Conference PaperPDF Available

Evolutionary design systems: a conceptual framework for the creation of generative processes

190
Evolutionary design systems: a conceptual framework for the creation of
generative processes
Patrick Janssen, John Frazer, Tang Ming-xi
School of Design
The Hong Kong Polytechnic University
Hung Hom, Kowloon
Hong Kong
ABSTRACT
This paper presents an conceptual framework for the construction of generative mapping processes as a
basis for creating active design tools in the domain of architecture. Such generative processes are seen
as key components within evolutionary systems that manipulate populations of alternative solutions in
order to discover previously unexplored possibilities. Solutions are represented in two forms: as highly
encoded genotypes referred to as design seeds and as decoded phenotypes referred to as design
proposals. The generative process maps the design seed to the design proposal. The discussion of
generative processes is in two parts. In the first part it is argued that any generative process that aims to
create a wide range of solutions that differ from each other in fundamental ways must focus on a
limited subcategory of possible designs. It is proposed that the endeavour to create active design tools
demands that the focus be on the designer's highly personalised style, called a design-schema. The
second part discusses how to uncover the essence of an architectural design-schema. In particular, it is
argued that implicit and familiar aspects of buildings must be scrutinised in order to reveal the
knowledge that is essential to capturing and codifying a design-schema. A range of rationalisations and
conceptualisations of built form are presented with examples to illustrate possible routes of analysis.
Finally, in conclusion, the possibility of discovering universal generators common to many divers
generative processes are discussed.
1 INTRODUCTION
A new phase of design tool is now under development. These tools are described as
active as opposed to passive in that they will become and integral part not only of the
manual design process but also of the cognitive design process. The task of these
software tools is not only to allow designers to analyse and evaluate, but also to
generate and explore alternative design proposals. Such tools aim to free designers
from 'design fixation' and the limitations of conventional wisdom, thereby allowing
them to explore a huge number of possible proposals for a design problem.
(Bentley1999)
191
1.1 The evolutionary process as a model for design
It is often presumed that, in order to automate any part of the design process, one must
start with a cognitive theory of how humans design. This is based on the assumption
that humans offer the only example of a successful design system. However,
alternative successful design systems do exist. One such system is biological
evolution in nature, which has been evolving biological designs that far exceed any
human designs in terms of complexity, performance and efficiency. Evolutionary
programs use biological evolution in nature as a source of inspiration, rather than a
phenomenon to be accurately modelled. The un-natural evolutionary system maintains
a population of alternative designs.
Genetic algorithms are today the best known and also the most widely used of
all evolutionary programs (Holland 1975). However, a number of researchers in
various specialised fields, including design, have found that most real-world problems
can not be handled with the classical genotype representation and the corresponding
genetic operators (Michalewicz 1996). When applied to complex problems, these
representations and operators do not allow the knowledge within the domain to be
succinctly described. In order to overcome this hurdle, stronger assumptions must be
made about the problem domain. Thus, whereas genetic algorithms require key
elements to remain domain independent, evolutionary systems in fields such as design
will typically allow them to become more complex specialised domain specific
components.
The mapping process from genotype to phenotype is a key area where domain
specific knowledge has been incorporated. The use of genetic algorithms in
optimisation problems required only a straightforward direct mapping from the binary
string genotype to the parameter being optimised that is the phenotype. However, in
attempting to capture a wide range of alternative solutions that differ from each other
in fundamental aspects, evolutionary systems are now employing highly complex
problem specific mapping processes.
In design systems, this process takes the form of a growth process that starts
with encoded design seeds and decodes them into fully developed design proposals.
Peter Bentley emphasises the biological analogy by referring to these growth
processes as embryogenies (Bentley 1999). John Frazer describes them as generative
processes. (Frazer 1979) Such a generative process is a crucial step within the overall
evolutionary design system. The evolutionary manipulation of the population can be
broken down into three stages: generating proposals, making predictions and creating
populations. The stage of making design proposals involves the generative process.
Thus each proposal for a built form is individually generated from the design seed in
response to a simulation of the environment and context within which the built form
would exist. The second stage involves making predictions of how the design
proposal would perform within the environment and context. These can either be
quantitative prediction calculated using various types of analysis software or the may
be qualitative predictions made by evaluation algorithms and human choice. In the
third stage, a new population of design seeds is probabilistically created through the
192
transformation and duplication of those design seeds in the previous generation that
produced the most favourable predictions.
2 THE GENERATIVE PROCESS
The design proposals for built form that are produced by such generative processes
must strike a balance between two opposing criteria. First, the design proposals must
all be recognisable as proposals for built form. The vast majority of three-dimensional
forms have no relationship to built form whatsoever. For example, if one were to
randomly place 'walls', 'floors' and 'roofs' in three-dimensional space, the likelihood of
creating something that resembled built form would be extremely small. If the
generative process were prone to creating such an unnecessarily wide range of
proposals, then the majority would be fundamentally inappropriate, thus rendering the
evolutionary system ineffective. Furthermore, the predictive process of analysing such
widely varying forms would become untenable. Thus, any design seed passing
through the generative process should either abort or result in a design proposal that
could, in a fundamental sense, be interpreted as built form, however inefficient that
proposal might be.
