This article was downloaded by: [York University Libraries]

On: 12 August 2014, At: 16:23

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Journal of Engineering Design

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/cjen20

Modelling the Design Process in Engineering and in

Architecture
a a
NIGEL CROSS & NORBERT ROOZENBURG
a
Faculty of Industrial Design Engineering, , Delft Technological University , Nigel Cross and
Norbert Roozenburg, 2628 BX Delft, The Netherlands.
Published online: 19 Apr 2007.

To cite this article: NIGEL CROSS & NORBERT ROOZENBURG (1992) Modelling the Design Process in Engineering and in
Architecture, Journal of Engineering Design, 3:4, 325-337, DOI: 10.1080/09544829208914765

To link to this article: http://dx.doi.org/10.1080/09544829208914765

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the
Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Journal of Engineering Design, Vol. 3, No.4, 1992 325

Modelling the Design Process in Engineering

and in Architecture

NIGEL CROSS & NORBERT ROOZENBURG

Downloaded by [York University Libraries] at 16:23 12 August 2014

SUMMARY Models of the design process in engineering seem to have converged upon a
consensus represented, for example, by the German VDI model. However, after starting
from common origins, models of the design process in architecture have diverged from the
engineering consensus, in response to criticisms from both theorists and practitioners.
There now appear to be significant differences between the engineering and architectural
design models. Criticisms of the consensus model of engineering design have also been
made, similar in some ways to the earlier criticisms of the architectural design models.
We discuss the similarities and differences between the two consensus models-in
engineering and architecture-and identify prescriptive vs, descriptive emphases. We
suggest that attempts should be made to reintegrate the two models to improve common
features of design education and practice across the disciplines.

1. Introduction
This paper is concerned with the differences that have arisen between models of the
design process in different disciplines, especially between engineering and architecture.
We find these differences both interesting and informative, especially since early
models of the design process in these disciplines had common origins and substantial
similarities. We propose in this paper to analyse the differences that have emerged and
to suggest the need to reintegrate the models.

2. Consensus Model of the Engineering Design Process

In engineering design, models of the design process have been developed since the
early 1960s. This development has converged to what might be called a consensus
model. The model is described in some VDI publications [1, 2] and in slightly different
versions in several textbooks (see, in particular, Pahl and Beitz [3] and Hubka [4]).
German engineering designers have been the main contributors to the development of
the consensus model. Comparable models, though less detailed and less precise, are
those of Van den Kroonenberg [5] in The Netherlands, and French [6] and Pugh [7] in
the UK.
The consensus model portrays the engineering design process as a sequence of
activities, leading to certain intermediate results: performance specification, function

Nigel Cross and Norbert Roozenburg, Faculty of Industrial Design Engineering) Delft Technological
University, 2628 BX Delft, The Netherlands.
 326 N. Cross & N. Roozenburg

structure, principal solution, modular structure (concept), preliminary layout, defini-

tive layout, documentation (see Fig. 1). The activities are usually grouped into four
phases: clarification of the task, conceptual design, embodiment design and detail
design.

Results

Clarify and define

1
the task I
Downloaded by [York University Libraries] at 16:23 12 August 2014

+ I
Determine functions
,
Specification
/
2 and their structures
I

• I
7
/ Function structure /

Search for SOlution

3 principles and their I
combinations I
~ /
t
I

Divide into realizable I

7 Principal solution
/
4 modules I
~ /
t
I

Develop layouts of I
7
Module structure
/
5 key modules I
t ~ I / Preliminary rovouts/
7
6
Complete overall I
layouf
I

~ /
I / Definitive loyoul

7
Prepare production and I
operating instructions
I
I / Product documents /

t
( Further realization

FIG. I. VDI model of the engineering design process.

