; Designing with Precedents
Learning Center
Plans & pricing Sign in
Sign Out
Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

Designing with Precedents


  • pg 1
									                      Designing with Precedents
                                              Gert Pasman
                                   Delft University of Technology
                            Subfaculty of Industrial Design Engineering
                                      Jaffalaan 9, 2628 BX Delft
                                           The Netherlands
                                   e-mail: g.pasman@io.tudelft.nl

1        Introduction
         This paper describes a research project as part of the IDEATE project [1], which is aimed at
studying and improving the conceptualising process of industrial design. The goal of the IDEATE project is
to gain a better understanding of this conceptualising process in order to establish an advanced electronic
environment, which supports the industrial designer in the generation and development of new product
         In a first study, which consisted of a number of interviews with designers in their own design
environment, a number of key considerations for designing such a user environment for creative ideation
were identified [2]. One of these considerations was that designers rely heavily on existing or precedent
designs as input for their idea generation. All of the subjects reported the collection of precedents in the
form of product samples, product catalogues, photographs, slides etc. to be a major activity during the
conceptual phase.
         Thus these precedent designs form an important source of knowledge in the design process.
When faced with a design problem, a designer can draw reference to existing solutions that have a
bearing on the current situation and transfer the design knowledge that they embody to support the
generation and development of new product forms.

2        Objectives
         The results from this first study initiated a special interest into the use of precedents in the
design process. From a methodological perspective the objective is to gain a better understanding of the
role and meaning of precedent designs. What role do precedent designs play in the creation of new
designs and how can this role be influenced? From a technological perspective the focus is on the
development of a computer system, which supports the use of precedents as a reference base. In this light
an image database system is being developed, which makes use of visual representations of precedent
designs which the designer can consult before as well as during the design process. In what way then
should such a database be structured and how would its interface have to look like?
3       Studies
        Through a number of different studies we are trying to get some answers to these questions.
They will be described here briefly.

3.1     Design experiments
        In design experiments the effects of providing designers with visuals of precedent designs while
actively working on a design task, are being studied.
        In one experiment subjects were given the following task: to design a portable, cordless
telephone, specifically for a female executive.     At the start of this task they were provided with
photographs of prior designed telephones they could refer to during the completion of the task.
        Two ways of organising these precedents, random vs. typological, were compared. Half of the
subjects was given five collections of twelve precedents each, which were grouped together based on
shared typical features regarding function, form or use. Thus each of the five collections represented a
design type, in which the specific characteristics of the individual precedents are secondary to the
characteristics of the design type. The specific characteristics of the individual design thus become
secondary to the characteristics of the design type, as represented by the collection. In this way the
problem-specific knowledge that is embodied in an individual design is taken to a more generic, problem-
independent level, thus affording a transfer of design knowledge to a new design situation. The other half
of the subjects was provided with the same precedents, this time, however, presented in one, randomly
organised stack. All subjects were asked to produce one conceptdrawing as the result of their design
work. These conceptdrawings were then assessed in order to compare the performance of both groups.
Figure 1 shows two of the five collections used in the typological condition.

                                                        Figure 1
         The results of the experiment showed that the concepts of the typological group reflected more
of the characteristics of the five design types as represented by the collections than the concepts of the
random group. Thus the typological organisation would enhance the transfer of design knowledge, that is
represented by the type. Moreover, the typological organisation did help the subjects in the typological
group to break away from their preconceived ideas by making the characteristics of the individual design
secondary to the more generic characteristics of the type, thus affording a more flexible design approach.
More information on this experiment can be found in [3].

