Thinking difference: Theories and models of

parametric design thinking

Rivka Oxman, Faculty of Architecture and Town Planning, Technion Israel
Institute of Technology, Haifa, 32000, Israel

The paper examines the uniqueness of seminal parametric design concepts, and
their impact on models of parametric design thinking (PDT). The continuity
and change within the evolution of design thinking is explored through review of
key texts and theoretical concepts from early cognitive models up to current
models of parametric design thinking. It is proposed that the seminal role for
parametric schema, as a strategic medium of parametric design thinking, is
formulated at the intersection of three bodies of knowledge: cognitive models of
typological and topological design in architecture; process models of digital
design; and tectonic order of material fabrication design (MFD).
Differentiation is introduced as a key design strategy of PDT and is
demonstrated through classification of prominent case studies.
Ó 2017 Elsevier Ltd. All rights reserved.

Keywords: computational models, parametric design, design cognition, design

knowledge, parametric design thinking

B
eyond being another tool for modeling complex forms, parametric
design is emerging as a unique and distinctive model of design.
Both research and praxis in parametric design are influencing the
emergence of parametric design theories that are currently undergoing a re-
formulation and an epistemological shift. In parallel, the development of cur-
rent tools and practices of parametric design are beginning to impact forms of
parametric design thinking (PDT). Current parametric design systems are
adapting to changing context under the impact of a new generation of script-
ing languages and techniques (Burry, 2011), relational topological schema,
associative geometries, and re-editing processes (Woodbury, 2010; Jabi,
2015) and computational process models of digital design (Oxman, 2006).

This theoretical review of parametric design thinking (PDT) is motivated by the

need to explore the uniqueness of parametric design methods, techniques, me-
dia and tools within the field of design studies in order to understand the
impact of parametric design on the emergence of new ways of thinking in
design.
Corresponding author:
Rivka Oxman In parallel with the chronological development of media and tools in design,
rivkao@gmail.com the theories and concepts of design thinking have evolved through both
www.elsevier.com/locate/destud
0142-694X Design Studies -- (2017) --e--
http://dx.doi.org/10.1016/j.destud.2017.06.001 1
Ó 2017 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 continuity and change in relation to accepted paradigms widely recognized
scientifically as ways of design thinking. The first scientific foundation of
design as a way of thinking appeared as a problem-solving paradigm in The
Sciences of the Artificial (Simon, 1969). A body of ideas presented by Robert
McKim in Experiences in Visual Thinking (McKim, 1972) focused on new vi-
sual aspects of design thinking. Ways of thinking in the discipline of architec-
tural design were demonstrated by Bryan Lawson in How Designers Think
(Lawson, 1980), and by Peter Rowe in Design Thinking (Rowe, 1987). These
works presented theories and methods of design that were considered basic
research in design studies. In the following years, design research has devel-
oped the relationships between design, cognition and computation, which
have become an important topic in design studies (Oxman, 1994; Oxman &
Gero, 1987; Oxman & Oxman, 1992).

Design thinking has been defined as a process of ‘creative strategies which de-
signers utilize during the process of designing’ (Visser, 2006); it has recently been
proposed as ‘a process of exploration and creative strategies’ in all design do-
mains and has been recognized as a new field in other emerging design prac-
tices (Dorst, 2012). Following the evolution of cognitive and computational
design research, the present study seeks to formulate the relationship between
selected models of design thinking and the impact of currently emerging
computational media. This is in order to characterize and evaluate the influ-
ence of novel digital design tools and technologies of parametric design in pro-
ducing distinctive novel forms of thinking in design.

In order to formulate the uniqueness of parametric design thinking, a system-

atic theoretical study is presented below comparing selected cognitive con-
cepts in both traditional paper-based models of design thinking and current
computational models of parametric design. The systematic analysis and defi-
nition of a chronological process of evolution of models of design thinking is
presented as part of the continuity and change in the evolution of models of
design thinking (from cognitive models to computational models to models of
parametric design thinking). Derived from this evolutionary framework of
design thinking, the key concepts and principles of parametric design thinking
are introduced, defined and illustrated. The process of tracing the transforma-
tion from traditional paper-based media to computational media-related
models of design thinking has enabled the systematic identification of current
seminal concepts and principles in parametric design thinking. The body of
generic concepts that have been found to be associated with media-related
strategies of design creativity in prominent models of design thinking have
also provided new perspectives and understanding of thinking processes in
design (Section 1).

As a result of the comparative study, unique cognitive concepts and principles

in models of parametric design thinking such as: Generic Schema; Relational

2 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 Logic; and Associative Relationships; have been identified and framed in the
current research as emerging models and methods of parametric design
thinking. The research study has focused around the intersection of three
research areas: parametric design models and tools, cognitive models of archi-
tectural design knowledge, and process models of digital design. The work
has identified concepts and principles of generic schema in PDT in each
of these areas of knowledge: the generic parametric schema (Section 2);
the knowledge-based design parametric schema (Section 3); and the
generic schema of information flow and process models of digital design (Sec-
tion 4).

Finally, the concept of differentiation, as part of a current body of disciplinary

knowledge in architectural design, is addressed within parametric design. The
processes of thinking differentiation are presented and demonstrated by
selected prominent architectural examples (Sections 6 and 7). Lastly, the
uniqueness of PDT as a model of design thinking, its limitations and future po-
tential, are presented and discussed in the concluding section.

1 Evolution of models of design thinking: from cognitive

to computational to parametric design thinking
Due to the rapid development of the role of digital media in design practice,
the study and understanding of the relationship between cognitive models
and computational processes in design has gained important research priority
in design studies (Oxman, 1995). In order to represent and understand the
uniqueness of PDT a theoretical study comparing the evolution of models
of design thinking was initiated. To conduct such a comparative study of
different theories and models of design thinking it is necessary to select and
define the cognitive activities that are among the most coherent and distinctive
components which characterize the main activities in the disciplines of design.

Creativity has been recognized as one of the main cognitive activities in design
studies that is addressed by almost all theories of design thinking (Boden,
1990; Cross, 1997). According to Dorst and Cross (2001) ‘in every design proj-
ect creativity can be found, if not in the apparent form of a distinct creative event,
then as the evolution of a unique solution possessing some degree of creativity’.
Based upon this assumption, accepted cognitive concepts of creativity in
different models of creative design were compared.

The main categories of creativity in design are shared by leading theories of

design thinking. For example, the concept of exploration space in design
was found as one of the main concepts appearing in all theories of design
thinking. While the intrinsic concepts of creativity and innovation referred

Theories and models of parametric design thinking 3

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 to in the semantic content of the term designerly ways of knowing (Cross, 1982,
2001, 2006, 2011) are shared by all design disciplines and across design scales,
the strategies and methods for exploring alternatives in a solution space may
be unique depending on the type of media technology.

The evolution of such key concepts has rapidly developed to adapt to the cur-
rent emerging technology of computational media. Thus it can be generally
observed historically that from hand-drawing and sketching, to code-based
scripting in digital design, and up to the current prominence of parametric
design thinking, emerging media and technology-related models of design
have resulted in the rapid development and change of the concepts, content
and procedures of design thinking.

The analytic objective of this review of theory has been to identify the evolu-
tion of design thinking from its origins to the current state of parametric
design thinking. Selected representative theories and models of design thinking
are presented and discussed in the following sections. Key cognitive concepts
of design thinking are analyzed in each of the prominent models of design
thinking. These include reference to a consistent set of concepts such as search
of solution space, exploration, emergence, reflection, modification, refinement,
adaptation and media including current computational design concepts such
as algorithmic design, scripting languages, etc. This tracing of the evolution
of the set of relevant concepts in the sequence of models demonstrates the
impact of new technologies of computational media on the processes of para-
metric design thinking.

1.1 Early models of design thinking ‘Designerly Ways of

Knowing’ (Cross, 1982, 2001, 2006, 2011)
These are among the early cognitive models of design thinking that appeared
in the 1980’s. Prominent among the large body of research in design thinking
are observations related to personal introspective thinking in sketch-based me-
dia. Among the objectives of observational research methods of designer’s
behavior was to define how the designer works through the complex iterative
cognitive processes of conceptualization, modification and refinement in a so-
lution space. Reflection upon the result of a design action has from the begin-
ning been a central concept in a personal design process where reflection means
identifying the good results and the bad results of the sketch ideas at any time.
Cross further identified sketch modification as part of a characteristic process
of refinement of a design idea. The linked cycle of design, reflection, modifica-
tion and re-design in an iterative process of successive refinement cycles of the
design idea was the archetype of cognitive models of design thinking as a
network of actions and reflection that exists as a significant stream of design
research until today.