Second, although the range of proposals must not be unnecessarily wide, it
must also not be restrictively narrow. This range must accommodate the required
scope of the evolutionary system as a whole. The scope must therefore be carefully
considered prior to the construction of the generative process. For example, a process
that creates timber-frame two bedroom houses is fine if that is all that is required.
However, if for some reason an extra bedroom is required or an alternative
constructional system is incorporated, the range of proposal would need to be
widened. This would consequently result in the generative process having to be
altered, or even recreated from scratch.
These two criteria affect the decision as to what the scope of the evolutionary
system should be. On the one hand, a generative process that has a very wide scope
will be more generally applicable. This generality will be valuable when new design
problems are tackled requiring proposals of a different kind. On the other hand, an
increase in scope will decrease the quality of the results achieved by the evolutionary
system as a whole.
2.1 Design-schemas
In order to create an evolutionary system that has any chance of producing high
quality design proposals that challenge an experienced architect, the generative
process must be imbued with detailed domain specific knowledge relevant to the
problem at hand. However, there is almost no knowledge that can be desceibed as
being strictly irrelevant to the field of architecture. The knowledge to be captured and
codified must therefore be restricted in some way. One possibility is to restrict the
scope of the generative system to a relatively specific subset or subcategory of all
built form. The question that remains is on what basis should this categorisation be
193
made? From a designers point of view, the most relevant category is the body of work
that they have produced, are producing and will produce in the near future. The key
aspect that unites this category of built form is the particular designers 'style'. John
Frazer describes this personal trademark as a design-schema. “Most designers employ
a methodology highly personalised yet can often be generic when the designer’s body
of work is taken as a whole. It is part of their working method and hence characterises
their ‘style’ by which they are known… This personalised but generic methodology
can be described as a design schema in that it is an abstract conception of what is
common to all designs. Inside the designer’s office, these implicit design-schemas
often become formalised. It is common to find sets of standard details in architects’
offices that serve to economise in time, ensure details are well tested, but also to
ensure a consistency of detailing and to reinforce the house style. In many offices this
extends to design procedures, approaches to organisation and so forth.” (Frazer 2000)
The existence of such personalised design-schemas, custom created by each designer,
has only recently had any effect on architectural design software. In an ideal world,
the most useful piece of software for a particular design project would be highly
specialised for the design-schema being used or even for the design problem. The
ultimate would be the piece of software with a single ‘do it’ button; the button that
always does exactly what you want. In the area of drafting software, a trend towards
more specialised types of software is developing resulting in totally customisable
drawing packages where highly generic core can be specialised to incorporate
templates, libraries of parts, drawing procedures, drawing interfaces and even
customised creation, manipulation and modification tools. The evolutionary paradigm
being presented assumes a similar but more extreme type of specialisation. The
evolutionary search system will remain largely schema independent and will therefore
enjoy general applicability. The generative process used by the evolutionary system
will, however, be highly schema specific. This approach requires methods for creating
generative processes for each design-schema but also for adapting generative
processes to allow for the more or less gradual evolution of existing schemas. These
tools thereby refocused the process of creating design proposals. Rather than directly
creating the proposal, a more abstract approach must initially be taken whereby the
underlying principles of the design-schema must first be defined. Only then can the
actual design proposal be tackled directly. The aim behind creating these design tools
is therefore not to duplicate or mimic existing traditional design process. Rather, the
aim is to create innovative tools that challenge the design process, allowing designers
to work in ways that were previously not possible. The tools to be developed reject
the traditional architectural methodology "on the grounds that, first, the present
architectural design process is fundamentally unsatisfactory in any known form and
not worth imitating and, second, imitating the human process is unlikely in any case
to represent the most imaginative use of a machine." (Frazer 1979)
2.2 Categorisations of built form
This design-schema orientated approach categorises built form according to the
designer that created the particular buildings. This is of course not the only possible
194
categorisation. Alternative choices include divisions that depend on geographic
location, on the constructional system used, on the historic style, on function or
purpose, on phenomenological experience, or on cultural consensus. Even Pevsner's
division of built form into the 'bicycle shed' category and the 'cathedral' category' is a
possible alternative (Pevsner 1976). Some theoreticians like L. Krier, Graves and
Venturi have proposed that the focus should not be on individual buildings but rather
the superordinate category of city and the subordinate category of architectural
element.
One theoretical paradigm that has entertained many of these divisions is the
idea of a type or typology in design. How built form can carry meaning can be
described as the fundamental issue at stake. The various theories surrounding the idea
of type in architecture are manifestations of much more basic cognitive mechanisms
of generalisation and categorisation. Paul Tesar identifies two aspects that are
required for meaning to arise: "The notion of type rests upon two mutually dependent
and conceptually inseparable aspects of: a thing (place, event) that bears some
resemblance to other things, and a group of persons who perceives this likeness and
conceptually subsumes these things as being of the same kind" (Tesar 1991).