This consensus model is fundamentally derived from the ways in which engineer-
ing design problems are conventionally perceived and modelled. Machines and pro-
ducts are technical systems that transform energy, material and information. The
functional behaviour of a technical system is fully determined by physical principles
and can be described by physical laws. The engineering design problem is to find and
define the geometry and the materials of the system in such a way that the required
 Modelling the Design Process 327

prespecified physical behaviour is realized in the most effective and efficient way.
Amongst others, Hubka and Eder [8) have set out these fundamentals in great detail.
Underlying this model is a structure based upon a systems engineering approach to
the development of complex systems. According to this approach, development
projects are to be structured in two 'dimensions'. The vertical dimension corresponds
to the origination phases in the life cycle of a product (such as feasibility study,
preliminary design, detailed design, planning for production, planning for distribution,
planning for retirement). The horizontal dimension is the problem-solving process that
takes place in every phase of the vertical structure: analysing and defining problems,
synthesizing solutions, simulating or predicting performance, and evaluating and
choosing the best system. This two-dimensional view can be traced to Hall [9) (see
Fig. 2); Asimow [10) was one of the first who applied this view to modelling the
engineering design process.
Downloaded by [York University Libraries] at 16:23 12 August 2014

STEPSOFTHE I 2 3 4 5 6 7
FINESTRUCTURE Problem Value Systems Systems Oprimiza- Decision Planning
(Logic) -----> Definition System Synthesis Analysis tion of AI· Making for Action
tematives
----------
PHASES OFTIJE
Design
(develop
(collect &
invent
(deduce
consequen- (iteration
(applica-
tion of
(to
implement
COARSE objectives altern- ces of al- of Steps I - value next phase)
STRUCTURE & criteria) erives) tematives) 4) system)
(Time)
1
Program Planning

2
Project Planning
(and preliminary
design)
3
System Develop-
ment (implement
oroicct clan)
4
Production
(or construction)

5
Disuiburion
(and phase in)

6
Operations
(or consumption)

7
Retirement
(and phase out)

FIG. 2. Hall's morphology of the systems engineering process.

However, the consensus model of engineering design tends to emphasize the

vertical dimension of this underlying structure. The horizontal dimension is not
strongly represented and possibly that is why this dimension is sometimes overlooked
-not so much by its authors (see, for instance, Pahl [11) and Hubka [4)) but by its
users and, above all, its critics, leading to faulty arguments and misinterpretations of
the model.
Two further characteristics of the consensus model should be mentioned. Firstly, it
assumes that design should proceed from the general and abstract to the particular and
concrete, i.e. the problem should be analysed in abstract terms, before any material
 328 N. Cross & N. Roozenburg

concepts are established, and those concepts are then gradually refined in increasingly
more detail. Secondly, it assumes that complex problems should be decomposed into
subproblems, for which subsolutions are to be found and 'synthesized' into overall
solutions for the design problem (see Fig. 3).

Overolt problem

Subproblems
Downloaded by [York University Libraries] at 16:23 12 August 2014

Individual problem,

Individual solufion,

Sub solutions

Overall solution

FIG. 3. VDI view of problem decomposition and solution synthesis.

By specifying in its vertical structure a set of intermediate results to aim for,

without prescribing in detail how these intermediate results are to be obtained, the
consensus model does not restrict designers to just one way of working. Instead, it tries
to organize the problem-solving behaviour of designers so that this behaviour will be
more effective and efficient than intuitive, unaided, unsystematic ways of working.
The consensus model is a 'methodology', though it is a weak or heuristic one. It is a
weak methodology because it is built upon weak knowledge (experience) about the
consequences of the possible actions to be taken (in terms of the insights gained with
respect to design decisions). It is a heuristic methodology because following it requires
'sensible' (knowledgeable or informed) interpretation by the designer of the vaguely
defined 'rules' and terms and-even if properly applied-success is not guaranteed.

3. Developments in Architectural Design Methodology

Early models of the design process in architecture were often very similar to models of
the engineering design process. For example, the model of architectural design
developed by Marcus and Maver [12] (see Fig. 4) had similarities to the model of
engineering design by Asimow. The editors of the 1962 Proceedings of the Conference
on Design Methods [13], Jones and Thornley, represented engineering design and
 Modelling the Design Process 329

architectural design, respectively, and both used similar models to help describe and
teach the design process. In industrial design there was also a similar basic model of
the design process represented by Archer [14] (see Fig. 5).

Design process,
2 Feasibility

3 Outline ptoposals
Downloaded by [York University Libraries] at 16:23 12 August 2014

4 SChemedesign

>-
Cl
o
<5
s:
o
Q.
5 Detail design
E
L-_H Apprnisal H~~

6 Production information

FIG. 4. Maver's model of the architectural design process.