4 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 1.2 Reflective practice as epistemological model ‘Reflection
in Action’ (Sch€
on, 1983, 1987, 1988)
Sch€on’s concept of reflection termed, Reflection in Action (Sch€ on, 1983, 1987,
1988), was another important original foundational concept in cognitive ap-
proaches to design thinking. The systematized depiction of the iterative pro-
cess through observation and visual documentation of the design process is
usually accomplished through observation by protocol analysis. In this
method the formalization and graphical documentation of the iterative cycles
of design reflection are followed by framing the situation by the designer and
evaluating a design action move for solution refinement as good or bad, and
moving forward to the next action. For Sch€ on the move is the action response
to reflection. The process proceeds through further acts of naming and framing
in which the design problem context imposes on the situation a coherence that
guides subsequent moves. In Sch€ on’s formalization of design processes, refine-
ment, or modification, is part of a moving-seeing-moving process. Design refine-
ment through a series of moves is characterized as a kind of knowing process of
moving-seeing-moving described variously as fast moving . thinking on your
feet while recorded and supported by the medium of paper-based sketches
(Cross, 2011).

A protocol of patterns of reasoning by designers using design rules can be

derived from underlying types, references, spatial gestalts, and experiential ar-
chetypes that represent design knowledge or holding environments for reflec-
tion in action (Sch€
on, 1988). Thus Sch€ on’s theories and research procedures
included the establishment of important concepts and research methods for
the future of design cognition research.

1.3 Design thinking in early computer-aided design models

‘Architecture, Media and Knowledge’ (McCullough,
Mitchell, and Purcell, 1990)
With the gradual departure from paper-based design media since the 1980’s
until today there is a growing and profound series of changes in models of
design thinking. The elements of this evolution occurred first as media techno-
logical developments. The evolution of computer-integrated design from one-
off design modeling systems to designer supportive design environments and
eventually to the truly generative geometric design machines (that parametric
systems are rapidly becoming) has witnessed a substantive transition in the na-
ture of design methods and processes. In the evolving relationship between
man and machine the profile of the designerly thinker, including the required
knowledge base, the skill set, the conceptual principles, and leading models
of design have been totally transformed. Another contribution in defining
this evolutionary process is the identification of the continuing role of the epis-
temological foundations of disciplinary knowledge such as typological knowl-
edge in evolving models of design thinking (Oxman & Oxman, 1992).

Theories and models of parametric design thinking 5

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 Early CAD systems were generally a representational medium for the 2-
dimensional and 3-dimensional modeling of designs that were sufficiently
geometrically well-conceived and visualized so that they might be computa-
tionally modeled. Initially, within the limitations of Euclidian geometry, the
CAD (or CAAD) modelers provided an interface for modifying views,
enabling the possibility of walk-throughs, etc. Thus the ability to explore the
one-off design was highly enhanced. Originally there existed no user interface
to support a re-editing of reflection and processes of re-representation of
computational 2D drafting and 3D modeling. With the technological and func-
tional advances from CAAD to Parametric Design Systems we can observe the
continuous mapping of early cognitive models of adaptation, or re-modeling,
into current re-editing of code-structures and models in parametric systems
which support contemporary interactive parametric models of design thinking.

1.4 Artificial Intelligence and cognitive models

‘Knowledge-based Computational Models of Design’
(Oxman, 1994, 2004; Oxman & Gero, 1987; Oxman &
Oxman, 1992)
This period of far-reaching technological developments has raised challenging
questions related to the impact of electronic media on new formulations of
models of design process and design thinking. Early research and development
on Artificial Intelligence (AI), and Knowledge-based Design (KBD), began to
provide perspectives related to cognitive structures, algorithmic systems and
computational mechanisms in design creativity and innovation. The cognitive
depth of these models of AI in design was to later become absorbed into the
development of advanced generations of computational design environments.

In early models of AI in design research, computational media associated with

knowledge-based design were applied to formulate and represent design knowl-
edge such as experiential knowledge and precedent-based knowledge (Oxman,
2004; Oxman & Gero, 1987). Models of cognitive processes of design thinking
such as refinement and adaptation were developed through computational
search mechanism of design. Knowledge types and sub-types related to generic
typological design knowledge were computationally formulated by employing
AI design media and tools (Oxman & Oxman, 1992). Such knowledge repre-
sentation capabilities were later to be incorporated into parametric design sys-
tems (Oxman & Gu, 2015).

1.5 Advanced computational design, algorithmic thinking and

scripting ‘Scripting Cultures’ (Burry, 2011); ‘Form þ Code’
(Reas & McWilliams, 2010)
The transition from traditional CAD to Computational Design to Theories of
the Digital in Architecture, (Oxman, R.E, and Oxman, R.M., 2014) has
contributed to contemporary models and methods of design thinking. Digital

6 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 media and computational design systems have integrated processes of simula-
tion, evaluation and fabrication within designer-authored scripted computa-
tional processes. Relevant principles from domains of science and
philosophy, mathematics and computer science have introduced new ways
of thinking and morphogenetic processes (Menges and Ahlquist, 2011;
Oxman, 2015). This reformulation of objectives and scientific foundations
has resulted in adding new principles, methods and processes to conventional
cognitive models of design thinking.

The integration of materialization technology including 3D fabrication and

robotics has advanced design to include both designing and making as integral
components in models of computational design thinking. Tectonic relation-
ships are informed by integrating design and materialization in the early
phases of conceptual design. The selection of material and materialization
techniques in early design stages is supported today by holistic process of in-
formation flow from concept to production. Furthermore, tectonic content
such as structure and material is becoming part of a novel type of design
thinking in what is currently termed, Material-based Design (Oxman, 2012).
The traditional role of the visual strokes of the designer’s pencil has now
been replaced by the computer’s algorithmic mode of thinking. Algorithmic
Thinking can be defined as a set of rules written by a source code of explicit
instructions that initiate computational procedures that generate digital forms.
Writing algorithmic code has become a fundamental component of a design-
erly way of knowing in models of algorithmic design. Scripting, or writing
code, provides a new way of design thinking. It demands the development
of cognitive and computational skill for understanding the formalization, rep-
resentation, and coding of design procedures. ‘Software is a tool for the mind’
(Reas & Fry, 2014). Scripting software programing languages allow the
designer to adapt, customize and reconfigure the behavior of the software
‘to fit his personal stylistic mode of thinking and working by scripting a dia-
logue’ (Burry, 2011).

In the context of algorithmic design, being reflective relates to the designer’s

ability to understand and control the computational and scripting tools.
‘Today, more than ever emphasis should be given to the intellectual founda-
tion of this knowledge. Being a scripting digerati is simply not enough’ (Ox-
man, N., in Scripting Cultures, Burry, 2011). Scripting languages have
become part of all leading computational design software. For example, the
Processing scripting language (Reas & Fry, 2014) enables visual designers
and artists to adapt, customize and reconfigure the structure of the software
to fit their personal stylistic mode of thinking. The scripting mode of creative
thinking thus supports the exploration of solution space as part of a unique
refinement search process. ‘The designer is no longer making choices about
single objects, but creating a matrix encompassing an entire population of
possible designs’ described as ‘explorative reflection’ (Reas & Fry, 2014;

Theories and models of parametric design thinking 7

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 Reas & McWilliams, 2010). Algorithmic code is thus related to the structure of
the visual representation.

1.6 Parametric design thinking ‘Generic Schema and

Associative Relationships’ (Woodbury, 2010)
Parametric design has currently emerged as an advanced form of design
computational support environment. It contains a synthesis of many of the
important design principles, concepts and methods that have appeared above.
Parametric design as an act of design thinking is based on the exploration and
re-editing process of associative relationships in a geometrical solution space.