The built forms that fall within particular design-schemas can also be
interpreted as types. The 'things' that are similar are a set of built forms being created
by the same design environment. The relevant group of persons is therefore those
people involved in the design process. Thus the design-schema is not a basic category
whose meaning is understood by society in general; instead it is a category whose
meaning is only recognised by a small subset of the design community. The schema
approach has been taken because it supports a personalised creativity not constrained
by past precedent, program, function, construction system, or other factors external to
the design environment. Of course, this does not mean that a generative process based
on a design-schema division excludes the possibility of historical reference, functional
typology or constructional pragmatism. It merely means that these concepts are not
built into the core of the process. For particular projects, further restrictions on the
scope of the generative process might indeed be found to be necessary. However, such
restrictions should remain easily removable and replaceable with new restrictions
without affecting the core design-schema.
2.3 Creativity
Whatever the delimiting factors of the design-schema are, the generative process must
capture and codify as accurately as possible the essence of that design-schema.
Quoting Walter Gropius: "A basic philosophy of design needs first of all a
denominator common to all… Will we succeed in establishing an optical 'key', used
and understood by all, as an objective common denominator of design?" Such a 'key'
would provide "the impersonal basis as a prerequisite for general understanding and
would serve as the controlling agent within the creative act." (Gropius 1962)
However, there is an obvious conflict between the 'controlling agent' and the 'creative
act'. Gropius himself states just prior to this phrase that such a key "can, of course,
never become a recipe or a substitute for art."
195
The conflict of control versus creativity is evocatively exposed by Le Corbusier's
discussion of the Modulor, his attempt at imposing a new system of dimensioning
upon the world. Just as the continuos phenomenon of sound had been cut up "in
accordance with a rule accepted to all, but above all efficient, that is flexible,
adaptable, allowing for a wealth of nuances and yet simple, manageable and easy to
understand", so Le Corbusier proposed to cut up space (Le Corbusier 1955). But he
was acutely aware of the issues of artistic freedom, as is made poignantly clear by his
own strident proclamation that "the 'Modulor' is a working tool, a precision
instrument; a keyboard shall we say, a piano, a tuned piano. The piano has been
tuned: it is up to you to play it well. The 'Modulor' does not confer talent, still less
genius. It does not make the dull subtle: it only proffers them the facility of a sure
measure. But out of the unlimited choice of combinations of the 'Modulor', the choice
is yours.'" (Le Corbusier 1955)
In discussing the Modulor, Robin Evans elucidates this conflict as follows:
"Any rule carries with it the eventual prospect of reduced liberty, but new rules can be
surprisingly unruly, cleaning away customs and habits that have stood in the way for
ages. Thus, for a time, perhaps quiet a long time, new rules can offer a way round the
obvious." (Evans 1994) Generative process that aim to capture and codify particular
design-schemas but that nevertheless also aim to enhance creative freedom should be
seen in this light; as processes that must themselves dynamically change over time in
order not to become the very 'customs and habits' that they initially 'cleaned away'.
3 BUILT FORM
Robin Evans starts his article Figures, Doors and Passages with the phrase "Ordinary
things contain the deepest mysteries" (Evans 1978) Since artificial intelligence was
formally initiated in 1956 at the Dartmouth conference, it has repeatedly fallen victim
to its own seductive claims by underestimating this simple truth. Two dangers are
worth highlighting. First, the wildly over optimistic claims about the future were the
result of underestimating the complexity of the domains that were being tackled.
Typically, general-purpose search mechanisms and reasoning systems were employed
in order to find solutions to highly complex problems for which detailed domain
specific knowledge was essential. Second, the over zealous claims about what has
already been achieved reflect the temptation to assert that the programs were learning,
thinking, being creative, making analogies, understanding stories, and so forth. The
source of these behaviours can invariably be traced to the representations supplied by
the programmer.
When attempting to create a generative process for built forms these two
dangers must be kept in mind. First, the familiarity of built form must not allow one to
assume that it is in anyway straightforward. The implicit and familiar aspects of
buildings must therefore be carefully scrutinised in order to uncover the deep and
implicit structures hidden within. Second, the generative process must be understood
as a computational representation carefully crafted and tuned by the human mind. The
196
aim of this crafting and tuning is to come as close as possible to giving away the
answer without inhibiting the unexpected.
3.1 Why is built form the way it is?
In order to uncover the essence of a particular design-schema, various types of
arguments and rationalisations of built form can be analysed. Such rationalisations
are, in many cases, normative guidelines that describe how built form should be rather
than how it is. In other instances they are analytical devices used to understand
existing built form. Some are highly specific to a small number of built forms. Others
are general theories that purport to be applicable to all built form. In science, theories
of any value must be both highly universal and analytical. However, for the purposes
of creating a generative process of built form all forms of rationalisation will be found
to be worthy of investigation.