Training

Brief Programming

I
Data collection

I
L...---+--Anolysis

I ---t-...J
Synthesis

I
Development -t--....J

Solution
l
Communication

FIG. 5. Archer's model of the industrial design process.

However, in the early 1970s, architectural design methodologists began to criticize

this shared view of the design process. Hillier et al. [15] were among the first to
 330 N. Cross & N. Roozenburg

question the orthodox view that designers should resist bringing their own preconcep-
tions to bear on the problem. They argued that designers not only do but inevitably
must 'prestructure' their problems in order to solve them. They suggested that the
prevailing 'analysis-synthesis' model of design (in which exhaustive problem analysis
must precede solution synthesis) was derived from a fallacious view of the role of
inductive logic in science and that designers have to rely on a form of preconception
-their prior knowledge of solution types. Hillier et al. therefore proposed a model of
design based (like Popper's view of science) on the centrality of conjectures. In this
'conjecture-analysis' model, the designer must first generate a solution conjecture
which is then subjected to analysis and evaluation, rather than analysis preceding
synthesis or conjecture. (Similar arguments and proposals in a systems engineering
context have also been made by Rzevski [16].)
Later, Darke [17] developed this model a little further. Based on interviews with
Downloaded by [York University Libraries] at 16:23 12 August 2014

architects, she proposed a 'generator-conjecture-analysis' model, which prominently

identified the role of a 'primary generator' in design. According to Darke, very early in
the design process the designer imposes (or identifies) a particular, strong generating
concept or a particular, limited set of design objectives. This again contradicts the view
that the design process must begin with an exhaustive problem specification, from
which solution concepts may then be synthesized. Instead, solution concepts are seen
to be based on 'primary generators' or 'prestructures',
March [18] also argued that the essential logic of the design process requires that
solution concepts are produced not by analytical, inductive or deductive reasoning but
by 'productive' reasoning (a form of Peirce's 'abduction'). (A comparable argument
with respect to the logic of design was developed by Eekels [19] and by Roozenburg
[20].) March outlined a rational design process as consisting of "(I) the creation of a
novel composition, which is accomplished by productive reasoning; (2) the prediction of
performance characteristics, which is accomplished by deduction; and (3) the accumula-
tion of habitual notions and established values, an evolving typology, which is
accomplished by induction". He proposed a 'P-D-I' model of the design process: an
iterative procedure of production-deduction-induction (see Fig. 6). The production
of a design proposal, March suggested, must be based on an initial statement of
requirements and on a presupposition or protomodel. There is similarity here to Hillier
et al.; they suggested a model comprising prestructures-conjecture-analysis and March
~uggests one comprising presuppositions-conjecture-analysis-evaluation.
The nature of design problems also came under scrutiny. Rittel and Webber [21]
argued that design problems in architecture and planning are inherently ill-defined (or,
as they called them, 'wicked') problems. They pointed to the limitations of the
analytical approach to wicked problems, and characterized design as a multidiscipli-
nary 'argumentative' process. It is now widely recognized that design problems are ill-
defined problems.
The result of these criticisms and reformulations of the design process in architec-
ture appears to have been a general rejection of any linear, sequential, analysis-synthe-
sis-evaluation scheme. There is no well-formulated consensus model of the design
process in architecture (nor any longer in industrial design) but we may conclude that
there has emerged a 'type model' with the following features:

• it has essentially a spiral structure;

• it .recognizes the importance of prestructures, presuppositions or protomodels as
the origins of solution concepts;
• it emphasizes a conjecture-analysis cycle in which the designer and the other
 Modelling the Design Process 331
Downloaded by [York University Libraries] at 16:23 12 August 2014

FIG. 6. March's model of the architectural design process.

participants refine their understanding of both the solution and the problem in
parallel;
• it assumes design problems, by definition, to be ill-defined problems.
These and other related arguments and developments have been presented by Cross
[22].
In retrospect, we might say that in architecture and industrial design the attention
of the design researchers and theorists shifted from the vertical (linear, procedural)
dimension of the design process to the horizontal (iterative, problem-solving) dimen-
sion.

