Content uploaded by Lennert Loos
Author content
All content in this area was uploaded by Lennert Loos on Oct 04, 2016
Content may be subject to copyright.
19th IABSE Congress Stockholm, 21-23 September 2016
Challenges in Design and Construction of an Innovative and Sustainable Built Environment
1
Enhanced decision making in the structural design process by means
of a dashboard approach
Lennert Loos
PhD researcher, Vrije Universiteit Brussel, Brussels, Belgium
Contact: lennert.loos@vub.ac.be
Kenny Verbeeck
Partner, Ney & Partners, Brussels, Belgium
Lars De Laet
Assistant Professor, Vrije Universiteit Brussel, Brussels, Belgium
Abstract
This paper proposes a novel approach for comparing different structural geometries based on data
visualisation and illustrates this with bowstring bridges as case study. The aim of the approach is
to have a quick and overall insight in different design proposals at the beginning of the design
process. The approach will use a dashboard that is a combination of charts. This dashboard will
enable the (structural) designer to have all information on the structural behaviour of a series of
design proposals at hand and to make informed design decisions based on the compact
dashboard.
Keywords: Structural Design, Performance Aided Design, Structural Informed Design, Bow-string
Bridges, Data Visualisation, Design Decision-making, Dashboard Approach
1 Introduction
Quite often the design of a structure is based on a
shape imposed by the designer. It is clear that
optimizing structural geometries is a field in itself,
but the question in the early design phase is to
estimate the impact of different structural
topologies and boundary conditions on the overall
behaviour of the structure.
A possible way to compare different structural
geometries is to model a geometry, conduct a
finite element analysis with all load cases and run
the code checks to dimension and verify the
structure. After the dimensioning, the model will
have a clear output: the weight of the structure.
Based on this parameter, one is able to compare
different design alternatives. However, this
approach might become very time consuming
when wanting to compare a whole series of
topological alternatives. An experienced structural
designer will start to point compare structures:
maximal normal forces, maximal bending
moments, maximal deformations, etc. Yet this
gives punctual information about peak
performance, while neglecting the view on the
overall performance and behaviour 10.
In this paper, a novel approach is presented using
two-dimensional models and lightweight FE-
19th IABSE Congress Stockholm 2016
Challenges in Design and Construction of an Innovative and Sustainable Built Environment
2
calculations followed by a dashboard of graphs.
This dashboard represents the overall behaviour
and performance of structural design alternatives
and enables the structural designer to compare
different design alternatives in a nuanced way,
rather than looking for the strict optimal solution
and ignoring the architectural needs. The
approach will be illustrated on the basis of a case
study of bowstring bridges. The tool is aimed to
offer the experienced engineer a synthetic view of
the whole, whilst not reducing the behaviour of
bowstring bridges to a single number. This tool is
part of ongoing research and has as main focus to
illustrate a methodology for comparing structures
in an early structural design phase. Previous
research on this topic, but focussing on gridshells,
is presented in [10].
2 Integrated methods for structural
design considerations
2.1 Traditional sequential approach
A succession of different sequential phases exists
in the traditional design process. The overall
geometry (shape, topology) is first conceived
(often by an architect). During this phase
architectural and functional criteria such as
aesthetics and boundary conditions are taken into
account. The structural behaviour and efficiency is
often only considered in a second phase, in which
the geometry is subjected to finite element
calculations and in which the results are analysed
(Figure 1). If the results are not satisfying, the
design is changed iteratively. If design
modifications are made in the geometric
modelling tools, those changes need to be
remodelled as well in the engineering tools, an
approach which is time consuming and can lead to
errors.
According to Clune et al. 2, the reason for this
sequential approach is the specificity of the
software applied by the architects and engineers.
Architects use geometrical modelling tools that
stimulate the exploration of the design space and
creativity, mostly without considering structural
efficiency, whereas engineers conduct finite
element calculations on a predefined shape. In
addition, the results generated with a finite
element analysis software can in general only be
checked case per case, making it hard for the
structural designer to compare design
alternatives. Also, only maximum values are often
taken into account with this approach, leaving into
the shadow the global structural behaviour.
Figure 1. Flowchart of a ‘traditional’ sequential
design process
2.2 Structural design tools and their focus
Geometrical properties, such as shape and
topology, are an important factor for a structure’s
performance 5. Hence structural considerations
should preferably be introduced in the first stages
of the design process.
Computational structural design tools tend to
enable a better integration of structural
performances during the early conceptual design
phase. Various structural design tools exist, all
with their specific focus, complexity, user
interactivity and required technical background.