The designer ‘designs’ the code of the parametric schema in order to design the
design object. The parametric schema is a unique type of mathematical model
that supports algorithmic processes of shape generation. Thus while there is
continuity with the cognitive characteristics of exploration, generation, reflec-
tion and modification in traditional paper-based design, the logic and sequen-
tial components of the design process have been transformed. Furthermore,
the built-in parametric variability of a knowledge-based cognitive schema in
a specific domain provides a powerful new medium for design generation
and innovation in all disciplines of design.

Parametric design has been defined as an exploration process of associative re-

lationships of geometric concepts (Woodbury, 2010). Parametric tools enable
reflection of both the associative logic and the geometry of the form generated
by the parametric software. The design interface provides a visual screen to
support visualization of the algorithmic structure of the parametric schema
to support parametric modification. A re-editing process of the parametric
schema becomes an act of skill in design thinking. After modification, the algo-
rithmic software can complete a re-editing process of a parametric structure
within the logic of the design schema.

2 Concepts and principles of the parametric schema in

PDT
The motivation of this review of existing theory has been to explore the con-
cepts and principles of parametric design and advance a new theoretical basis
for understanding the current evolution of parametric design thinking.

The research study has focused around the intersection of three areas of
knowledge: parametric design models and tools, cognitive models of architec-
tural design knowledge, and process models of digital design. The work iden-
tified the following unique types of generic schema in each of these area: the
algorithmic schema in parametric design; the cognitive schema of typological
and topological knowledge in Architecture, and the computational schema of
digital processes and information flow of evolutionary, performance-based

8 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 and generative process models of design. The work explores the impact of
these unique types of schema upon styles of design thinking from conception
to production.

Parametric design thinking and the definition of its concepts and principles in
design can be defined by the intersection of the three perspectives illustrated in
Figure 1. The parametric schema is presented below, including concepts of
algorithmic and scripting design models of PDT discussed in Section 2.1
and 2.2. Concepts of a knowledge-based cognitive schema in PDT are pre-
sented and discussed in Section 3 and the computational schema of digital pro-
cesses and information flow in PDT is introduced and discussed in Section 4.

2.1 Concepts of algorithmic and scripting in PDT

Parametric design can be defined as a formation process of parametric struc-
tures of associative geometry that generates the geometry of desired objects of
design. Terms and concepts such as associative relations and topological geom-
etry are implemented in a parametric algorithmic schema.

The act of scripting is an essential component in formulating a general theory

of PDT. In parametric schema the designer interacts with the parametric
modeling system (such as Rhino/Grasshopper) employing visual coding sym-
bols. The image of the design is then generated by the Rhino 3D modeling
component of the system. The parametric capability of the Grasshopper system
enables generation and modification of the design simply by changing param-
eters rather than the need to re-write substantial amounts of code. This is

Figure 1 Diagram of inter-

secting fields of research in
PDT

Theories and models of parametric design thinking 9

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 characteristic of a typical process of visual reasoning in which a parametric
modification of the script maintains the main parametric relationships that
have been initially defined. To summarize, the key-concepts and principles
of parametric design in PDT are defined below:

 Explorative mechanism of cognitive processes of refinement and adaptation

represented by an algorithmic parametric schema:

The formulation of algorithmic parametric schema is a fundamental cognitive

capability of creativity of the computational designer. In the process of
designing, the parametric schema can be modified and adapted. The cognitive
role and the logic of a generic schema is proposed as a re-representation cogni-
tive process of adaptation and refinement.

 A versioning medium for exploring parametric variations of associative rela-

tionships and topological structures:

The exploration power to produce variants within a parametric schema is a

dominant characteristic of the parametric design process. The parametric
schema supports form-generation strategies for exploring variations such as
shape relations, structural organisation, scale and formal patterns.

 Automatic algorithmic re-editing mechanism supporting models of

re-representation:

Current parametric design systems are replacing traditional CAAD re-

modeling with re-editing computational mechanisms to complete recognized
thinking processes, providing a function similar to ‘reflection in action’. Para-
metric tools, such as Grasshopper, also provide an exploration mechanism in
which the history of the internal logic of all modification moves can be traced
and reflected upon.

2.2 Concepts of visual representation in PDT

Traditional representations in design are focused upon the visual representa-
tion of the design object. Traditional visual representations are non-explicit
with respect to presenting the structural logic that underlies processes of
form generation and the logic of making the object under design. This logic
becomes explicit in parametric design systems. In addition to this expanded vi-
sual content in parametric systems, additional specific design processes now
exist including the integration of performance analysis and performative simu-
lation with generative processes. Parametric systems can provide various types

10 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 of informing processes including a visualization of both code and form of
generative and performative processes of design.

The constructive logic becomes explicit by visual display of the separating

parts of the schema (Davis, Burry, & Burry, 2011). In parametric design there
are two types of visual display. A display of the 3D visual image is generated
simultaneously to interactive programing by the visual scripting code. Para-
metric variations of the image can be updated and generated simultaneously
in a visual display in parallel to the code modification.

The parametric schema supports topological variability that can be applied to

both process and form. The parametric design process thus contains two parts:
the design of the parametric schema for the design of desired solution and the
interactive process of modification of the parameters of the schema in order to
search through the family of sub-types designs. The solution space in modifi-
cation and refinement is updated, simultaneously, on the scripting code and
the visual display. As a result, the technology and the methods of parametric
design have become a keystone of digital design thinking. Current software
packages and parametric plug-ins are designed to provide such interactive
interfaces.

Most cognitive theories of design refer to the cyclical and iterative model as the
dominant model of design. In almost all known models such as analysis-syn-
thesis, generation-representation-evaluation, and reflection in action the reflec-
tion phase is followed by re-representation of the visual representation of
geometrical properties of the object itself. In order to operate in parametric
design environments designers must adapt a different type of thinking. Design
modification in traditional ways of thinking, is usually achieved by visual op-
erations of re-drawing and/re-modeling of the object of design. While tradi-
tional models of design thinking refer to the element itself, in parametric
design re-editing refers to design process represented by a set of algorithmic
rules defined by the designer.

In parametric design the design object can be modified by an associative set of

algorithmic rules. The designer can modify parametric values or change the re-
lationships of the rules and let the system drive the whole process of visual re-
representation modification of both algorithmic code and the 3D geometric
model. Re-editing as a re-representation process is fundamentally different
from the modification of traditional paper-based and 3D modeling of geomet-
ric representations. Furthermore, re-editing one parameter will generate a
computationally modified structure of the associative rule set. This skill re-
quires a different logic of thinking relative to algorithmic design thinking
(Woodbury, 2010).

Theories and models of parametric design thinking 11

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 3 The cognitive model of design knowledge in the
parametric schema in PDT
In traditional paper-based design in processes of schema-refinement and adap-
tation there is a non-explicit modification process of typological knowledge.
This has been exemplified through a case study developed in prior research
(Oxman, 1990, 1997; Oxman & Oxman, 1992). In traditional approaches of
architectural and engineering design the design process is based on the modi-
fication and adaptation of typological knowledge (for example of dwelling
types in architecture and type of bridges, or type of columns and beams in
engineering).

In contrast, the definition of a parametric schema using typological rule struc-

tures is an explicit act in parametric design. It involves the knowledge of stra-
tegies and parametric methods of rule-based design. The knowledge of
scripting is fundamental in re-editing processes of rule-modification in
response to changing programmatic factors according to a modified program-
matic context.

The cognitive logic of a generic parametric schema as a basis for exploration

processes of solution space is by a process of parametric re-editing. Re-editing
of a parametric schema (adaptation of parameters within the schema) can be
considered an explicit process of re-representation. The unique cognitive logic
of both, the typological parametric schema and the topological schema are a
basis for exploration, creativity, and innovative processes.

3.1Cognitive design schema of typological knowledge in

PDT
A Parametric Design Schema of disciplinary typological knowledge is a
computational form of the representation of generic domain knowledge in
design. The cognitive role and the logic of a generic schema (referring in this
case to the architectural knowledge of a classical temple typology), can be
demonstrated through the example of the design of a Classical Greek Temple,
reflecting the different plan styles, geometrical series and proportions (Oxman
& Gu, 2015) (see Figure 2).