4. Differences between the Models

After starting from common origins, there now appear to be significant differences
between the consensus or type models of the process of engineering design and that of
the process of architectural design. A major difference is the apparently linear,
sequential nature of the engineering model vs the spiral, cyclical nature of the
architecture model. Models of the engineering design process tend to emphasize the
sequence of stages through which a project is expected to progress (e.g. concept-
embodiment-detail stages), whereas the models of architecture and industrial design
emphasize the cycle of cognitive processes that the designer is required to perform
(e.g. productive-deductive-inductive thinking). In emphasizing the sequence of stages
that is expected to occur during project development, the engineering model is more
prescriptive; in emphasizing the thought-processes that have to be employed by the
designer, the architecture models are more descriptive.
In both architecture and engineering, it was at first proposed that solution concepts
should be synthesized only after rigorous and exhaustive analysis of user requirements
and other basic features of the problem. Architectural design methodologists now
 332 N. Cross & N. Roozenburg
stress the importance of generating solution concepts early in the design process,
drawing upon presuppositions. This shift is also a reflection of the prescriptive-
descriptive split.
Another example of divergence from a once shared view is concerned with
representations of problem structure. The engineering design consensus model is based
on a view of the problem structure as essentially a tree-like structure; the overall
problem can be broken down into distinct subproblems, each with sub-subproblems,
etc. After initially assuming that architectural and planning problems could also be
hierarchically decomposed as trees, Alexander [23] soon specifically rejected this view
[24], and most planners, architects and industrial designers now accept this.
There may be some straightforward reasons for these differences between the
models; In particular, the knowledge domains of the professions are different. Archi-
tects lack, for their essential problems, the equivalent of the engineering sciences that
Downloaded by [York University Libraries] at 16:23 12 August 2014

are available for engineering designers; the architects therefore have to rely much more
on trial-and-error design procedures. Architects also tend to view their design prob-
lems as inherently ill-defined problems, whereas engineers' problems are more usually
well defined.
There may also be fundamental differences related to preferences in cognitive
styles between engineers and architects. Such a dichotomy might arise because of
differences between engineering's science-based, problem-focused education and archi-
tecture's arts-based, solution-focused education, as suggested by Lawson [25]. Cross
[26] identified cognitive styles (such as serialist vs holist, convergent vs divergent)
with 'designing styles' and suggested that models of the design process by different
authors also could be characterized as representing different cognitive styles. Thus,
some models represent a serialistic design strategy, while others are holistic, some
represent a convergent design strategy, while others a divergent design strategy.
The characteristics of the consensus models of engineering design and architectural
design are contrasted in Table 1.

TABLE 1. Comparison of characteristics of the engineering and

architecture models

Characteristics of the Characteristics of the

engineering model architecture model

Assumes problems are (or can be) Assumes problems are ill-defined
well defined
Systematic, expert process Opportunistic, argumentative process
Starts with problem-analysis; avoids Starts with solution-conjecture; accepts
preconceptions prestructures
Linear Cyclical
Tree-like problem structure Lattice problem structure
Prescriptive of design behaviour Descriptive of design behaviour

s. Criticisms of the Consensus Model of Engineering design

The consensus model of engineering design has not been without criticism. To begin
with, it has been observed that the model has been developed with the design of new,
innovative technical systems in mind. Therefore it pays too much attention to the
conceptual design phase at the expense of the phases of embodiment and detailed
 Modelling the Design Process 333

design [11, 27]. In practice, many product design projects can or must do without the
'invention' of new technical principles and start from known, 'proven' concepts.
However, the consensus model offers little procedural advice with respect to embodi-
ment and detailed design. It has even been questioned whether detailed procedural
models for these phases may exist [28, 29], because the decisions to be taken in these
phases are strongly interrelated, owing to the complexity of technical systems. The
process in these phases is essentially one of continuously refining a concept, jumping
from one subproblem to another, anticipating decisions still to be taken and correcting
earlier decisions in the light of the current state of the design proposal. Therefore, it
seems that the iterative problem-solving model mirrors these phases better than does a
linear procedural model.
As in architecture, the model of engineering design has also been criticized from an
empirical point of view. In practice, the behaviour of engineering designers very
Downloaded by [York University Libraries] at 16:23 12 August 2014