Each of the existing structural design tools has its
own goal and is very suited for a specific use: think
of interactive graphic statics, real-time numerical
structural analysis, form-finding tools and
definitions, …
However, we are looking for an approach whereby
various designed proposals can be calculated
quickly in a familiar FEA software, after which the
load bearing behaviour of all cases can be
compared with each other rapidly and easily in
one general interactive overview. Most
19th IABSE Congress Stockholm 2016
Challenges in Design and Construction of an Innovative and Sustainable Built Environment
3
importantly, the comparison of structures should
be possible without the need to calculate the total
weight of the structure. First comes understanding
the structure, afterwards dimensioning that
structure.
2.3 Interactive data visualisation
Data is currently generated at very high volumes
and the generation rate is still speeding up. The
exploration and analysis of these vast volumes of
data is increasingly difficult. However, information
visualisation and visual data mining often helps to
deal with a lot of information. The advantage of
visual data exploration is that the user does the
most important part: he is directly involved in the
data mining and interpretation process 4. As
Joyce states in his paper 3, data visualisation is
the link between graphic design and quantitative
information. Besides that, the discipline of data
visualisation searches for the cognitive
understanding and interpretation of graphical
figures by people 3.
Nowadays data visualisation is used a lot in all
different kind of fields. Main reason is the
availability of relative easy-to-use tools. Think of
the well-known chart-makers implemented in MS
Office. However also more complex visualisations
can be found in a whole range of topics; think for
instance of the popular publications “Information
is Beautiful” and “Knowledge is Beautiful” by Mc
Candless 67. New tools are making the way for
more user-interactive dynamic visualisations on
the web. Several newspapers (i.e. New York
Times) picked up the potential of this new
information medium and show increasingly
interactive visualisations on their websites. The
new web standards implemented on modern web
browsers and the relative ease of scripting these
visualisations are making interactive data
visualisations more accessible 3. In this paper
the JavaScript library “D3” 1 is used.
Before elaborating on the developed approach,
we first present the case studies that will serve to
illustrate the approach.
3 Case study: bowstring bridges
3.1 Geometries
As mentioned before, bowstring bridges will be
used as subject to discuss the general thoughts of
the dashboard approach. Various bridge
geometries have been generated (Figure 3) in the
interactive design environment of Grasshopper3D,
a plugin for Rhino3D, CAD software developed by
Robert McNeel & Associates. Not all of the design
proposals are form-found. The very preliminary
and approximate form-finding of some models has
been performed by means of the Grasshopper
plugin Kangaroo. This serves the opportunity to
generate quickly a variety of geometric models
that can easily be implemented in the further
workflow of the comparison tool.
The span of all bowstring geometries is 85 meters
and the ratio span/height equals 5 for all designs.
The calculation models are considered in two
dimensions and also the structural analysis is
conducted in 2D. The bowstring bridges consist of
a continuous arch, a continuous deck and
connecting cables. The arch and deck-ends are
connected with hinges. The bridge itself is simply
supported at both ends of the deck. The
connections between the cables and the bridge
deck divide the deck in 19 equal parts for all
models.
3.2 Structural analysis
A general typical section for the arch, the deck and
the cables is applied equally in all structural
models, after which the structural models are
linearly calculated in the finite element software
SCIA Engineer (Nemetschek) under a reference
load. This reference load consists of 18 identical
equally distributed point loads at on the deck
(Figure 2) for each bowstring model. This
reference load is a fixed assumptive self-weight
for all cases.
Figure 2. The used structural system for each
design alternative
19th IABSE Congress Stockholm 2016
Challenges in Design and Construction of an Innovative and Sustainable Built Environment
4
Figure 3. The considered bowstring configurations and their respective model name.
4 Dashboard approach
The structural numerical outcome of these quick
FE calculations is post-processed to gather the
information of all models in a dashboard overview
(Figure 5). When considering a whole series of
structural models and possibly wanting to add
more models, it is important to make this process
as generic and automatic as possible.
The interactive dashboard allows the designer to
hide and show different structural models and
their represented structural information in order
to have all essential parameters for comparison
and interpretation at one single place. Moreover,
this way of nuanced comparisons facilitates the
choice of proposals that both satisfy the
architectural and structural criteria, rather than
looking for the strict optimal solution whilst
ignoring architectural needs. In what follows, the
used dashboard and its components will be
discussed and illustrated with an example.