The roots of a typological parametric schema of the Classical Greek Temple can
be considered as a rule-based representation of classical notation rules of
Greek Architecture that were historically passed on without formal representa-
tion methods of design knowledge (Carpo, 2016). The Classical notation rules
provide an example of the historical transmittance of a typological schema.
Such verbal dissemination of typological rules of proportions and combinations
of elements including procedural and geometric rules that can be understood as
an early form of using a non-formal cognitive typological schema of design
thinking. According to Carpo (2016) Deleuze observed and recognized the

12 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
Figure 2 Classical variations
of the generic order of the
Greek Temple

generic power (of parametric notations) to describe and generate the whole set
of typological variations of a generic object, later termed by him as the ‘objec-
tile’. Today, this process of formulation of rule-based knowledge is similar to
the definition of the parametric typological schema.

Two examples (see Figures 3aec and 4) are used to illustrate the generic typo-
logical schema of Classical Architecture.

3.2Cognitive design schema of topological knowledge in

PDT
The cognitive logic of exploring variations in creative and innovative processes
differs in typological and topological design. These two types of logic represent
the distinction between a generic typological schema, referring to the explor-
ative design of sub-types by changing of parametric variables, versus a
topological-schema, referring to a topological explorative type of design of ver-
sioning types by changing the values of their associative relationships (Oxman
& Gu, 2015).

The exploration process of topological parametric design relies on the genera-

tive power of the system to create an infinite range of topological versioning
types that can be mediated by parametric modifications of associative relation-
ships. In nature, topological form is maintained and transformed by environ-
mental forces. Processes of growth and evolution of natural systems can be
defined and traced by topological relationships. In natural environments, com-
plex biological systems can grow and evolve by exchanging information with
their environment. Natural principles of topology were employed by D’Arcy
Thompson in his noted study of evolutionary processes in developmental
biology (Thompson, 1917). His comparative analysis represents the

Theories and models of parametric design thinking 13

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
Figure 3 Exploration processes of typological rule-based knowledge: classical parametric variations of Greek columns (3a; 3b; 3c) 3D Rhino
model: left screen; Grasshopper visual code: right screen. Credit: Liz Leibovitch, Michael Weizmann, Anais Siman, Digital Experimental
Studio, Technion

topological relations between forms of sub-species that could be deformed into

one another by topological transformations.

Typological thinking is argued by Jan Knippers as limiting the creative

approach to exploration processes. Knippers (2013) argues that the creative
style of thinking is replaced today by experimental design tested by material-
ization logic. This new experimental logic is supported by topological para-
metric design.

14 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
Figure 4 Topological curve
design. Credit: Yarden Ha-
dad, Inbal Tamir, Anat
Saar, Digital Experimental
Studio, Technion

In contrast to a varied typological schema of the classical column, a topolog-

ical parametric schema is founded on versioning moves (Figures 4 and 5) that
can be represented by rules of associative relationships and dependency. Un-
derstanding how to manipulate and explore associative relationships and depen-
dencies in topological geometry are among the central concepts and principles
of parametric design thinking.

The creative power of topological design is illustrated by the following two ex-
amples. The two design examples of a topological curve (Figure 4); and a
three-dimensional wall illustrates the process of parametric design using topo-
logical versioning strategies to design and evaluate a three-dimensional, tri-
partite parametric wall of open mesh structure (Figure 5).

The exploration process of versioning strategies in different variations of the

parametric wall design and the effect of the modulation of the central open
wall zone upon the base and header zones has been developed and explored
by re-editing the parametric schema; Figure 5a-top, illustrates a tri-partite
mesh of a differentiated pattern; Figure 5b-middle illustrates a uniform
mesh zone; Figure 5c-bottom illustrates a hybrid zone of a differentiated
mesh pattern.

4 Information flow processes in PDT

In order to represent processes of parametric design there is a need to explicate
the flow of information and the embedded logic in process models of

Theories and models of parametric design thinking 15

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
Figure 5 Exploring versioning strategies in a topological parametric wall design: (a) A tri-partite differentiation. (b) A uniform differentiation
(c) A hybrid differentiation zone 3D Rhino model: left screen; Grasshopper visual code: right screen. Credit: Liz Leibovitch, Michael Weiz-
mann, Anais Siman, Digital Experimental Studio, Technion

parametric design. This includes the explication of knowledge structures, the

order of flow of information, and the parametric logic of design.

The process models and information flow diagrams (Figures 6 and 7) provide a
generic formulation to represent the development of computational design
processes over time (Oxman, 2006). The formulation is presented in the

16 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
Figure 6 Generic schema of
process models and informa-
tion flow (2006) represented
by links between Representa-
tion (R) Evaluation (E) Per-
formance (P) Generation
(G) R. Oxman, Design
Studies (2006)

context of the current research in order to map diverse process models of para-
metric design.

The centrality of the designer in this schematic formulation (marked as D) im-

plies that the traditional role of the designer as a visual thinker is maintained in
parametric design (with respect to both code and form). Figure 6 (Oxman,
2006) includes the processes of a visual re-representation (marked as R), eval-
uation (marked as E), performance (marked as P), and generation (marked as
G) in which both the non-implicit and/or the explicit-flow of digital informa-
tion employ a central visual representation. Directions of the flow of informa-
tion are represented by arrows. Additional symbolic designations provide for
the notation of specific additional components of PDT processes.

The symbols in Figure 7 illustrate modes of interaction, links with visual rep-
resentations, and types of interaction with digital media (in parametric design
the diagram represented in Figure 7 employs the term digital media, refers to
algorithmic languages, and scripting code). Their specific use in parametric
design is explicated in the following diagrams (Figure 8) of various models.

4.1 Process models of information flow supported by

parametric tools in PDT
Understanding and developing models of information flow in various process-
based models of digital design (see Figure 8) are providing a foundation for

Theories and models of parametric design thinking 17

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
Figure 7 Graphical symbols of the generic schema. Credit: R. Oxman, International Journal of Design Studies, 27 (3) pp. 229e265

Figure 8 Digital process models of information flow in digital design. Credit: R. Oxman, International Journal of Design Studies, 27 (3) pp.
229e265

understanding the current development of ‘plug-in’ parametric design tools

which support digital process models of parametric of design thinking.

Design thinking in parametric design relies on tools that provide visualization

of code and form of coded structure of the parametric schema and the 3D geo-
metric model of the design object.

Models of performance-based form generation are currently supported by spe-

cific visual ‘plug-in’ tools in parametric design environments. Plug-ins such as
Kangaroo (Form-finding with Computational Physics); Karamba (Structural
Design); ARCH-CFD (Computational Fluid Dynamics); Pachyderm (Acous-
tical Simulation); Ladybug and Honeybee (Sustainable Environmental
Design), etc. exist in order to provide analytical capability to support specific
process models of PDT. In all plug-in tools of parametric design such as Kan-
garoo, Karamba, Ladybug, and Honeybee, the interactive display of the visual
image is generated in parallel to an interactive programing window for modi-
fying the visual scripting code. Furthermore, parametric variations of the im-
age can be updated and generated simultaneously in the visual display in
parallel to code modification in a re-editing process. Current software

18 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 packages and parametric plug-ins are designed to assist processes of re-editing
relationships replacing the traditional cognitive process of re-representation.

5 Material fabrication design (MFD) as a schema of

disciplinary architectural knowledge in PDT
Today the convergence of design, engineering, and architectural technologies
is breeding a new material practice in experimental design domains such as ar-
chitecture. This shift is defined as a dynamic synthesis of principles of spatial,
structural and material ordering that are integrated through materialization
and fabrication technologies. Current discourse on materials includes concepts
of materialization processes that are changing operative concepts of design
tectonics in design thinking.

5.1 The integration of material fabrication in parametric

design
Tectonics is becoming a seminal concept in parametric design due to current
digital technology and computational informing processes that enable the
mediation between design and fabrication. Tectonic relationships are being
informed by holistic types of the logical flow of information from conception
to production. This multi-stage process has been termed, Informed Tectonics
(Oxman, 2012). The evolving processes of information have been defined by
researchers in this field as a computational informing process that enhances
the tectonic relationships between form, structure and material properties
within the logic of fabrication and robotic technologies (Oxman, 2011).