For example, the radical analyses of Ronchamp by Robin Evans provide
archetypal examples of a rationalisation of built form that is relevant to only a small
number of buildings by Le Corbusier. It was Le Corbusier who laid down the
challenge within Ronchamp: "The Modulor is everywhere. I defy the visitor to give,
off hand, the dimensions of the different parts of the building." However, after careful
analysis, Evans rejects the Modulor as the generator of the freeform shape of carcass
of the building. Instead, he discovers a hidden generator: the ruled surface. "Not
Modulor measure but rigid lines of ruled surfaces and translated arcs lay behind the
free form of Ronchamp. The Modulor, good fable that it is, tells a story about a
hidden regulatory agency, but is not that agency itself… the ruled surface had usurped
the Modulor. It was a secret; a secret not meant for modest folk" (Evans 1994) If Le
Corbusier had created a generative process for himself, the rationalisation of ruled
surfaces would no doubt have been an important element. What other rationalisations
might he have used? Possibilities include the structural capabilities of reinforced
concrete, the skills of local labour or maybe the path of the sun in the celestial sphere
at that particular location.
The various rationalisations that have been proposed over the last two
thousand years reveals that any theory must first assume a conceptualisation of what
built form is. On the one hand, the order of conceptualisation is concerned with what
built form is, in a factual sense. On the other hand, the order of rationalisation is
concerned with various interrelated ideas and arguments about built form. The latter
presumes the former; arguments and ideas about built form must necessarily make use
of a conceptualisation of built form. Some examples will be given in order to show
how these rationalisations are central to understanding why built form is the way it is.
3.2 Conceptualisations of built form
The conceptualisations used by the various rationalisations range from physical to
abstract. This range of conceptualisations can be usefully categorised into four levels
of abstraction. These will be referred to as material form, spatial form, referential
form and configurational form. A brick wall is a material form that is both tactile and
197
visible. A room defined by four brick walls is a spatial form. Spatial forms are more
abstract in that they are no longer tactile, but they are nevertheless still visible. Walls,
roofs and spaces can be arranged in a certain way to create a built form that can be
labelled as a church. Although a particular church might be both tactile and visible,
the generic concept of what is and what is not a church is much more ephemeral. The
‘church’ phenomenon exists only as conceptual category. The built form of the actual
church refers to the ‘church’ conceptual category. The reference to the conceptual
category is achieved via the built form’s particular arrangement of material and spatial
form. Finally, the church might be part of a monastery consisting of various carefully
related and interconnected spaces. This reveals the most abstract type of form;
configurational form. Whereas the referential form relies on the visual and tactile
properties of built form, configurational form relies on the interrelationships,
adjacencies, proximities and views between the various spatial forms. Configurational
form thus consists of various types of connectivity between spatial forms.
3.3 Rationalisations of built form
Procedural rationalisations attempt to explain built form by focusing on various
processes that lead up to the final product. Included are such long-term processes as
the broad paradigm shifts in worldviews, developments in the socio-cultural condition
of society, the gradual evolution of various architectural styles and changes in
procedures and techniques of building construction. Shorter-term processes include
the design of the particular building, the actual sequence of construction for the
building and the transformation and alteration of the building after it has been
completed.
In Figures, Doors and Passages Robin Evans gives an illustrative example of
how cultural transformations have affected configurational form. Evans aims to shows
how the configurational form of domestic architecture embodies concepts such as
privacy, comfort and independence that are dependent on social, cultural and
technological conditions. Evans starts by examining the Villa Madama, designed by
Raphael and Antionio da Sangallo and partly completed in the first half of the
sixteenth-century. He notices two characteristics in the plan that "we would nowadays
never do." First, the rooms tend to have many doors. Second, there is no distinction
between circulation spaces and inhabited spaces. A complete inversion is shown to
have occurred in what was perceived to be convenient that was reflected in the
configurational form of domestic architecture; from the matrix of highly permeable
interconnected rooms to corridor systems serving terminal rooms (Evans 1978).
Environmental rationalisations attempt to explain built form by focusing on
the influence of the environment upon the built form. The largest of these
environments is the cosmos and its universal laws. Throughout history, many
theoretical systems of laws and rules have been proposed as being universal.
Inevitably, in each case it was discovered that these laws and rules were not quiet so
universal after all and were in fact highly relativist. At a more planetary scale,
included are those influences that are specific to what Buckminster Fuller calls
spaceship earth. Particularly influential at this scale are factors such as the properties
198
of global materials like steel and glass, the influence of gravity and 'global village'
cultural tendencies that transcend the local culture. Below the planetary scale, ever
decreasing neighbourhoods of influence can be identified. At the scale of the country
or state, building regulations might be paramount. At the scale of the local city or
town, the abundant availability of certain building materials and associated skills
might be most important. Finally, of most relevance at the scale of the particular site
might be the shape of the building next door or the cost of land per square meter.
An example of an environmental rationalisation that affects material form
concerns the availability of building materials and building systems in association
with the skills required to manipulate them. In a vernacular sense, this is often
translated to the use of certain local stone or perhaps the skills involved in making a
thatch roof. However, such local craft based labour intensive building trades have all
but disappeared. Instead they have been replace by a group of organisations and
industries performing widely differing functions. Groak splits the parties involved in
the building culture into four sectors - the client organisation, the manufacturers, the
designer/specifiers and the assemblers. (Groak 1992) Of these four sectors, it is the
manufacturing industry that will define what materials are available for use and it
therefore has a large impact on the material form. Groak writes: "There is an
extraordinary array of suppliers of materials - such as cement, timber, bricks, thermal
insulants, impervious membranes, etc. - and manufacturers of components - such as
window-wall assemblies, roof trusses, radiators and boilers, switches and cabling,
door and window furniture, etc." (Groak 1992) The design of material form has, to a
large extent, become a question of choosing and combining such manufactured and
prefabricated materials and components.
