaviours – Designing with combinatorial equilibrium models, in
Proceedings of Design Modelling Symposium 2015, Copenhagen, 2015, paper in review-process.
[18] Otto F., Mitteilung des Instituts für leichte Flächentragwerke: Experimente. IL 25, Universität
Stuttgart, 1990, p.0.5.
[19] Schlaich M., Computerunterstützte Bemessung von Stahlbetonscheiben mit Fachwerkmodellen,
Dissertation, ETH Zürich, 1989.
[20] Sobek W., Radical Sources of Design Engineering, in The new Structuralism: Design,
Engineering and Architectural Technologies. Wiley, 2010, p.33.

View publication stats