One of the emerging aspects of materialization and fabrication processes has

been the ability to alter the conventional order of tectonic processes in archi-
tecture from form-structure-material, to material-structure-form. This
reversal of order has changed the traditional conceptual model of design
thinking. In the traditional model of architecture, materialization follows con-
ceptual design of form and the design of structure. Today, due to the growing
importance of digital fabrication, the selection of materialization techniques
can be an essential part of the inception stage in design. This development dic-
tates an order of theoretical priority for material design in the conceptual
thinking phase of design. Thus the new procedural order: from material to
structure to form has relocated material, materialization and material fabrica-
tion to a level of priority in the PDT processes of architecture and other design
disciplines.

5.2 Material fabrication design (MFD)

The current priority of materialization processes and fabrication technologies
in architecture are breeding a new field of design research termed: Material
Fabrication Design (MFD) (Oxman, 2012, 2016). By developing concepts,
principles, and models of MFD designers are given the freedom to create

Theories and models of parametric design thinking 19

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 new material-related experimental methods of design thinking that were previ-
ously considered difficult, or impossible, to implement (Figure 11).

As presented above, and in the following examples of leading architectural

projects, MFD is defined today as a computational informing process that en-
hances tectonic relationships between form, structure and material within the
logic of materialization and fabrication technologies. In computational in-
forming process, information is translated directly into the control data that
drives the materialization processes. The three key concepts of MFD are
briefly described.

Material Design is defined as understanding and exploiting the properties and

structural behavior of materials in the processes of design. Relationships with
form, structure and construction are developed emphasizing the synthesis of
complex orders in the design of elements such as building structures, building
facades and envelopes.

Materiality is defined as digital tectonic content mediated by fabrication and

robotic techniques. Categories of fabrication and materialization represent
technological extensions based on both conventional and new materials and
their relationships with form, structure and construction. Digital Materiality
is considered as digital processes of materialization that contribute to new
types of digital tectonics (Gramazio, Kohler & Oesterle, 2010; Oxman, 2010,
2012).

Fabrication technologies originally developed as media for Rapid Prototyping

(RP). Today, the relationships between novel design technologies and tech-
niques of fabrication are rapidly creating new ways of design thinking in

Figure 9 Parametric technique of assembly parts (Laser-cutting by Rhino Grasshoper). Credit: Liz Leibovitch, Michael Weizmann, Anais Si-
man, Digital Experimental Studio, Technion

20 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
Figure 10 Architectural para-
metric structure of MFD: Us-
ing scale of fabrication
techniques as principles of a
structural scale design.
Credit: Metropol Parasol,
Seville, Spain Designed by
J€
urgen Mayer (2011). View
of the parametric space grid
structure. Source: J€urgen
Mayer; Photos: Nikkol Rot
Holcim

Figure 11 Formation Model (R). Experimental design of the parametric differentiation pattern for a high-rise building. Credit: Hadid-
Schumacher Studio, the Angewandte Department of Architecture, Vienna (2007)

design domains such as Architecture, Industrial Design, Engineering and

other design disciplines. Designers are expanding the scope of design thinking
by making design decisions related to the choice of technology and production
method in response to new design complexities and innovative design. Fabri-
cation processes are currently enabled within parametric software tools. The
data for making assemblies of digital models for digital manufacturing are ac-
quired directly from the parametric system (Figure 9).

In MFD processes, the formal and behavioral content of materials has become
a dominant component of design conceptualization processes. The reversal of
conventional information flow of tectonic order in design has given priority to
materialization techniques of fabrication in thinking models of parametric
design.

Theories and models of parametric design thinking 21

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 Principles of parametric fabrication techniques can affect the formal and
behavioral content of material structures that become a dominant part of
both the conceptualization stage and the materialization stage as in the Met-
ropol Parasol in Seville (Figure 10) designed by J€
urgen Mayer in 2011.

6 Thinking differentiation in architectural design

Relative to specific design objectives, parametric systems provide diversified
explorative, context-based strategies of formal, generative and performative
design in a parametric solution space. In contrast to traditional architectural
composition processes, parametric design in architecture can be controlled
by patterns of differentiation. Exploration, modification and refinement
of code-based geometrical structures represented by a differentiation
pattern, support various types of patterned processes of the object under
design.

Differentiation in architecture may be considered as the local specialization of a

repetitive formation. The objective of the specialization of a part of a regular
system is to provide new types of formal, functional, performative, and struc-
tural properties and material behavior. From the point of view of novel modes
of design thinking, the potential of using geometric patterns in PDT has had
an impact on both style and model of design thinking. The functional and
aesthetic power of differentiation is beginning to make a significant parametric
design thinking strategy of current design in architecture.

In recent years, in a series of articles in professional journals, the term Para-

metricism has been proposed by Patrick Schumacher to indicate an architec-
tural approach ‘seen as a global type which draws its impetus from
technological advancement and computational evolution’ (Schumacher, 2008).
Helen Castle (Castle, 2016) has referred to these design characteristics as, ‘a
style that succeeds the modern movement’.

Schumacher has stated that: ‘we pursue the parametric design paradigm all
the way, penetrating into the corners of the discipline. Systematic adaptive
variations, continuous differentiation (rather than mere variety), and dy-
namic parametric figuration concerns all design tasks from urbanism to
the level of tectonic detail, interior furnishing and the world of products.’
(Schumacher, 2008). In a recent volume of the British journal, Architectural
Design (AD), entitled Parametricism 2.0, published and guest edited by Pat-
rick Schumacher (Schumacher, 2016), a more realistic and less dogmatic
view of parametric differentiation in architecture, urbanism and industrial
design is proposed by Schumacher and other practice-based design re-
searchers. Here, the potential of Parametricism is considered to help solve
environmental problems and social issues through formal and functional

22 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 solutions derived from exploiting pattern mechanisms of parametric differ-
entiation. As such, Parametricism can be described both as a parametric
design thinking strategy as well as a process-driven model of parametric
design thinking (PDT).

In the following sections, the potential of parametric differentiation is pro-

posed and exemplified as a computational model of parametric design
thinking.

6.1 Process models of differentiation in parametric design

Differentiation is a term in the field of calculus, and continuous differentia-
tion can be considered as the ‘geometrical transcoding of parametric varia-
tions into differential geometries driven by iterative versioning processes of
parametric design in a continuous rhythm’ (Jabi, 2013). The exploitation
of differentiation as a design strategy in parametric systems is unique as a
design medium that can support a versioning series within otherwise contin-
uous fields. This strategy can be applied in both parametric design and fabri-
cation. Within processes such as form-finding, performance-based and
generative design, differentiation patterns provide a medium for versions
of exploration moves that are interactively initiated by the designer. That
is, within each parametric version of the schema the designer can interac-
tively modify the rhythm of geometric structures as seen in the example in
Figure 5aec above.

In each one of the following diagrams the role of a dual visual interface sup-
porting code input and resultant 3D visual representation, is presented and
illustrated. Each of the illustrative diagrams presented below represents the
three components of information flow process in parametric design: the first,
presents (by the same visual image) a symbolic differentiation code (left
side); the second presents a unique process model of information flow (mid-
dle) and the third represents a selected image of a 3D model in the field of
architecture (right side). In each of the three characteristic parametric
design processes, the information flow diagram (center) differs (Figures
11e13).

The diagram in Figure 12 presents an evaluation process model of the design

of a cylindrical structure employing FEA Finite Element Analysis software
(load forces are displayed in color). Optimizing the relationship between
form and structure is achieved by exploring the particular form of structural
differentiation within the geometrical pattern of the design.

In parametric generative processes the 3D model is generated in reference to

the input of a visual scripting code. Parametric differentiations of the tree

Theories and models of parametric design thinking 23

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
Figure 12 Evaluation Model (E). Representation and parametric differentiation of structural properties in design. Credit: N. Oxman

Figure 13 Generation Model (G). Parametric recursive differentiated branching as a design medium of generative design. Credit: Mor Herman,
Digital Experimental Studio, Technion

structure presented below can be modified and generated simultaneously to

code modification of selected differentiation patterns (Figure 13). The diagram
illustrates the visual interface for interactive code input and the resultant 3D
representation of the geometrical model.

The design of a dynamic model of performance-based design requires repre-

sentation of temporal dynamic solar properties. By using a pattern of para-
metric differentiation the dynamic behavior of the louver system can be
represented (Figure 14). In all dynamic and responsive systems, the potential
for differentiation provides a design medium for the simulation and explora-
tion of performance/time responsive systems.

