Thursday, August 28, 2008

Discussion for Week 2 - September 4 - Design process knowledge

The papers for this week focus on what designers do from a process perspective. To guide discussion, consider the following questions (click on the link at the bottom to post your reflections):

- What do designers "do"? What does this suggest about what they "know" (in other words, what might be reasons for the ways they "do" design)?

- How does this relate to ideas about "good design"?

- How does this compare to what you know about design or to the representations of design we discussed today?

The readings:

  • Adams, R. S. (2002). “Understanding design iteration: Representations from an empirical study,” Proceedings of the International Conference of the Design Research Society, September, London.
  • Atman, C. J., Chimka, J. R., Bursic, K. M., & Nachtman, H. L. (1999). A Comparison of freshman and senior engineering design processes. Design Studies, 20 (2), 131-152.
  • Dorst, K. & Cross, N. (2001). “Creativity in the design process: co-evolution of problem-solution.” Design Studies, 22 (5), pp. 425-437.
  • Mehalik, M.M. & C. Schunn (2006). “What constitutes good design? A review of empirical studies of design processes.” International Journal of Engineering Education, 22 (3), Special Issue on Learning and Engineering Design.

8 comments:

§adieLovingtonNibblesworth said...

What do designers do? And what does this suggest about what they know? How does this relate to ideas about “good design” ? How does this compare to what you know about design or to the representations of design we discussed today?
This is a quadruple edged sword upon which any student can fall after this week’s readings. Dorst and Cross have explicitly suggested that design possesses an innate creativity, thus it follows designers are the originators of said creativity. I feel that the Dorst paper sets the tone for the nature of designing as he notes that time spent on defining leads directly to a more creative design; however, the key impetus behind creativity being in the concept of the “creative event” may not necessarily equal ORIGINAL designs. Given that original design and creative design may be at odds, I feel that Dorst would still say that creativity is a fundamental component of good design, although he is not explicit enough in the reading for me to support that claim directly. In class this week I saw representations of the creative process as endemic to the nature of design in many of the diagrams perused, albeit sometimes implicit.
The work by Atman and the subsequent work by Adams both take a study in design and extrapolate information about the designers that is complimentary to Dorst’s work. Atman notes that the seniors, aka the more experienced designers, are generally more efficient and this overall efficiency scales directly to quality of design. The subsequent work by Adams further explores the Atman study in the framework of iteration web diagrams to explicitly show “what iteration looks like.” Whereas Atman notices that there is no huge difference in overall time spent on any one facet of design between the senior and freshman designers (save with scrutinizing and endgame communication,) Adams notes clear demarcation in effective iteration strategies within her design framework. The work of Adams notes that beyond the time spent in any one facet of design, the flow and purpose of transitory iteration yields robust information that can more specifically indicate a “good” designer and a “good” design. I especially enjoyed the Adams paper since it unified many of the themes of iteration I tried to invoke in my drawing from class; the difference being that Adams’ framework yields more ground to design being a non-linear, iterative process, whereas my notion of iteration was stratified into independently self-iterating, linear chains.
Mehalik and Schunn’s work does not appear to be in the same vein as Atman’s, Adams’, and Dorst’s for the sake that its goal is different. Mehalik begins by attacking the notion of the design framework as being inaccurate in representing true design. While this is implicitly noted in the methodologies of Adams’ and Atman’s work, a framework is useful in both their works to prove overarching themes of and concepts such as that of diagnostic and transformative processes. Mehalik’s work bolsters the arguments in the other papers this week by noting that design studies and designers should not be characterized by the preliminary work they perform at the beginning of a design process. Also, Mehalik further expands upon the “uniqueness” of design thinking by defining it as an expansion of problem based learning that includes design and planning. The importance of design and planning as discussed by Mehalik resounds well with the concept of iteration being necessary to design as invoked by Adams, and is conjures images of good design as being “creative” from Dorst’s work. The strength of Mehalik’s argument lies in the meta-analysis used to separate the specifics of the design papers from what characteristics of design strengthen the design itself. I have to say that I am glad that Mehalik found graphical representation “significant” to good design, but am disappointed that it was reported less-frequently than other elements. In class I relied heavily on graphical representations, although I noted that the majority of the case studies viewed in class had few graphical interpretations.
The fundamental question yet to be answered by all of the research this week is just how a student who, in the nomenclature of Adams, is considered a non-canonical “C” category student, transforms into an “A” category student by the end of his senior year.