seldom resembles the behaviour prescribed by the consensus model. Several authors
[27, 30, 31] have stated that, contrary to what the consensus model assumes, working
from abstract problem formulations to concrete solutions and splitting problems into
subproblems are iterative and recursive processes that rely upon anticipations of
possible solutions. As such these observations do not disqualify the model, because it is
a prescriptive model that intends to structure, and not to predict, design behaviour, but
there is not much sense in prescribing 'impossible' behaviour. However, there is until
now little empirical evidence in favour of the effectiveness of the model contrasted to
conventional intuitive ways of working, or against it. Empirical research into the
engineering design process is only beginning to gain momentum, so opinions on the
value of the model as a heuristic method for engineering design are still largely based
on personal experiences and beliefs in its rationality.
The model has also been criticized from a methodological point of view. For
instance, function structures might be of limited heuristic value [32], partitioning the
design process into a large number of small steps might lead to an uncontrollable
explosion of possible solutions [30] and often, when appearance is important, the total
form of a product has to be determined before or in relation to the form of the parts
(as 'dictated' by the chosen principles [2, 33]).

6. Towards Reintegration of the Models

We believe that there are good reasons for trying to get the consensus or type models
to converge, i.e. for seeking to integrate the two types into a common version again. It
is clear that both models have particular strengths and weaknesses. For example, it is
obvious that all designers need to progress their projects in a sequence of stages,
similar to the engineering model; it is also obvious that designers must employ varying
cognitive procedures during the design process, as in the architectural model. In
contrast, a weakness of the engineering model is that it emphasizes problem analysis
and specification, perhaps at the expense of innovative solution generation; a weakness
of the architecture model is that it emphasizes early solution conjectures, perhaps at
the expense of adequate problem clarification. A more developed, generalized model of
the design process would integrate the strengths of both models, while avoiding their
weaknesses.
In particular, it is in design education that integrated models are now urgently
required. If a design education curriculum is based on extreme versions of either type
model, then not only do we risk training future designers with impoverished under-
standing of the nature of design and with crippled design ability, but also we shall lose
 334 N. Cross & N. Roozenburg

the benefits and progress that genuinely have been made in the course of developing
both the consensus models over the recent decades. These comments apply not only to
the education of professional designers but also-and perhaps especially-to the
rapidly growing development of design in general education (i.e. in the schools), where
the educational weaknesses of simplified models of the design process have already
been criticized [34].
An integrated model is still a long way off although some authors have attempted
to move towards it [35]. The present authors have also each attempted some
integration in their recent books. Roozenburg and Eekels [36] deal with both type
models. They consider the basic design cycle (see Fig. 7), a model of the problem-
solving process in design, as the more fundamental model, and use this cycle (instead
of a linear procedural model) as the frame of reference for the discussion of
methodological problems and methods.
Downloaded by [York University Libraries] at 16:23 12 August 2014

function

value of the design

FIG. 7. Roozenburg and Eekel's model of the basic design cycle.

Cross [37] has proposed a 'hybrid model'-it has descriptive and prescriptive
traits-(see Fig. 8) with the following features. Firstly, a symmetrical relationship is
assumed between problem and solution, and between subproblems and subsolutions.
This attempts to demonstrate and indicate that the relationship is not one way from
problem to solution, but that problem definition is often dependent upon solution
concepts. This is an acknowledgement that making solution conjectures is often a
 Modelling the Design Process 335

means of helping to clarify the problem. At lower levels, there are similar interactions
between identifying subproblems and generating subsolutions. The upper and lower
double-headed arrows in Fig. 8 indicate that the designer's thinking will oscillate to-
and-fro between problem/subproblems and solution/subsolutions.
Downloaded by [York University Libraries] at 16:23 12 August 2014

FIG. 8. Cross's model of the engineering product design process.

Secondly, there is a hierarchical relationship between problem and subproblems,

and between solution and subsolutions. This attempts to demonstrate and indicate that
an important and inescapable part of problem clarification is the activity of decompos-
ing it into subproblems, for example, by identifying subfunctions and specifying
performance requirements. On the other side of Fig. 8 is indicated the necessity of
building the overall solution from subsolutions, for example, by generating, combining,
evaluating and choosing appropriate subsolutions. The descending and ascending
arrows at the left and right of the diagram indicate that the designer will expect to
progress from problem to subproblems and from subsolutions to solution.
Finally, within this overall framework there is a proposed set of design activities
(which are related to a set of design methods). These activities and methods promote
and assist the design process, whether this is exploring the problem-solution relation-
ship, decomposing problems into subproblems or synthesizing subsolutions. It should
be noted that the activities could also be presented as a sequential process, starting
with 'clarifying objectives' and ending with 'improving details'.