Intentional rationalisation attempts to argue that the built form is, to some
degree, a result of the purpose, function or target performance of a built form. In the
field of architecture, the Functionalist movement of the first half of the twentieth
century has been a highly conspicuous form of intentional rationalisation. The
limitations of Functionalism as an architectural theory has long since been
acknowledge. However, the idea that the form of an artefact is in some way related to
'what it is for' has not been rejected. Furthermore, the interpretation of 'what it is for'
has become much broader, encompassing such issues as cultural and social
appropriateness and change over time. The term 'intended function' refers to this
broader concept of the intended purpose, function or target performance of built form.
An example of an intentional rationalisation that affects both spatial and
configurational form is what is referred to by Bill Hillier as 'generic function'. Hillier
argues that the spatial and configurational form of all buildings is constrained by very
specific kinds of laws which relate 'generic function'. "Generic function refers not to
the different activities that people carry out in buildings or the different functional
programmes that buildings of different kinds accommodate, but to aspects of human
occupancy of buildings that are prior to any of these: that to occupy space means to be
aware of the relationship of space to others, that to occupy a building means to move
about in it, and to move about in a building depends on being able to retain an
intelligible picture of it. Intelligibility and functionality defined as formal properties
of spatial complexes are the key 'generic functions', and as such the key structures
199
which restrict the field of combinatorial possibility and give rise to the architecturally
real" (Hillier 1996).
4 CONCLUSION
Research into the use of evolutionary system in design has been steadily increasing.
This paper argues that the generative process within such evolutionary systems must
focus on the designer's highly personalised design-schema. In order to uncover the
essence of an architectural design-schema, it was suggested that implicit and familiar
aspects of buildings must be scrutinised in order to reveal the knowledge that is
hidden within them. This approach is based on the fact that domain specific
knowledge must be the foundation of any design system that aims to challenge the
designer. This paper therefore advocates focusing on a relatively limited subcategory
of built form that can be captured and codified within a generative system.
Nevertheless, this approach does not exclude the search for that which is
common to all generative processes of form. The current approach envisages the
development of a large number of generative processes, each specialised to its own
particular design-schema. However, in the long term these systems will inevitably
have certain qualities in common. In effect, generators that transcend any one
particular generative process. For example, all design domains make use of the
concept of symmetry. The number of symmetry groups in one, two and three
dimensions are universal and timeless and might therefore be presumed to be present
in a similar form within all generative process.
What is the best method of uncovering such universal generators? One
approach to this quest might propose a long-term centralised research initiative that
attempts to uncover the universal generators of all design-schemas. However, this
strategy would be unlikely to produce any short-term tangible results beyond a
proliferation of conference papers. An alternative approach might propose an
evolutionary methodology suggesting that the most effective way forward is to ignore
the big picture and instead allow parallel development of a wide range of domain
specific generative processes. In time, the universal generators will become obvious,
as they will be the common denominators of the surviving generative processes. What
is proposed is to employ this short-term methodology in tandem with a long-term
overview. While creating various generative processes for particular design-schemas,
the relative generality of each generator within the system will be analysed.
Generators can be rated from the most specific to the most universal. Eventually,
some generators might be discovered to be immutable laws of a universal reality.
5 ACKNOWLEDGEMENTS
Our research project is supported by a UGC PhD project grant from the Hong Kong
Polytechnic University.
200
6 REFERENCES
Bentley, P. (1999) An Introduction to Evolutionary Design By Computers, in Bentley,
P. (ed.), Evolutionary Design by Computers, Morgan. Kaufmann Publishers
Inc, San Fransisco, pp. 1-73
Evans, R. (1978) Figures, Doors and Passages, in Translations From Drawings to
Buildings and Other Essays, Architectural Association Publications, London
1997.
Evans, R. (1995) The Projective Cast: Architecture and its Three Geometries,
Massachusetts Institute of Technology.
Frazer, J., Connor, J and J. (1979) A Conceptual Seeding Technique for Architectural
Design, PArC79, Proceedings of International Conference on the Application
of Computers in Architectural Design, Berlin, Online Conferences with AMK.
pp 425-34.
Frazer, J. (1995) An Evolutionary Architecture, Architectural Association
Publications.
Groak, S. (1992) The Idea of a Building, E & FN Spon, London.
Gropius, W. (1962) The Scope of Total Architecture Collier Books, New York
Hillier, B (1996) Space is the Machine, Cambridge University Press, Cambridge.
Holland, J. (1975) Adaptation in Natural and Artificial Systems. The University of
Michigan Press, Ann Arbor.
Le Corbusier (1955) Modulor II in Peter de Francia and Anna Bostock (trans.)
Modulor I and II, Cambridge, Mass., 1982.