3.2      Product categorisation and assessment
         What kind of types do designers use when categorising products? Are these types induced by the
characteristics of the products or are they product-independent? How much agreement is there to be
found between designers and what is the influence on this of a specific context, as presented in a given
design brief? Through product categorisation and assessment tasks we are hoping to get more insight into
these issues.
         Through an experiment the influence of a specific context, such as a design brief, on the product
categories identified by designers as well as on their degree of agreement, was studied. Subjects were first
asked to categorise different sets of images of precedent designs according to their own preferences,
followed by a ranking of the same sets of images on a number of bipolar scales. Two conditions were used
in the experiment. Half of the subjects were given a brief, describing a specific design situation in which
they had to place themselves during the experiment. The other half of the subjects were given no external
context. Figure 2 shows an example of the categorisation task.

                                                     Figure 2

         Comparing the results on the ranking task showed that no major differences were found within
as well as between both groups. More information on this experiment can be found in [4].
3.3      User interface design
         From a technological perspective work has been done on the design and development of an
image database to support the industrial designer while conceptualising. A number of proposals for the
user interface of such a database have been developed. Traditionally databases are very strict and rigorous
in the way they enable their users to search the content as well as in the presentation of the found items.
Users are forced to verbalise their questions in terms of specific jargon they are often not familiar with.
The results of their queries are often presented to them in a sequential way, forcing the user to focus on
each finding individually, thus lacking an overview. Since we find this kind of behaviour in conflict with the
explorative and divergent nature of the conceptual phase, we are trying to come up with different
         An example of such a solution is the MDS-interactive interface. It makes use of a statistical
method, named Multi-Dimensional Scaling (MDS), for visualising similarity relationships between objects in
a database. It creates a 2D, 3D or even higher-dimensional arrangement of objects based only on
information about the amount of similarity between object pairs. Thus similar objects appear closely
together, dissimilar ones far apart.
         The user can now make a query by indicating a position on the MDS-layout. The database will
then find a sample from its content, which relates to the ones shown on the screen as indicated by this
position. The objects on the screen will arrange themselves then dynamically in order to find the best
possible fit. In the same way objects can be removed by dragging them out of view, thus creating a visual
dialogue for exploring the collection. Several simple, interactive prototypes have been implemented in
Macromedia Director™, followed by some first user testing. The result of which suggest that the dialogue
of adding and removing works quite well, but the specific design of the interface needs some improving.
Figure 3 shows an example of one of the prototypes. More information on this research can be found in

                                                  Figure 3
4       Future research
        Through this variety of studies we have already acquired a better understanding of the role and
meaning of precedent designs in the generation and development of new form concepts. From a
methodological perspective a way of organising precedents that affords knowledge transfer to new design
situations, has been developed and tested. Studying the categorisation and assessment of precedents by
designers provided us with more insight into the usefulness of the proposed organisation. From a
technological perspective the development of several user interface proposals has answered some of our
questions regarding the design of a computer support system using precedents.
        In the future we want to combine the methodological perspective with the technological
perspective by using a simple image database in an actual design situation. The database would then be
organised according to our proposed method as well as having elements of the suggested interface.

[1]     Hennessey, J. "The IDEATE Project: Exploring Computer Enhancements for Conceptualising" in:
        White, T., Tzonis, A. (Eds), "Automation Based Creative Design", Elsevier, Amsterdam (The
        Netherlands), 1992.

[2]     Kolli, R. Pasman, G. Hennessey, J. "Some Considerations for Designing a User Environment for
        Creative Ideation", Proceedings of Interface '93, Raleigh (USA), 1993.

[3]     Pasman, G. Hennessey, J. "Random versus Typological Organization of Precedents in a Design
        Task", Proceedings of the 4th Design Thinking Research Symposium, Massachusetts Institute of
        Technology, Boston (USA), 1999.

[4]     Pasman, G. Hennessey, J. "The Influence of Context on the Assessment of Precedents in Product
        Design" Proceedings of the 4th Asian Design Conference, Nagaoka (Japan), 1999.

[5]     Stappers, P.J. Pasman, G. "Exploring a Database through Interactive Visualised Similarity Scaling",
        CHI 99 Extended Abstracts, Pittsburgh (USA), 1999.

To top