24 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
Figure 14 Performance Model (P) in the design of a dynamic system of louvers. Credit: Shoham Ben Ari, Digital Experimental Studio,
Technion

7 Thinking difference: classes of architectural types and

forms of differentiation patterns in PDT
Differentiation gradient thinking within the topology of a parametric pattern
has become a significant medium of design in current architectural design. Dif-
ferentiation in parametric design can be treated as a feature of a versioning
process that allows for difference to occur within a continuous field, or rhythm
(Jabi, 2010). In contrast to parametric instances in typological parametric
schema, differentiation can be defined as a type of topological parametric
versioning schema that differentiates a formal topological pattern of the
design in response to functional and contextual environmental goals and
constraints.

As has been defined above, differentiation patterning in architectural design is

a unique strategy of PDT. In order to understand the application and purposes
of parametric differentiation patterns we have defined types of differentiation as
a medium of designing in architecture.

The following brief study presents an approach to the definition and classifica-
tion of specific differentiation approaches in specific contexts, goals and con-
straints in architectural design. The study is presented below through the
illustration and classification of a series of types of differentiation in architec-
ture. The following case studies represent a topological basis for the selection
of the specific differentiation pattern. Topological relationship of form, struc-
ture, tectonic model, and types of materialization (MFD) enabled both design
and materialization of complex forms by designing a specific differentiation
pattern as a medium of architectural design. In addition to functional objec-
tives, each of the architectural projects presents forms of stylistic distinctive-
ness and performative behavior. In each one of the represented case studies

Theories and models of parametric design thinking 25

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 unique architectural differentiation patterns are presented by the intersection
of the following categories:

- Process Model of Digital Design

- Architectural Model of Tectonic Order (MFD)
- Differentiation Pattern
- Architectural Design Medium

The next sections present the architectural design strategy of differentiation

design in selected architectural case studies, each of which may be considered
as a novel strategy of design thinking in PDT.

7.1 Differentiation of a gradient geometric pattern e as a

modifier medium of diverse functionalities
Case Study: The Broad Museum; Los Angeles, 2015; Diller Scofidio and
Renfro

- Process Model of Digital Design: Geometrical Formation Model

- Architectural Model of Tectonic Order: Form-Structure-Material
- Differentiation Pattern: Gradient mesh of rhomboidal surface
- Architectural Design Medium: Modifier medium of diverse functionalities

In the recently constructed Broad Museum in California (Figure 15) the inte-
rior gallery space of the museum is provided with a continuous solar control
and modulation system. In the case of this building, this is a fixed system.
Termed the ‘Veil’ by the architects, this continuous parametric rhomboidal
surface pattern covers the exterior wall and the roof of the museum. The
mesh continuity is differentiated in geometry for lighting control of the

Figure 15 Differentiation of
geometric pattern as a modi-
fier medium of diverse archi-
tectural functionalities: The
Broad Museum Los Angeles
2015, Visualization of Main
Entry Facade. Credit: Broad
Art Museum; Diller Scofidio
and Renfro

26 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 skylights of the roof. Furthermore, the function of the wall surfaces is locally
modified for architectural purposes such as opening up of the entrance area
and in order to provide exterior exposure of the glass wall and public areas
of the building. A large sculptural indentation on the second floor of the fa-
çade further modulates light at the pedestrian arrival point to the exhibition
levels of the building.

7.2 Differentiation of a structural mesh as a medium - of an

ecological responsive skin
Case Study: Louis Vuitton Store, Macau, 2007; Zaha Hadid Architects

- Process Model of Digital Design: Performance-based Ecological model

- Architectural Model of Tectonic Order: Structure-Material-Form
- Differentiation Pattern: Gradient packing of circular elements
- Architectural Design Medium: Structural mesh medium of responsive skin

The experimental design for the Louis Vuitton Store in Macau (Figure 16)
from 2007 presents certain additional characteristics of differentiation in
PDT. As with the case of the Broad Museum, the Zaha Hadid design for
the Louis Vuitton store is a protective mesh structure for the external skin
of the building. The typological structure in this case is a mesh structure of cir-
cular packing of a gradient system of elements.

The functional behavior in this experimental project was intended to provide

an ecologically responsive differential pattern. Within the structural thick wall
of the evolutionary elements it is possible to modulate and control direct solar
penetration, as well as the to control viewpoint and to provide for the

Figure 16 Differentiation of a
structural mesh as a medium
of an ecologically responsive
facade Louis Vuitton Store,
Macau, Zaha Hadid Archi-
tects 2007. Credit: Zaha Ha-
did Architects

Theories and models of parametric design thinking 27

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 introduction of other elements such as signage, external lighting, etc. In this
experimental design, the wall structure has become the structural support of
the roof element. Thus, the skin is both performative as well as an ecological
responsive design system in the PDT of this project.

7.3 Differentiation of expanded aluminum metal screen e as

a medium of modulation
Case Study: Messe Basel New Hall, 2013; Herzog de Meuron

- Process Model of Digital Design: Material Formation Model (MFD)

- Architectural Model of Tectonic Order: Material-Structure-Form
- Differentiation Pattern: Expanded metal aluminum mesh
- Architectural Design Medium: Modulation medium for program and
function

In the design and construction of the New Hall of the Bern Messe of 2013, by
the Swiss architects, Herzog de Meuron (Figure 17), an expanded metal
aluminum mesh screen provides a similar function for the two upper levels
of exhibition halls within this building. The building, the function of which
is to provide three new independent halls for the Bern Messe, is treated as three
superimposed buildings. The metal mesh screen is differentiated in form by
bending and folding the structure of the screen in a process similar to the
‘expanded metal’ well known in building. This much larger mesh is a skin
that can be treated as a gradient surface and can be differentially modulated,
thus providing a distinguishing architectural form and view possibilities for
each of the three halls.

Figure 17 Differentiation of a
metal mesh screen as a me-
dium of distinguishing pro-
grammatic functions Messe
Basel New Hall, Herzog de
Meuron 2013. Credit: Herzog
de Meuron

28 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 7.4 Differentiation as medium of experimental structural
design
Case Study: Venice Architecture e Biennale, 2016; Philippe Block (BRG at
ETH, Zurich)

- Process Model of Digital Design: Performance based structural design

model
- Architectural Model of Tectonic Order: Material-Structure-Form
- Differentiation Pattern: Differentiation of form in bending
- Architectural Design Medium: Medium of structural form finding
optimization

The BRG Block Research Group led by Philippe Block at the ETH Zurich has
become noted for its experimental structural design using parametric design as
a basis for a new approach to design that transcends the typologic and calcu-
lation approaches of structural building types. His approach exploits para-
metric structural form finding through digital parametric processes and
performance models of structural design. Parametric models constructed by
algorithmic scripts can generate designs by changing geometrically associated
parameters according to given structural forces and properties of materials of
design in different contextual conditions. This is accomplished both through
digital simulation and the physical experimentation of experimental models.
This follows upon much pioneering work of proto-parametric structural de-
signers such as Gaudi, Frei Otto, and, in the case of the study below, the shell
structures of Felix Candela and Heinz Isler (Block, 2016).

This exhibition at the Venice Biennale of 2016 (Figure 18) illustrates an

example of contemporary experimental vault structure replacing the tradi-
tional masonry materials and the traditional methods of ribbed vaults with
new material fabrication methods and techniques. The project demonstrates
how new structural generative processes of form finding and optimization
methods can generate geometry of compressive vaults that are absorbed by
tension ties. Following historical precedents, these vaults demonstrate a new
approach to PDT in vault design and construction that saves constructional
weight and environmental impact and can be created by 3D print technology.