Michalewicz, Z. (1996) Genetic Algorithns + Data Structures = Evolution Programs.
Springer-Verlag.
Pevsner, N. (1976) A History of Building Types, Princeton University Press,
Princeton.
Tesar, P. (1991) The Other Side of Types, in Rockcastle, G. (ed.), Type and the
(Im)possibilities of Convention, University of Minnesota College of
Architecture and Landscape Architecture, Minnesota, pp. 165-175.
... According to Fischer (2007), the computer (and specifically a parametric approach) can either 'enable' or 'restrict' the design process, depending on its use as a 'medium' or as a 'tool'. When taking into account the way in which parametric models are constructed, Janssen et al. (2000) defined a creative solution as one which is not known beforehand, is useful for the project, and is not a mere application of a formula. ...
Article
Full-text available
Parametric methods from different modeling paradigms were proposed to students in a design studio. The influence of these methods on creativity was studied using qualitative methodology. ?Generation? was found to stimulate ?abundance? of ideas and to lead to different evolution and flexibility, as well as to higher ?engagement? in the design process.
... The key aspect that unites this category is the particular designer's 'style'. This personal trademark has been described as a 'design-schema' (Janssen 2000, Frazer 2001). The design schema should not be interpreted as a superficial quality merely concerned with form and appearance of a design proposal. ...
Article
Full-text available
Teaching methodologies for parametric design are being researched all over the world, since there is a growing demand for computer programming logic and its fabrication process in architectural education. The computer programming courses in architectural education are usually done in a very short period of time, and so students have no chance to create their own designs. This paper describes a course in which metaphors are used as a teaching methodology in parametric design, in order to let students create their own designs and learn the basic elements of parametric programming language in a short period of time with deductive reasoning. In this course, it was intended to teach visual programming language to undergraduates. Advancing under the metaphor theoretical framework, the students obtained experience in achieving form-finding process for their projects in accord with the certain constraints. Using this methodology, the students, who experienced all design stages from 3D modeling to the digital fabrication, additionally were able to develop their ability for versatile thinking and the use of more than one tool in combination, in the early years of their architectural education.
Book
Winner of the 1997 Alice Davis Hitchcock MedallionAnyone reviewing thehistory of architectural theory, Robin Evans observes, would have to conclude thatarchitects do not produce geometry, but rather consume it. In this long-awaitedbook, completed shortly before its author's death, Evans recasts the idea of therelationship between geometry and architecture, drawing on mathematics, engineering, art history, and aesthetics to uncover processes in the imagining and realizing ofarchitectural form. He shows that geometry does not always play a stolid and dormantrole but, in fact, may be an active agent in the links between thinking andimagination, imagination and drawing, drawing and building. He suggests a theory ofarchitecture that is based on the many transactions between architecture andgeometry as evidenced in individual buildings, largely in Europe, from the fifteenthto the twentieth century.From the Henry VII chapel at Westminster Abbey to LeCorbusier's Ronchamp, from Raphael's S. Eligio and the work of Piero della Francescaand Philibert Delorme to Guarino Guarini and the painters of cubism, Evans exploresthe geometries involved, asking whether they are in fact the stable underpinnings ofthe creative, intuitive, or rhetorical aspects of architecture. In particular heconcentrates on the history of architectural projection, the geometry of vision thathas become an internalized and pervasive pictorial method of construction and that, until now, has played only a small part in the development of architecturaltheory.Evans describes the ambivalent role that pictures play in architecture andurges resistance to the idea that pictures provide all that architects need, suggesting that there is much more within the scope of the architect's vision of aproject than what can be drawn. He defines the different fields of projectivetransmission that concern architecture, and investigates the ambiguities ofprojection and the interaction of imagination with projection and itsmetaphors.