7. Conclusion
We have discussed some positive and negative features of both consensus models. The
consensus model of engineering design is essentially a concise prescription of the tasks
in a design process. It is strong in its rationality-as founded in the theory of technical
systems-but has some shortcomings with respect to the cognitive processes that take
place in the heads of designers. In contrast, the models that nowadays prevail in
architecture reflect design as it is carried out by practitioners. These models are
primarily descriptive and, hence, they offer little guidance to those who believe-as we
do-that better ways of working than those in practice are possible and worthwhile to
develop.
 336 N. Cross & N. Roozenburg

We have argued for the need to make a reintegration of the consensus models of
engineering and architecture, building on the strengths of each. Good models will be
built upon rationality adapted to the properties and features of the tasks to be
performed, and to the cognitive characteristics of the designer. This calls for an
integration of the insights that have been gained from design methodology in both
engineering and architecture, if design practice in general is to benefit from these
insights. Above all, it is in education that models of the design process are needed that
are neither overly prescriptive nor weakly permissive, but are reliable, robust and
formative of good design behaviour.

REFERENCES
Downloaded by [York University Libraries] at 16:23 12 August 2014

[I) VDI-Richtlinie 2222 Blatt 1 (Entwurf): Konzipieren technischer Produkte (Dussel-

dorf, VOl, 1973, Endgltige Ausgabe, 1977).
[2) VDI-Richtlinie 2221 (Entwurf): Methodik zum Entwickeln und Konstruieren
technischer Systeme und Produkte (Dusseldorf, VDI, 1985).
[3) PAHL, G. & BEITZ, W. (1986) Konstruktionslehre; Handbuch fur Studium und
Praxis, zweite, neuerarbeitete und erweiterte Auflage (Berlin, Springer).
[4) HUBKA, V. (1989) Principles of Engineering Design (New York, Springer).
[5) KROONENBERG, VAN DEN H. (1986) CAD applications in the creative phases of
the methodical design process, in: Proceedings of the Second International Confer-
ence on Computer Applications in Production and Engineering, CAPE '86, Copen-
hagen, pp. 339-348 (Amsterdam, North Holland).
[6) FRENCH, M.J. (1985) Conceptual Design for Engineers, 2nd edn (London, Design
Council).
[7) PUGH, S. (1991) Total Design: Integrated Methods for Successful Product Engineer-
ing (Wokingham, Addison-Wesley).
[8) HUBKA, V. & EDER, W.E. (1988) Theory of Technical Systems: a Total Concept
Theory for Engineering Design (Berlin, Springer).
[9) HALL, A.D. (1962) A Methodology for Systems Engineering, 2nd edn (Princeton,
NJ, Van Nostrand).
[10) ASIMOW, M. (1962) Introduction to Design (Englewood Cliffs, NJ, Prentice Hall).
[11) PAHL G. (1990) Notwendigkeit und Grenze der Konstruktionsmethodik, in:
HUBKA, V. & KOSTELIC, V. (Eds) Proceedings of the International Conference on
"Engineering, ICED '90, Dubrovnik, pp. 15-30 (Zurich, Heurista),
[12) MAVER, T.W. (1970) Appraisal in the building design process. in: MOORE, G.T.
(Ed.), Emerging Methods in Environmental Design and Planning (Cambridge,
MA, MIT Press).
[13) JONES, j.c. & THORNLEY, D.G. (Eds) (1963) Proceedings of the Conference on
Design Methods (Oxford, Pergamon).
[14) ARCHER, L.B. (1984) Systematic method for designers, in: CROSS, N. (Ed.)
Developments in Design Methodology (Chichester, Wiley).
[15) HILLIER, W., MUSGROVE, J. & O'SULLIVAN, P. (1984) Knowledge and design, in
CROSS, N. (ed.), Developments in Design Methodology (Chichester, Wiley).
[16) RzEVSKI, G. (1981) On the design of a design methodology, in JACQUES, R. &
POWELL, J. (Eds), Design:Science:Method-Proceedings of the 1980 Design Re-
search Society Conference, pp. 6-17 (Guildford, Westbury House).
[17) DARKE, J. (1982) The primary generator and the design process, in CROSS, N.
(Ed.), Developments in Design Methodology (Chichester, Wiley).
 Modelling the Design Process 337