Carolynn (CJ) said...

While there are many “common practices” in design, it seems that each designer does design in a different way because each person experiences different thought processes. Even though design methodology and design processes are taught, much of design just happens as thoughts come to mind or as ideas are triggered. I think that what designers “do” may largely influenced by what they “know”, but also by the way they think – in ways they have been trained to think and in intuitive methods as well.

As mentioned in the article by Atman, design involves thinking in multiple areas. It is not just thinking about one subject, such as math, but rather involves creativity, analysis, reasoning, etc. While some people may excel more in one type of subject or area more than others, “good design” seems to require a combination of knowledge and application in many areas. Thinking about design in this sense, I wonder if students who are considered more “well-rounded” have a tendency to perform better in design (particularly engineering design) than students who are much stronger in one specific subject. Another question that came to mind for me was: Do students who are very strong in academics but do not participate much in other activities (sports, extracurricular activities, etc.) usually do better or not as well in design activities (maybe especially in the area of creativity)?

This also relates to the article by Adams about iteration in design. The article mentions that “freshman and senior subjects in this study did not significantly differ across academic and personal backgrounds” in terms of certain given measures. It continues to state, “therefore, it is not likely that differences in the representations can be attributed to personal characteristics.” I am still curious, however, in what ways personal characteristics can influence differences in design processes and design thinking. It is often difficult to remove bias or influence of these individual traits and thought patterns from studies involving humans, especially when considering thought processes. -C.J.

little-T truth said...

The first set of readings begins to prompt us to think about design, designers, and the relationship between characteristics and process of design and the design outcome. Design is a process that engineers, novice and experts, engage to solve problems that are “ambiguous, ill-structures, and may have multiple solutions” (Adams 2002).

Mehalik and Schunn provided a good review of the individual steps that can be associated with the design process as conceived through the meta-analysis of published studies on design. I found this article to be a good beginning point for considering the design process in a more interactive format. A clear perspective of the different stages of design was identified and somewhat defined. I clearly recognize the existence of the various stages identified in the article and further agree that there is some degree of significance associated with certain design elements influencing quality design outcomes.

One interesting assertion that emerged from the readings was that there appears to be a point in time where a problem is defined and the designer should move on with the process. There is a clear need for a designer to define the problem in a way that provides personal clarity to what they embrace as the task to be accomplished. The time in this space is important and occurs in an iterative fashion that is highly transformative as the designer vacillates between process steps – gather information, communication, generate ideas, decision, feasibility evaluation, and modeling – in an attempt to manipulate the initial problem space, that has its own solution, into a series of partial problem spaces (i.e. Maher model) that the designer feels provides them the best opportunity to present a solution to the initial problem. The difference between an expert, a novice, and a “seasoned novice” resides in the quality of connections in the iterations performed in defining the problem. Interestingly, the notion of creativity that may be associated with a design may be linked to the iterative capability of designer in the structuring and redefining of the problem, since this element of the design process provides the designer in a unique starting point that may result in what is perceived as a creative solution.

Other questions/comments of interest from the reading:

• Creativity as an evolved manifestation of an engaging a task?

• Levels of creativity?

• Representation of experts and novices in the papers. Is 5 years, four of which is in school, enough to earn a label of “expert?” I think of these individuals as “seasoned novices,” but not experts. Experts have to have had time to evolve in their process thinking. It is the evolution through experience by manipulating their mental models that alters their approach to ill-structured problems that makes them an “expert” since they are no longer beholden to the foundational design process provided in the educational routine of engineering.

Neha Gupta said...