Article
In "An Evolutionary Architecture", John Frazer presents an overview of his work for the past 30 years. Attempting to develop a theoretical basis for architecture using analogies with nature's processes of evolution and morphogenesis. Frazer's vision of the future of architecture is to construct organic buildings. Thermodynamically open systems which are more environmentally aware and sustainable physically, sociologically and economically. The range of topics which Frazer discusses is a good illustration of the breadth and depth of the evolutionary design problem. Environmental Modelling One of the first topics dealt with is the importance of environmental modelling within the design process. Frazer shows how environmental modelling is often misused or misinterpreted by architects with particular reference to solar modelling. From the discussion given it would seem that simplifications of the environmental models is the prime culprit resulting in misinterpretation and misuse. The simplifications are understandable given the amount of information needed for accurate modelling. By simplifying the model of the environmental conditions the architect is able to make informed judgments within reasonable amounts of time and effort. Unfortunately the simplications result in errors which compound and cause the resulting structures to fall short of their anticipated performance. Frazer obviously believes that the computer can be a great aid in the harnessing of environmental modelling data, providing that the same simplifying assumptions are not made and that better models and interfaces are possible. Physical Modelling Physical modelling has played an important role in Frazer's research. Leading to the construction of several novel machine readable interactive models, ranging from lego-like building blocks to beermat cellular automata and wall partitioning systems. Ultimately this line of research has led to the Universal Constructor and the Universal Interactor. The Universal Constructor The Universal Constructor features on the cover of the book. It consists of a base plug-board, called the "landscape", on top of which "smart" blocks, or cells, can be stacked vertically. The cells are individually identified and can communicate with neighbours above and below. Cells communicate with users through a bank of LEDs displaying the current state of the cell. The whole structure is machine readable and so can be interpreted by a computer. The computer can interpret the states of the cells as either colour or geometrical transformations allowing a wide range of possible interpretations. The user interacts with the computer display through direct manipulation of the cells. The computer can communicate and even direct the actions of the user through feedback with the cells to display various states. The direct manipulation of the cells encourages experimentation by the user and demonstrates basic concepts of the system. The Universal Interactor The Universal Interactor is a whole series of experimental projects investigating novel input and output devices. All of the devices speak a common binary language and so can communicate through a mediating central hub. The result is that input, from say a body-suit, can be used to drive the out of a sound system or vice versa. The Universal Interactor opens up many possibilities for expression when using a CAD system that may at first seem very strange.However, some of these feedback systems may prove superior in the hands of skilled technicians than more standard devices. Imagine how a musician might be able to devise structures by playing melodies which express the character. Of course the interpretation of input in this form poses a difficult problem which will take a great deal of research to achieve. The Universal Interactor has been used to provide environmental feedback to affect the development of evolving genetic codes. The feedback given by the Universal Interactor has been used to guide selection of individuals from a population. Adaptive Computing Frazer completes his introduction to the range of tools used in his research by giving a brief tour of adaptive computing techniques. Covering topics including cellular automata, genetic algorithms, classifier systems and artificial evolution. Cellular Automata As previously mentioned Frazer has done some work using cellular automata in both physical and simulated environments. Frazer discusses how surprisingly complex behaviour can result from the simple local rules executed by cellular automata. Cellular automata are also capable of computation, in fact able to perform any computation possible by a finite state machine. Note that this does not mean that cellular automata are capable of any general computation as this would require the construction of a Turing machine which is beyond the capabilities of a finite state machine. Genetic Algorithms Genetic algorithms were first presented by Holland and since have become a important tool for many researchers in various areas.Originally developed for problem-solving and optimization problems with clearly stated criteria and goals. Frazer fails to mention one of the most important differences between genetic algorithms and other adaptive problem-solving techniques, ie. neural networks. Genetic algorithms have the advantage that criteria can be clearly stated and controlled within the fitness function. The learning by example which neural networks rely upon does not afford this level of control over what is to be learned. Classifier Systems Holland went on to develop genetic algorithms into classifier systems. Classifier systems are more focussed upon the problem of learning appropriate responses to stimuli, than searching for solutions to problems. Classifier systems receive information from the environment and respond according to rules, or classifiers. Successful classifiers are rewarded, creating a reinforcement learning environment. Obviously, the mapping between classifier systems and the cybernetic view of organisms sensing, processing and responding to environmental stimuli is strong. It would seem that a central process similar to a classifier system would be appropriate at the core of an organic building. Learning appropriate responses to environmental conditions over time. Artificial Evolution Artificial evolution traces it's roots back to the Biomorph program which was described by Dawkins in his book "The Blind Watchmaker". Essentially, artificial evolution requires that a user supplements the standard fitness function in genetic algorithms to guide evolution. The user may provide selection pressures which are unquantifiable in a stated problem and thus provide a means for dealing ill-defined criteria. Frazer notes that solving problems with ill-defined criteria using artificial evolution seriously limits the scope of problems that can be tackled. The reliance upon user interaction in artificial evolution reduces the practical size of populations and the duration of evolutionary runs. Coding Schemes Frazer goes on to discuss the encoding of architectural designs and their subsequent evolution. Introducing two major systems, the Reptile system and the Universal State Space Modeller. Blueprint vs. Recipe Frazer points out the inadequacies of using standard "blueprint" design techniques in developing organic structures. Using a "recipe" to describe the process of constructing a building is presented as an alternative. Recipes for construction are discussed with reference to the analogous process description given by DNA to construct an organism. The Reptile System The Reptile System is an ingenious construction set capable of producing a wide range of structures using just two simple components. Frazer saw the advantages of this system for rule-based and evolutionary systems in the compactness of structure descriptions. Compactness was essential for