7.5 Differentiation as medium of multi-functional

performative design
Case Study: Chaise Lounge, 2008e2010; Neri Oxman (Mediated Matter
Group at MIT) in Collaboration with Craig Carter 2009

- Process Model of Digital Design: Performance based generation

- Architectural Model of Tectonic Order: Digital Material-Structure-Form

Theories and models of parametric design thinking 29

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
Figure 18 Differentiation as
medium of experimental
structural optimization
Beyond Bending e Venice
architecture Biennale 2016;
Philippe Block, BRG ETH
Zurich. Credit: BRG at
ETH/Iwan Baan

- Differentiation Pattern: Voronoi Differentiation

- Architectural Design Medium: Medium of ‘multi-functional performative
design’ of Variable Properties Design (VPD) of structural surface pattern

Neri Oxman at MIT and her research group (Mediated Matter/Media Lab)
are inspired by nature where form-generation is driven by growth processes,
topological versioning and variable properties of material in the behavior of
natural systems. The term Material Ecology coined by Neri Oxman seeks
the integration of form, material and structure by incorporating algorithmic
form-finding strategies (Oxman N., 2011), and the invention of fabrication
techniques of 3D fabrication technologies.

Figure 19 Medium of ‘Multi-

Functional Performative
Design’ of Variable Proper-
ties Design (VPD) of struc-
tural surface pattern and
digital materiality Chaise
Lounge, Beast, 2010; Neri
Oxman (Mediated Matter
Group, Media-Lab, MIT) in
collaboration with W. Craig
Carter (Material Sciences
MIT). Credit: Production:
Stratasys 3D printed with
Stratasys multi-material 3D
printing technology; Photo
credit: Neri Oxman

30 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 The design of the chaise, Beast (Figure 19), is an example of replacement of
traditional information flow from Form-Structure-Material to Material-Struc-
ture-Form that incorporates a novel fabrication technology termed, ‘Variable
Properties Design (VPD). VPD is considered to be a design model, a method-
ology and technical framework by which to model, simulate, and fabricate ma-
terial assembly with varying parametric properties designed to correspond to
multiple and continuously varied functional constraints’ (Oxman, 2008). This
approach can be regarded as a unique computational model of PDT. This
unique thinking process in which the invention of a novel tool, technology
or technique reflects and guides the creation of the object itself can be regarded
as a novel model of PDT (see Figure 19).

In the design for the chaise lounge (Beast, 2010) differentiation is achieved by
the differentiation of Voronoi tiling of a parametric surface that provides a
multi-functional performative design approach to the structural surface pattern
of the chaise; this is accomplished through Variable Properties Design (VPD).
That is, the 3-d printing process is guided by performance-based differentiated
properties of a digital material. This is illustrated by the design of the contin-
uous skeletal area of the shell structure (black), the support structure (grey)
and comfort and softness supports (white) at the areas of bodily pressure.

7.6 Differentiation as a design medium of structural

sectioning - materialized by fabrication techniques
Case Study: Metrapol Parasol Seville, Spain (2011); J€
urgen
MayereHermann and Arup Engineering

- Process Model of Digital Design: Performance-based Model

- Architectural Model of Tectonic Order: Form-Fabrication-Structural
Material
- Differentiation Pattern: Force Form Differentiation
- Architectural Design Medium: Medium of structural sectioning, material-
ized by a form-force pattern and a parametric fabrication technique

Figure 20 Urban plaza, Met-

ropol Parasol, Seville, Spain.
J€
urgen MayereHermann and
Arup Engineering (2011).
Credit: J€
urgen Mayere
Hermann and Nikkol Rot for
Holcim

Theories and models of parametric design thinking 31

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 Metrapol Parasol designed by J€urgen MayereHermann and Arup Engineering
(Figure 20) is the name of a partially covered, multi-functional urban plaza in
Seville, Spain. The design of J€urgen Mayer and Arup Associates essentially
provides three vertical zones of function: an elevated plaza; a ground level
of archeological findings; and the visually dominant major element of an
elevated wooden grid structure e the parasol e that is both an urban-scale
shading element as well as an elaborated viewing platform raised on six large
mushroom columns, or vertical structures.

The cloud-like form of the parasol structural network has been influenced by
the structural forces, programmatic requirements, and environmental condi-
tions. The material of the structural system was formulated after the selection
of the fabrication technique. Finally, the form and the dimensions of the struc-
tural sectioning were informed by structural force properties represented by a
differentiation structural model of a force-field pattern.

The urban-scale parasol is designed as a wooden grid-lattice structure imple-

mented by parametric fabrication technique to create an organically flowing
form in response to structural loading. The diversity of the structural sectioning
was designed by a Force-Form differentiation pattern. The negotiated process
between form and structural forces is reminiscent of Gaudi’s use of catenary
experiments in the structural design of the Sagrada Famila church in
Barcelona.

To summarize, differentiation patterns of PDT as a design strategy in architec-

tural and urban scale are, as yet, little studied or developed in architectural
thinking. Such PDT applications in both scales are a future field for research
in design.

8 Summary and conclusions: contributions to research

and praxis of PDT
Beyond being another tool for modeling complex forms, parametric design
was presented in this research paper as a unique and distinctive model of crea-
tivity and innovation in parametric design thinking (PDT).

The systematic theoretical analysis and definition of the chronological evolu-

tion of models of design thinking from the origins of cognitive models and up
to PDT demonstrated both the continuity and change in the evolution of models
of design thinking. The presentation of the transformational evolution process
of thinking models in design; from traditional paper-based media to computa-
tional media-related models of design thinking has identified seminal shared
cognitive concepts during a substantive change to media-related methods,
principles and techniques in PDT.

32 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 The research study presented above explored knowledge of three intersecting
fields of knowledge: parametric design models and tools; cognitive and compu-
tational models of architectural knowledge; and process models of digital
design. Unique types of generic schema in each of these areas: the algorithmic
schema in parametric design; the cognitive schema of typological and topolog-
ical knowledge in architecture, and the computational schema of digital pro-
cesses and information were defined and illustrated. Contemporary process
models such as formation, evolution, performance-based; and generative pro-
cess models of design have been demonstrated as holistic processes of design
thinking from conception to production.

Parametric design has been presented as a new paradigm of design thinking. The
evolution of leading concepts of design thinking from typological thinking to
topological design thinking in creative design is one of the most distinctive
changes in design thinking in the design disciplines. The centrality of topolog-
ical variants in design and topological versioning in the design medium of the
parametric schema has been demonstrated as a seminal theoretical and opera-
tive methodological concept in PDT.

Understanding and learning a new symbolic role of imagery in design is among

the foundations of PDT. Sketching by code is not only a possibility, but prom-
ises to become a new norm of skill and knowledge. The traditional role of the
visual image in paper-based design is replaced by both the integration of visual
form and visual code in parametric design. The relationships between the visual
image of design form and visual scripting code in parametric tools have been
demonstrated by, and illustrated in, diagrams. This relationship is supported
by a visual interface for cyclical exploratory re-editing and interpretation pro-
cesses in PDT models of architectural design. Thus PDT is an evolutionary pro-
cess of explorative design of the parametric schema.

Integrating PDT and MFD as a new class of disciplinary knowledge is chal-

lenging the culture of design in all fields of the design discipline. Disciplinary
knowledge of novel types of material design, and fabrication and materializa-
tion technologies (MFD) in parametric design represent a unique form of tec-
tonic thinking in current PDT. Materiality relates to tectonic design mediated
by fabrication and robotic techniques. Digital Materiality is considered as dig-
ital processes of materialization that contributes to new types of digital tec-
tonics such as with advanced 3D printing technologies. Understanding the
evolution and change of concepts of MFD contributes to novel processes in
design thinking models in parametric design.

Leading concepts of topological design thinking offer unique design methodo-

logical approaches in PDT. Mediated by recent parametric design tools, para-
metric design can be seen as a new operative model of design thinking. While

Theories and models of parametric design thinking 33

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 the traditional typological mode of thinking (of generic types of forms and
functions in disciplines such as architecture and structural design) may limit
creative processes of exploration, differentiation strategies in a topological
mode of thinking supports new types of creative thinking in innovated design.
Understanding how to manipulate and explore associative relationships in to-
pological parametric schema is emerging as both of a model and a style of
design in PDT.

Differentiation patterns are an ensemble collection of the intersection of prior

models. The ensemble of prior models now includes the process models of dig-
ital design and information flow; the disciplinary models of tectonic order, mate-
rialization in MFD models, and the selection and implementation of
differentiation patterns. For disciplines such as architecture there is now a chal-
lenge to develop and formulate new disciplinary knowledge (such as new types
of differentiation patterns) in the transitioning to a normative production of
knowledge in parametric design.