4.1 MN-chart
The MN-chart displays the axial force and bending
moment in all structural elements of the
numerical model; each dot represents one
structural element. The horizontal axis represents
the bending moments while the axial forces are
located on the vertical axis. This representation
gives a first understanding of the overall
behaviour of the structures, each in its own
colour. An illustration of the principle is given in
Figure 4. We clearly observe three different main
structural types of a bowstring bridge: arch, deck
and cables (Fig. 4). The cables are taking relative
small tensile axial forces only, the arch is mainly in
compression and the deck is in tension while
exhibiting relative large bending forces. The area
of the dots is a function of the proportion M/N.
This is an important parameter if we are looking
for structures with the smallest bending moment,
compared to their axial force. It can be clearly
observed that cable members are loaded more
efficient (axial, rather than by bending) than the
deck elements. By having this chart, the structural
designer immediately gets a view on the different
design proposals and the consequences of their
geometries on the structural behaviour.
Figure 4. MN-chart including 4 bridge models
represented in blue, pink, green and yellow. The
MN-chart turns out to be useful for a general
interpretation of different design proposals.
Bridge deck
Cables
Bow elements
19th IABSE Congress Stockholm 2016
Challenges in Design and Construction of an Innovative and Sustainable Built Environment
5
Figure 5. Complete dashboard containing the MN-chart, 4 line charts
and the bar charts. Models A, B, C and D are visualised.
4.2 Bar charts
Additionally to the MN-graph, a bar chart with
information on the distribution and maxima of
internal forces is given. A bar chart can be found
at the bottom of Figure 5. For all main elements of
the bowstring bridge (arch, deck and cables), the
maximum internal forces are listed for each design
proposal. In addition, the integrals of the internal
forces over the element lengths of the arch, deck
and cables are included per model as well. These
integrated forces give a better insight in the
needed structural mass compared to considering
maximal forces only. All values are also graphically
represented by coloured bar charts that can be
sorted by value, aiming for a more readable and
Internal forces of the arch Internal forces of the deck
19th IABSE Congress Stockholm 2016
Challenges in Design and Construction of an Innovative and Sustainable Built Environment
6
intuitive way of comparing. In a glance, the
designer can make some intuitive observations
and categories based on the performance of the
structures.
4.3 Line-graphs
For a more detailed view of the distribution of the
internal forces in the entire structure, the
dashboard includes a collection of line-graphs,
representing the normal forces and bending
moments along the structure’s length. Because
the arch of bowstring bridges typically exhibit
large axial compression forces while the deck is in
tension, the dashboard contains separate graphs
for both the deck and arch elements, resulting in 4
line-graphs (Figure 5). Since the bow lengths of
the various models differ slightly, one can choose
to normalise the total length of the arch, or to use
the actual scaled lengths on the horizontal axis.
5 Case study
To illustrate the new approach, the four models B
(blue), G (pink), L (green) and M (yellow) will be
discussed and compared by using the dashboard
(Figure 3).
A reasonable way of comparing these bowstring
models would be to look for the model with the
smallest maximal bending moment in the arch.
However, we selected four different design
alternatives that have a similar maximal bending
moment in the arch. Nevertheless, the MN-chart
in Figure 6 shows that the various models differ in
structural behaviour. Looking at the bar charts in
Figure 7, it becomes even clearer that focussing
on maximum values alone does not provide a
complete view of the structural efficiency since
the 'non-peak' values would be ignored.
Model M, as an example, has a lower maximal
bending moment in its arch than models B and G
(see first column figure 7). However the integral of
bending forces along the complete arch is higher
than the other three alternatives. This will have an
influence on the required structural material. The
same occurs with the bending moments of the
deck. As can be seen in the MN chart and the bar
chart, models B, L and M have similar maximal
bending moments in the bridge deck. Yet, model B
has a smaller integral (distribution) of bending
moments along its deck.
These are considerations that are important to
have a more overall but nuanced view for
intelligent decision-making during the preliminary
design phase. In addition, the MN chart allows the
designer to cluster elements for preliminary
dimensioning. The bending moments in the deck
of model G know a large spread (which indicates
more different sections to be applied), where the
bending moments in the deck of model M are very
similar.
Figure 6. MN-Chart of models B (blue), G (pink), L
(green) and M (yellow)
Figure 7. Bar charts representing the maximal internal forces and integrals of internal forces for each model.
