Knowledge Transformation in Conceptual
Design: An Approach to Build a Model of CAID Knowledge System
TAN Hao, PhD
Candidate
School of Design, Hunan
University, Changsha, China.
e-mail: htan@hunu.edu.cn
Prof. ZHAO
Jianghong
School of Design, Hunan
University, Changsha, China.
e-mail:zhaoma@public.cs.hn.cn
Abstract
Conceptual design is a process of transformation
from the interpreted initial state into the requirement-matching goal state, in
which knowledge transformation plays a key role. From a perspective of design,
a model of knowledge transformation is built, and there are two stages in
knowledge transform, one is knowledge analysis charactered by discovering and
finding in realm of theory, and the other is knowledge synthesis charactered by
inventing and making in realm of practice. Then, knowledge and their
transformation in conceptual design are discussed in the paper in terms of four
cognitive levels: physical, perceptual, functional and conceptual. And a
protocol experiment was adopted to demonstrate the model of knowledge
transformation. Based on the model of knowledge transformation, the paper
presents a conceptual model of a CAID knowledge system, which offers design
aids for designers throughout two stages, and may generate new forms and
designs if with AI technologies, like Generative Design.
Key Words: Knowledge
Transformation, Conceptual Design, Cognition, Knowledge
1. Knowledge and Design
Design is a kind of business in which skills
and expertise are acquired after learning basic techniques, assimilating domain
specific and general knowledge, and absorbing past good cases. The acquired
expertise, however, is often tacit and implicit. Even experienced designers cannot
be clear what kind of expertise they use in designing and how. [1] This is one
of the problems in design researches.
In order to solve the problem, Gero, Kavakli,
Teeravarunyou, Sato have introduced the concept of information and knowledge
into design research and made some progress. [2-4] Knowledge is a type of
integrative and meaning-independent information which can be learned, applied
and be fed back and be used to forecast things. [5] The essential
characteristic of knowledge is that knowledge can be put into practice.
Knowledge plays a key role in conceptual design, such as aesthetic knowledge,
functional knowledge, user knowledge, sponsor knowledge, etc. [6] In some case,
conceptual design is one stage of knowledge transformation process that designers
transform their aesthetic knowledge, functional knowledge, user knowledge etc.
to products, and products express knowledge to users. (Fig.1)
Design process becomes a
process of knowledge transformation.
Fig.1 Knowledge
transformation from designers to users
2. Two
Stages of Knowledge Transformation in Conceptual Design
Fig.2 Knowledge using and
building
Conceptual design is a process of transformation
from the interpreted initial state into the requirement-matching goal state. In
conceptual design process, knowledge transformation is mainly during the
designers’ works. Knowledge is used to create objects, and objects are
evaluated to build knowledge [7]. Knowledge using and knowledge building are
two ways of knowledge transformation in conceptual design. (Fig.2) By the
transformation, knowledge changes from inner state into outer state, and
products changes from original state into goal state.
Fig.3 Knowledge
transformation cycle
From designers’ perspective,
conceptual design involves two stages: one is knowledge analysis charactered by
discovering and finding in realm of theory, and the other is knowledge
synthesis charactered by inventing and making in realm of practice. In the two
stages,
knowledge
using and knowledge building respectively do their works. (Fig.3)
In the realm of theory, designers
use design knowledge and explore the design methods, models and cases by design
research in order to acquire the design knowledge and gain the design model and
case. Then designers test or demonstrate the model and case to build their
design knowledge.
In the realm of practice, designers
use design knowledge and design products by their design skills and experiences
in order to express the design knowledge and gain products. Then designers
research users and markets to build their new design knowledge.
In Fig. 3, there are two knowledge
transformation cycles, which are ideas acquisition in realm of theory and ideas
expression in realm of practice. In each knowledge transformation cycle, there
includes knowledge using and knowledge building. For instance, in realm of
practice, knowledge using is a process of designing a new product. Obviously,
knowledge is the bridge linking the realm of theory and practice, the process
of finding and the making. Knowledge plays a key role in conceptual design.
3. An Experiment
In order to gain a detailed view of
the knowledge transformation, a protocol analysis experiment is put forward.
From human information process, designers’ mental process can be divided into
four cognitive levels: physical, perceptual, functional and conceptual. [7] By
analysis the protocol reports in the four levels, a knowledge transformation
model can be introduced.
Fig.4 One part of cognition action code analysis
of protocol reports
Five designers were invited to
take part in the experiment and heir task is to design one kind of NC machine
tools. All the design process is recorded by video. By protocol analysis, we
divide the entire protocols into small segmentation. For each segment, we code
cognitive actions of designers into the four categories. [1] (Fig.4)
Fig.4 is one part of the whole
protocol reports and it shows the process of mental analysis of designing the
scales of the machines. As it is shown in Fig.4, knowledge building and
knowledge using appear in design process and knowledge is the main elements of
design. On the other hand, the levels of information processing have an
inherent dependency on each other; and a design action on upper level should be
inherently dependent on that in lower level. If designers are not satisfied
with the works (after designers evaluation, it’s in conceptual level), they
begin a new process of knowledge transformation.
4. A Model of a CAID Knowledge
System
Based on the model of knowledge
transformation, we build a conceptual model of a CAID knowledge system. (Fig.5)
In the conceptual model, CAID
knowledge system can be divided into two parts: knowledge building parts which
applies knowledge aids and knowledge using part which applies design aids. Each
part has 4 knowledge transformation levels, which are physical, perceptual,
functional and conceptual. The four levels perceptively help designers’
designing in design process. On the other hand, theory knowledge and practice
knowledge are also presented in the stages of design.
By the model of CAID knowledge
system, knowledge can be transformed in design process. Furthermore, in some
case, the CAID knowledge system may generate new forms and designs if with Gene
Code, Case-based reasoning etc. AI technologies, like Generative Design.
Fig.5 Conceptual model of a CAID knowledge system
References
[1] Masaki Suwa, Terry Purcell, John
Gero. Macroscopic analysis of design processes based on a scheme for coding
designers’ cognitive actions. The Great Brain: Design Studies (19), 1998
[2] Manolya Kavakli, John S. Gero. The
structure of concurrent cognitive actions: a case study on novice and expert
designers. The Great Brain: Design Studies (23), 2002
[3] John S. Gero. An approach to the analysis
of design protocols. The Great Brain: Design Studies (19), 1998
[4] S Teeravarunyou，K Sato.
Object-Mediated User Knowledge Elicitation Method. In: proceeding of the 5th
Asian International Design Research Conference. Korea，2001
[5] Devlin，Keith. Infosense. Turning
Information into Knowledge. New York: W.H. Freeman and Company, 1999
[6] Jose Carlos Teixeira. Applying Design Knowledge to Create Innovative
Business Opportunities. Ph.D. Essays in Illinois Institute of Technology, 1999
[7] Charles L. Owen. Design research: building
the knowledge base. The Great Brain: Design Studies (19), 1998
[8] Suwa, M and Tversky, B. How do architects
interact with their design sketches in exploring design ideas. Proceedings of
4th Australasian Cognitive Science Conference 1997 Newcastle, Australia (1997)
in press