Parametricism as formulated by Schumacher as a new style, is in fact both, a

new style and model of design. It represents both a new body of knowledge and
new methodologies in the design disciplines. PDT should exist as a shared
design paradigm among all specific domains (e.g. architectural, structural en-
gineering, and materialization) producing holistic conceptual thinking pro-
cesses from conception to fabrication. This is a new horizon of theory and
pedagogy for the future of design thinking.

The tools and practices of parametric design are beginning to impact new
forms of development in the institutions of education and practice of design
culture. Scripting and tool-making are becoming required forms of knowledge
in research, education and practice. Is this to become core knowledge of design
in academia and practice?

Finally, parametric design has been presented as a new paradigm of design

thinking that integrates and advances certain of the concepts of prior models
of design thinking. Design research in this area should become more strategic,
computationally informed, and performance-based; it should be oriented to
the production of knowledge relative to specific programmatic and functional
requirements given by specific contexts. It is the integration of generic knowl-
edge such as types of generic parametric schema with new forms of disciplinary
knowledge such as MFD that is creating new forms of disciplinary knowledge
in design.

Acknowledgements
I would like to express my gratitude to the reviewers for their incisive recom-
mendations for revisions and improvements. Their enlightened critique has
had a profound effect upon the level of quality, clarity and understanding of

34 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 the article. Hopefully, this will expand the research into the relationship be-
tween design thinking and mediated design for the readership of the Interna-
tional Journal of Design Studies.

I would like to thank my students that took part in my Digital Experimental

Studio and credit their works and contribution to our research study on para-
metric design thinking.

References
Boden, M. (1990). The Creative Mind. London: Weidenfeld and Nicolson.
Burry, M. (2011). Scripting Cultures: Architectural Design And Programming. AD
Primers, Wiley and Sons.
Carpo, M. (2016). Parametric notations. In P. Schumacher (Ed.), Parametrism
2.0. Architectural design (AD), 86(2) (pp. 25e29). Wiley and Sons.
Castle, H. (2016). Editorial. In P. Schumacher (Ed.), Parametrism 2.0. Architec-
tural design (AD), 86(2) (pp. 5). Wiley and Sons.
Cross, N. (1982). Designerly ways of knowing. Design Studies, 3(4), 221e227.
Cross, N. (1997). Descriptive models of creative design: Application to an
example. Design Studies, 18(4), 427e455.
Cross, N. (2006). Designerly Ways of Knowing. London: Springer.
Cross, N. (2011). Design Thinking. Oxford-New York: Berg.
Davis, D., Burry, J., & Burry, M. (2011). Untangling parametric schemata:
Enhancing collaboration through modular programming. In Proceeding of
the 14th International Conference on Computer Aided Architectural Design.
Liege: University of Liege.
Dorst, K. (2012). Frame Innovation: Create New Thinking by Design. Cambridge,
MA: MIT Press.
Dorst, K., & Cross, N. (2001). Creativity in the design process: Co-evolution of
problem-solution. Design Studies, 22(5), 425e437.
Gramazio, F., Kohler, N., & Oesterle, S. (2010). Encoding material. In
R. E. Oxman, & R. M. Oxman (Eds.), The new Structuralism: Design, engineer-
ing and architectural technologies, architectural design (AD), 80(4) (pp.
108e115). Wiley and Sons.
Jabi, W. (2013). Parametric Design for Architecture. Laurence King Publishing.
Knippers, J. (2013). From model thinking to design process. In B. Peters, &
X. Kestelier (Eds.), Computation works, architectural design (AD), 2 (pp.
74e81). Wiley and Sons.
Lawson, B. (1980). How Designers Think. Architectural Press.
McCullough, M., Mitchell, W., & Purcell, P. (Eds.). (1990). The Electronic Design
Studio: Architecture, Media and Knowledge in the Computer Era. London En-
gland: MIT Press.
McKim, R. H. (1972). Experiences in Visual Thinking. Institute of Educational
Science.
Menges, A., & Ahlquist, S. (Eds.). (2011). Computational Design thinking. Wiley
and Sons.
Oxman, R. (1990). Prior knowledge in design, A dynamic knowledge-based model
of design and creativity. Design Studies, 11(1), 17e28.
Oxman, R. (1994). Precedents in design: A computational model for the organi-
zation of precedent knowledge. Design Studies, 15(2), 141e157.
Oxman, R. (1995). Observing the observers: Analyzing design activity. Design
Studies, 16(2), 275e284.

Theories and models of parametric design thinking 35

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001
 Oxman, R. (1997). Design by re-representation: A model of visual reasoning in
design. In O. Akin (Ed.), Special issue on prescriptive and descriptive models
of design, design studies, 18(4) (pp. 329e347).
Oxman, R. (2004). Think-maps: Teaching design thinking in design education.
Design Studies, 25(1), 63e91.
Oxman, R. (2006). Theory and design in the first digital age. Design Studies, 27(3),
229e265.
Oxman, R. (2008). Performance-based design: Current practices and research is-
sues. International Journal of Architectural Computing, (IJAC), 6(1), 1e17.
Oxman, N. (2010a). Structuring materiality: Design fabrication of heterogeneous
materials. In R. E. Oxman, & R. M. Oxman (Eds.), The new Structuralism:
Design, engineering and architectural technologies. Architectural design (AD),
80(4) (pp. 78e85). Wiley and Sons.
Oxman, R. (2010b). Morphogenesis in the theory and methodology of digital tec-
tonics. In R. Motro (Ed.), Morphogenesis in the International Journal of space
structures, IJSS, 51(3) (pp. 195e207).
Oxman, N. (2011). Variable property rapid prototyping. The International Journal
of Virtual and Physical Prototyping, 6, 3e31.
Oxman, R. (2012). Informed tectonics in material-based Design. Design Studies,
33(5), 427e455.
Oxman, N. (2015). Templating design for biology and biology for design. In
A. Menges (Ed.), Material synthesis. Architectural design (AD), 85(5) (pp.
100e107).
Oxman, R. (2016). MFD: Material-based design informed by digital fabrication.
In M. Loyola (Ed.), Architecture and technology, revista materia, 13 (pp.
105e109).
Oxman, R., & Gero, J. S. (1987). Using an Expert System for Design Diagnosis and
Design Synthesis, Expert Systems. Wiley Online Library.
Oxman, R., & Gu, N. (2015). Theories and models of parametric design thinking.
In Proceeding of the 33rd International Conference of ECAADE Conference,
Vienna, pp.2e6.
Oxman, R. E., & Oxman, R. M. (1992). Refinement and adaptation in design
cognition. Design Studies, 13(2), 117e134.
Reas, C., & Fry, B. (2014). Processing: A Programming Handbook for Visual De-
signers and Artists. MIT Press.
Reas, C., & McWilliams, C. (2010). FormþCode in Design, Art, and Architecture.
Princeton Architectural Press.
Rowe, G. P. (1987). Design Thinking. Cambridge: The MIT Press.
Sch€on, D. A. (1983). The Reflective Practitioner: How Professionals Think in Ac-
tion. New York: Basic Books.
Sch€on, D. A. (1987). Educating the Reflective Practitioner: Towards a New Design
for Teaching and Learning in the Professions. Jossey-Bass.
Sch€on, D. A. (1988). Designing: Rules, types and worlds. Design Studies, 9,
181e190.
Schumacher, P. (2008). Parametrism as a Style e Parametrism Manifesto. Lon-
don: New Architectural Group.
Schumacher, P. (2016). Parametricism. In P. Schumacher (Ed.), Parametrism 2.0.,
architectural design (AD), 86(2) (pp. 8e17). Wiley and Sons.
Simon, H. A. (1969). The Sciences of the Artificial. Cambridge, Mass: MIT Press.
Thompson, D. W. (1917). On Growth and Form. Cambridge University Press.
Visser, W. (2006). The Cognitive Artifacts of Designing. Lawrence Erlbaum
Associates.
Woodbury, R. (2010). Elements of Parametric Design. Routledge.

36 Design Studies Vol -- No. -- Month 2017

Please cite this article in press as: Oxman, R., Thinking difference: Theories and models of parametric design thinking,
Design Studies (2017), http://dx.doi.org/10.1016/j.destud.2017.06.001