19th IABSE Congress Stockholm 2016
Challenges in Design and Construction of an Innovative and Sustainable Built Environment
7
Arch – My (kNm)
Arch – N (kN)
Deck – My (kNm)
Deck – N (kN)
Figure 8: Line charts with the continuing of the
internal forces
When looking at the line graphs in Figure 8, one
immediately gets an insight in the distribution and
peaks of internal forces in the decks or the arches.
It becomes clear that the selected design
proposals have different types of bending
moment diagrams in their arches. The question
now is to compare these alternatives, based on
the line charts of the bridge deck and the arch.
Model B (blue) has the smallest bending integrals
for both its arch and deck. The designer may
however prefer a more fluent bending diagram
such as model L (green) or model M (yellow)
whereby the bending moments do not change a
lot over the design (for reasons of dimensioning).
This discussion shows the more nuanced and
interactive way of comparing structural
geometries in the first conceptual design phase.
For reasons of clarity, only four models have been
lighted out. Of course this interactive dashboard
approach lends itself to incorporate many
structural geometries that can be switched 'on' or
'off' in the visualisation of the tool.
6 Conclusions
In an early structural design stage, it is important
to quickly and easily assess the structural
behaviour of various design alternatives. When
considering bowstring bridges, various geometries
can be imagined. In order to make informed
design decisions, these geometries should be
analysed before proposing a viable design
solution.
In the traditional engineering process, different
structures will be compared after all load cases
have been taken into account and the
dimensioning of all elements has taken place
according to the current Codes. This process is
time consuming and complex, whereby the user
needs to make comparisons based on a lot of
various screens and extreme values.
To avoid this, we proposed a quick and easy
approach whereby the comparison of the
structural behaviour of a large amount of design
proposals is facilitated. The difference in structural
behaviour is reflected by the use of one single
reference load case. After the structural
calculation the generated data is plotted in a
dashboard of graphs, like an overall MN-graph,
bar and line charts. The bar charts not only
illustrate the maximum occurring internal forces
and bending moments, but also the integral of the
internal forces and bending moments. The line
chart represents the forces and moments along
the structure’s length. These graphs prove to be
interesting for bowstring bridges. It takes some
effort to create the appropriate dashboard.
However, future research will search for the right
representation methods and will indicate which
graphs are useful for comparing other structural
typologies.
This dashboard allows comparing roughly at a
glance the structural behaviour of the various
proposals. By combining the dashboard including
an automated graphical plot of all the occurring
forces and a table containing key numerical
values, the structural designer has all the tools at
hand to quickly assess design alternatives.
With this approach, the structural designer can
assess more quickly the different design proposals
based on a more complete understanding of the
structure. The goal is to replace a 'computed
therefore it is' approach for a knowledgeable
approach in the structural design thinking..
19th IABSE Congress Stockholm 2016
Challenges in Design and Construction of an Innovative and Sustainable Built Environment
8
7 References
[1] Bostock M., Ogievetsky V., & Heer J. D³
data-driven documents. Visualization and
Computer Graphics. 2011; IEEE Transactions
on, 17(12), 2301-2309.
[2] Clune R., Connor J.J., Ochsendorf J.A.,
Kelliher D. An object-oriented architecture
for extensible struc-tural design software.
Computers & Structures. 2012; 100-101; 1-
17.
[3] Joyce S. C. Web Based Data Visualisation
Applied to Creative Decision Making in
Parametric Structural Design. Proceedings
of the International Association for Shell and
Spatial Structures (IASS) Symposium “Future
Visions” Amsterdam, Netherlands, 2015.
[4] Keim D.A. Information Visualization and
Visual Data Mining. IEEE Transactions on
Visualization and Computer Graphics 8. N° 1
(March 2002); 1–8.
[5] Larsen O.P. & Tyas A. Conceptual Structural
Design: Bridging the Gap between
Architects and Engineers. Thomas Telford,
London, United Kingdom, 2003.
[6] McCandless D. Information is Beautiful.
Collins, London, United Kingdom, 2012.
[7] McCandless D. Knowledge is Beautiful.
Collins, London, United Kingdom, 2014.
[8] Murray S. Interactive Data Visualization for
the Web. O’Reilly Media, California, U.S.,
2013.
[9] Verbeeck K., Loos L., De Laet L. How to re-
open the black box in the structural design
of complex geometries. Proceedings of the
International Conference on Structures and
Architecture. Guimarães, Portugal, 2016.
[10] Verbeeck K., Muller L., De Laet L. Structural
Design by a Dashboard approach.
Proceedings of the International Association
for Shell and Spatial Structures (IASS)
Symposium “Future Visions” Amsterdam,
Netherlands, 2015.