When we think about "what exactly a designer 'knows' to be able to 'do' the things they do..", it is compelling to first understand what is it - "a good design" that we think a designer knows how to do. How do we judge a piece of work is a good design? I would commend Mehalik and Schunn for reviewing a range of journal articles to come to close proximity to the idea of what can constitute a good design. They try to investigate the elements of a design process that is proposed to lead a good design. While considering all these aspects of a design, it is needless to mention that all the stages are iterative so to lead to a good design. Also, Adams clearly states that iteration constitutes an effective design practice aiming to lead to a 'good design'.

Although the time spent during this iterative process or the time spent on the problem scoping significantly relates to the quality of the work, it is hard for me to believe that there is any definitive rule of thumb for dividing the time into the several stages of a design or the iteration process. Adams exemplifies the score of a high school senior who dedicates more time iterating than the other designers. As Dorst elucidated on creativity being fundamental to designing, it occurs to me if creativity is also the reason for iteration. Adams suggests that the design educators could use iterative activity in teaching to develop design strategies. While I completely identify with the idea of iteration, it keeps me wondering how iteration may be affecting the creativity of the designers. Dorst and Cross say that a 'creative event' can lead to a good design. 'How far a creative event can lead to an original idea that is also a good design?' is a difficult one to answer. An original idea may not be unique as found out by Dorst and Cross and 'uniqueness' does not necessarily mean that the design idea is a good and valuable for practical purposes given the design constraints.

Atman, Chimka, Bursic and Nachtmann have shown through their findings that experience can be significant when it comes to a good design like the seniors had higher quality designs as they gathered more information, considered more alternatives and iterations through their design steps which was not observed for the freshman designs. It seems to me that the lack of confidence and knowledge base amongst the freshman could have been the reason for their inability to gather as much informationand think over the design as creatively as the seniors.

From these readings, I think that a large portion of design skills are based on the designers’ practical knowledge and ability to think creatively. An iterative design process can certainly help to understand the problem clearly and come up with alternative solutions that may lead to a good design even if not as original.

Aidsa said...

There is no doubt about the importance of design in engineering, especially in engineering education. As Adams mentioned, design is a “prominent feature in how we educate engineers, accredit engineering programs, and one way of describing the competency of engineering graduates and practitioners”. I totally agree, as an educator I have seen how students struggle when faced to design a solution of a “problem” or situation. Also I have seen how us, the educators struggle in teaching students the engineering design process to be followed. I believe that this occurs because we (engineering faculty) tend to teach the way we were taught. There is no pedagogical expertise required to become an engineering faculty which affects negatively the quality of engineering education.

Atman et. al. described several characteristics of successful design processes as follows: (1) “the use of a prescribed methodology that allows for flexibility and opportunistic design, (2) the effective use of transitions among design steps, (3) the development of good conceptual models, including effective problem scoping”. In the paper they compared the design processes for freshman and senior students. My experience as a professor was with freshman engineering students where I noticed how freshman students spend a lot of time analyzing the problem thus they were unable to progress and design the solution to the problem. Their goal is to solve the problem from the first attempt and they “get stuck” analyzing the problem, seeking for the “optimal solution”.

This relates to one of the two of the aspects that called my attention form this week’s readings referred to as iterations and creativity. Adams defined iterations as “a symbolic feature in design models that represents a process of revisiting and resolving design conflicts”. I have seen how difficult is for freshman engineering students to analyze a design and come up with the “optimum solution”. But as Mehalik and Schunn explained, the “design process is an open-ended search process of deciding when a solution is good enough to be called complete”. It is important for a designer to understand when to stop searching for the “optimum solution” because there will always be something else that could be done.

Aidsa

Junqiu said...

“Design is a process including multiple steps and features. According to Atman et al, design is a central element of engineering and all engineers are some kind of designers, such as improving process, developing software, planning a project, or creating a new material.” While trying to discover what constitutes good design, Mehalik and Schunn created a list of process elements to form a framework of design. Fifteen design process elements were identified.
While studying the difference in designing between freshman and senior engineering students, Atman et al found that senior engineering students did better job in: 1) gathering enough information covering an adequate number of categories; 2) considering multiple alternatives during design development; 3) transitioning between design steps; 4) paying adequate attention to each step in the process. Mehalik and Schunn discovered that “explore problem representation”, “use/iteractive design methodology” and “explore alternatives” are the most important features of the design process after integrating design literatures. Those features fit with Atman’s finding pretty well. A good design should explore or gather enough information about the problem to be solved; should develop multiple alternatives and should use iterative design or paying enough attention to transitioning between design steps. Adams also found that iteration is a very important feature of effective design.
Design to me is basically to solve an open ended problem with limited resources.

celia said...