the early computational work when computer memory and storage space was scarce. However, compact representations such as those described form very rugged fitness landscapes which are not well suited to evolutionary search techniques. Structures are created from an initial "seed" or minimal construction, for example a compact spherical structure. The seed is then manipulated using a series of processes or transformations, for example stretching, shearing or bending. The structure would grow according to the transformations applied to it. Obviously, the transformations could be a predetermined sequence of actions which would always yield the same final structure given the same initial seed. Alternatively, the series of transformations applied could be environmentally sensitive resulting in forms which were also sensitive to their location. The idea of taking a geometrical form as a seed and transforming it using a series of processes to create complex structures is similar in many ways to the early work of Latham creating large morphological charts. Latham went on to develop his ideas into the "Mutator" system which he used to create organic artworks. Generalising the Reptile System Frazer has proposed a generalised version of the Reptile System to tackle more realistic building problems. Generating the seed or minimal configuration from design requirements automatically. From this starting point (or set of starting points) solutions could be evolved using artificial evolution. Quantifiable and specific aspects of the design brief define the formal criteria which are used as a standard fitness function. Non-quantifiable criteria, including aesthetic judgments, are evaluated by the user. The proposed system would be able to learn successful strategies for satisfying both formal and user criteria. In doing so the system would become a personalised tool of the designer. A personal assistant which would be able to anticipate aesthetic judgements and other criteria by employing previously successful strategies. Ultimately, this is a similar concept to Negroponte's "Architecture Machine" which he proposed would be computer system so personalised so as to be almost unusable by other people. The Universal State Space Modeller The Universal State Space Modeller is the basis of Frazer's current work. It is a system which can be used to model any structure, hence the universal claim in it's title. The datastructure underlying the modeller is a state space of scaleless logical points, called motes. Motes are arranged in a close-packing sphere arrangement, which makes each one equidistant from it's twelve neighbours. Any point can be broken down into a self-similar tetrahedral structure of logical points. Giving the state space a fractal nature which allows modelling at many different levels at once. Each mote can be thought of as analogous to a cell in a biological organism. Every mote carries a copy of the architectural genetic code in the same way that each cell within a organism carries a copy of it's DNA. The genetic code of a mote is stored as a sequence of binary "morons" which are grouped together into spatial configurations which are interpreted as the state of the mote. The developmental process begins with a seed. The seed develops through cellular duplication according to the rules of the genetic code. In the beginning the seed develops mainly in response to the internal genetic code, but as the development progresses the environment plays a greater role. Cells communicate by passing messages to their immediate twelve neighbours. However, it can send messages directed at remote cells, without knowledge of it's spatial relationship. During the development cells take on specialised functions, including environmental sensors or producers of raw materials. The resulting system is process driven, without presupposing the existence of a construction set to use. The datastructure can be interpreted in many ways to derive various phenotypes. The resulting structure is a by-product of the cellular activity during development and in response to the environment. As such the resulting structures have much in common with living organisms which are also the emergent result or by-product of local cellular activity. Primordial Architectural Soups To conclude, Frazer presents some of the most recent work done, evolving fundamental structures using limited raw materials, an initial seed and massive feedback. Frazer proposes to go further and do away with the need for initial seed and start with a primordial soup of basic architectural concepts. The research is attempting to evolve the starting conditions and evolutionary processes without any preconditions. Is there enough time to evolve a complex system from the basic building blocks which Frazer proposes? The computational complexity of the task being embarked upon is not discussed. There is an implicit assumption that the "superb tactics" of natural selection are enough to cut through the complexity of the task. However, Kauffman has shown how self-organisation plays a major role in the early development of replicating systems which we may call alive. Natural selection requires a solid basis upon which it can act. Is the primordial soup which Frazer proposes of the correct constitution to support self-organisation? Kauffman suggests that one of the most important attributes of a primordial soup to be capable of self-organisation is the need for a complex network of catalysts and the controlling mechanisms to stop the reactions from going supracritical. Can such a network be provided of primitive architectural concepts? What does it mean to have a catalyst in this domain? Conclusion Frazer shows some interesting work both in the areas of evolutionary design and self-organising systems. It is obvious from his work that he sympathizes with the opinions put forward by Kauffman that the order found in living organisms comes from both external evolutionary pressure and internal self-organisation. His final remarks underly this by paraphrasing the words of Kauffman, that life is always to found on the edge of chaos. By the "edge of chaos" Kauffman is referring to the area within the ordered regime of a system close to the "phase transition" to chaotic behaviour. Unfortunately, Frazer does not demonstrate that the systems he has presented have the necessary qualities to derive useful order at the edge of chaos. He does not demonstrate, as Kauffman does repeatedly, that there exists a "phase transition" between ordered and chaotic regimes of his systems. He also does not make any studies of the relationship of useful forms generated by his work to phase transition regions of his systems should they exist. If we are to find an organic architecture, in more than name alone, it is surely to reside close to the phase transition of the construction system of which is it built. Only there, if we are to believe Kauffman, are we to find useful order together with environmentally sensitive and thermodynamically open systems which can approach the utility of living organisms.
A Conceptual Seeding Technique for Architectural Design, PArC79
  • J Frazer
  • J Connor
Frazer, J., Connor, J and J. (1979) A Conceptual Seeding Technique for Architectural Design, PArC79, Proceedings of International Conference on the Application of Computers in Architectural Design, Berlin, Online Conferences with AMK. pp 425-34.
Modulor II in Peter de Francia and Anna Bostock (trans.) Modulor I and II
  • Le Corbusier
Le Corbusier (1955) Modulor II in Peter de Francia and Anna Bostock (trans.) Modulor I and II, Cambridge, Mass., 1982.
Type and the (Im)possibilities of Convention
  • P Tesar
Tesar, P. (1991) The Other Side of Types, in Rockcastle, G. (ed.), Type and the (Im)possibilities of Convention, University of Minnesota College of Architecture and Landscape Architecture, Minnesota, pp. 165-175.