(18] MARCH, L.J. (1984) The logic of design, in: CROSS, N. (Ed.) Developments in
Design Methodology (Chichester, Wiley).
(19] EEKELS, J. (1983) Design processes seen as decision chains; their intuitive and
discursive aspects, in: Proceedings of the International Conference on Engineering
Design, ICED '83, Copenhagen, Vol. 3 (Zurich, Heurista).
(20] ROOZENBURG, N. (1992) On the logic of innovative design, in: CROSS, N., DORST,
K. & ROOZENBURG, N. (Eds), Research in Design Thinking, pp. 127-138 (Delft,
Delft University Press).
[21] RITTEL, H.W.J. & WEBBER, M.M. (1984) Dilemmas in a general theory of
planning, in CROSS, N. (Ed.) Developments in Design Methodology (Chichester,
Wiley).
(22] CROSS, N. (Ed.) (1984) Developments in Design Methodology (Chichester, Wiley).
(23] ALEXANDER, C. (1964) Notes on the Synthesis of Form (Cambridge, MA, Harvard
Downloaded by [York University Libraries] at 16:23 12 August 2014

University Press).
(24] ALEXANDER, C. (1966) A city is not a tree, Design, 206, pp. 46-55.
[25] LAWSON, B. (1984) Cognitive strategies in architectural design, in: CROSS, N.
(Ed.) Developments in Design Methodology (Chichester, Wiley).
(26] CROSS, N. (1985) Styles of learning, designing and computing, Design Studies,
6(3).
(27] FRANKE, H.J. (1985) Konstruktionsmethodik und Konstruktionspraxis; eine kri-
tische Betrachtung, in: Proceedings of the International Conference on Engineering
Design, ICED '85, Hamburg, Vol. 2, pp. 910-924 (Zurich, Heurista),
(28] ANDREASEN, M. (1983) Gestaltungstheorie und Methoden, in: Proceedings of the
International Conference on Engineering Design, ICED '83, Copenhagen, Vol. 1,
pp. 189-196 (Zurich, Heurista).
(29] PAHL, G. (1983) Vorgehen beim Entwerfen, in: Proceedings of the International
Conference on Engineering Design, ICED '83, Copenhagen, Vol. 1, pp. 209-219
(Zurich, Heurista).
(30] RIEHM, U. (1983) Einige konseptuelle Miingel der Konstruktionswissenschaft
und ihre Auswirkungen auf CAD, in: Proceedings of the International Conference
on Engineering Design, ICED '83, Copenhagen, Vol. 1, pp. 314-326 (Ziirich,
Heurista),
(31] RUTZ, A. (1985) Konstruieren als gedanklicher Prozess, Dissertation (Technical
University of Munich).
(32] SCHREGENBERGER, J.W. (1985) Neue Impulse fiir die Konstrucktionsmethodik,
in: Proceedings of the International Conference on Engineering Design, ICED '85,
Hamburg, Vol. 2, pp. 893-898 (Ziirich, Heurista).
(33] TJALVE, E. (1981) Form design; a systematic approach, in: Proceedings of the
International Conference on Engineering Design, ICED '81, Rome, Vol. 1, pp.
559-571 (Ziirich, Heurista).
[34] JEFFEREY, J.R. (1990) Design methods in craft, design and technology, Journal of
Art and Design Education, 9(1), pp. 57-70.
[35] TOVEY, M. (1986) Thinking styles and modelling systems, Design Studies, 7(1),
pp. 20-30.
[36] ROOZENBURG, N. & EEKELS, J. (1991) Productonaoerpen: srructur en methoden
(Utrecht, Lemma).
[37] CROSS, N. (1989) Engineering Design Methods (Chichester, Wiley).