What do designers do? How do they approach the design task?
My first reaction to these questions is: who knows that answer? Of course, we may expect designers to tell us everything about design, like what they are thinking, what strategies they are using, etc. However, we know that even we ourselves, acting as both engineer and educational researcher, are struggling with the above questions.
Though we can not expect a complete answer from designers, we still would like them to “talk”, talk through their work and performance. Then we engineering educational researchers will try to translate, to explore, to better understand the design, even if only a corner of the big picture.
This week we have four readings and each of them shows a different aspect of design. Atman et al (1998) compare the freshman and senior performance as they approach an open-ended design problem. Three major stages of design process are discussed and compared in the article: problem-scoping, developing alternative solutions and project realization. Adams(2002) reexamine the data to illustrate that design is also a iterative process. Dorst and Cross (2001) study creativity in design process, which is a higher demand in engineering design. Mehalik and Schunn summarize some previous research, aiming to identify what constitute a good design.
One thing attracts my attention is that I noticed ,in Atman et al’s paper, the authors state that they find many interesting correlations among variables in freshman date, such as a strong correlation between quality score and total time spent in design. However, there are fewer interesting correlations among the senior data than there are in the freshman’s. I think this may reflect the development of expertise in engineering design in some degree. It is not difficult to imagine that freshmen, who haven’t received some particular training in design, only master some specific skills and knowledge. Their expertise is unbalanced and incomplete and they do not familiar with the design process. For example, the correlation of the quality score with the percentage of time spent in problem definition is -0.5. This may illustrate that students lack certain skills in defining a problem. Similarly, a positive relation may indicate that these freshmen have already had some initial skills in that aspect. Compared with freshmen, seniors have already had developed some important and useful skills and knowledge that they can rely on. They are already familiar with each step of the design process and they know good performance in each stage can contribute to a good design. This may explain why there are more moderate correlations between quality score and several other variables in senior data.
In terms of what constitute a good design, I think there are different ways for evaluation. Besides the design elements that summarized by Mehalik and Schunn, some design skills can also have some positive connection with a good design, such as problem-decomposition strategy, depth-first method. And I am still a little skeptical about Mehalik and Schunn’s way to identify what constitute a good design as it ignores some context. For example, Atman et al suggest that freshmen who spent a large proportion of their time defining the problem did not produce quality designs. In contrary, Dorst and Cross said that defining and framing the design problem is a key aspect of creativity, which means defining the problem contributes a lot to design. Then, how will somebody categorize “defining the problem” only based on these two simple statements? Is it bad or good?

Junqiu said...

Both Brereton and Stumpf talked about how the social interaction process shaped the designing, but with different research methods and pespectives.
Brereton took a qualitative approach to analyze a vedio tape about a designing process of a engineering team. In order to get many perspectives of the designing team, an interdisciplinary team was organized to watch the tape segments looking for interesting practices. The vedie tapes showed that designers in the vedio display very different preference towards both the ways of approaching a designing problem and the solution to a designing problem. Through the social interaction, negotiation was finally made through persuation to reach a final designing solution. One point is very interesting in the paper is how the team members seeking to fill in different role gaps they perceived missing in the group. For example, when two of the designer tried to persuade each other on a specific design, a third designer appeared as an arbitrator to facilitate their arguing, in order to get conflicts solved.
In conclusion to the paper, the content of an evolving design depends heavily on negotiation strategies and other more subtle and ubiquitous social process.
Stumpf took a quantative perspective to study how reflection-on-action can be used to help reflection-in-action in a designing process. Blahblah… to be continued.