Wednesday, September 3, 2008

Addition to Week 2 assignments

Thanks to everyone for being patient with schedules this week.

To catch up on time lost this week we will use the first hour of class to revisit the readings from Sept. 3 (Atman et al, Adams, Dorst & Cross, Mehalik & Schunn). To prepare for this - you have an assigment that should be posted to the blog by Monday evening. There will also be an assignment on Week 3 readings but that is in a separate comment.

On Blackboard for Wk 2 - Sept. 3 are three files to help you synthesize ideas from the readings.
1. Download the file "process model activity".

2. On the first page is a diagram of a process model - this is a diagram a design educator uses in his design class to help students think about "what is a good process, and why?" For this diagram - what do you think is "good" about this process and what are opportunities for improvement? Think about these ideas as you move onto the next step.

3. On the remaining pages are "design step" diagrams for some of the experts we talked about last week. The codes on the left (PD, Gath, etc.) are the same codes used by Atman et al that represent steps in a design process. The tickmarks represent time spent in those activities over the duration of the time they spent designing a playground. Based on what you understand of the readings - which timelines would you anticipate would be examples of "effective" processes (i.e., high quality solutions) and which would be examples of "ineffective" processes? Why?

4. How do your ideas of “good process” relate to process models of design (yours, those described in design texts, the experts you discussed last week)? Some of you might find that there are contradictions, some of you might see them as the same.

5. If you want to think ahead, there is a file on Blackboard called "least-most expert analysis". this is an overview of what the experts consider least important and most important about a set of ideas people associate with design activity. What do you see? Any surprises?


celia said...

Good about the process:
The whole process seems like a “simplified” design process, which is linear without iteration. This model may be appropriate for the first-year students, who do not have much experience about engineering design. So a simple and clear model could help them acquire a basic understanding of what is engineering design. Furthermore, several important stages in design have been included and stressed in this model.

Opportunities for improvements:
The present activities can be improved in several aspects: In the first two steps, students are asked to create concept through brainstorming and chosen the most promising one based in needs. Here, with students acquiring more design knowledge and skills, they should also be encouraged to create concept based on their domain knowledge and design experience, not just brainstorming. In addition, as this model is not a complete one, we can see that there are still a lot of stages missing in the whole process, like problem framing, etc. Besides, it should also be emphasized that design is an iterative process, not as linear as it shows in the program.

The timelines of expert 03-14 and expert 07-35 are good examples of effective design because :
1. more iterations 2 considerable time spent in modeling 3 sufficient amount of time spent in early stage of considering more constraints.

Ineffective process: Exp 04-26
1 few iterations 2 very few requests for information 3 lack of problem scoping. 4 the whole process seems to be linear, which is dominated by modeling.

I must say that now it seems to me that experts have their distinct way of design. We always say ** may contribute to a good design, based on our observation or expert’s description. Thus, what we are talking about good design is all about the common points across experts’ performance. In terms of individual expert, which method he/she will use may depends on the problem, context or constraints, etc.

A little bit surprised to see that problem decomposing is the least important. Maybe experts tend to consider problem as a whole? They do not need to decompose it?

CJ said...

One good thing about the diagram is the phase including brainstorming and creating many different concepts. The model is very simplified, which may be easier for students at lower levels to use, and the activities that are listed were placed in a logical order.

Some opportunities for improvement are largely due to the simplification of the process. A large portion of time is allotted to selecting the concept, but no effort is made to formalize the concept and finalize the ideas before building the prototype. There is no analysis phase (in terms of engineering modeling, etc.), but maybe this is for designers without experience doing engineering analysis and calculations. Building a prototype would likely require more time than 1 week, depending on the complexity of the project. It would likely be beneficial to have some additional time for revisions to the design and building the final design, especially if no analysis is conducted to evaluate the design before prototyping. Documentation should occur throughout the entire process such as recording ideas, sketches, alternatives, reasoning, requirements, objectives and goals, etc.

The important elements for good design practices as noted by Atman and/or Mehalik include: gathering information, exploring the problem representation, considering/exploring multiple alternatives, iterating, transitioning between steps, and paying adequate attention to each step in the process.

The processes which best seemed to incorporate these “good design” practices include 03-14, 07-35, and 01-01. Each of these designers seemed to do a great deal of gathering information and spent a significant portion of their time modeling. It seems that 01-01 used iteration as well because generating ideas continues throughout almost the entire process. 03-14 also considered feasibility and evaluated alternatives.

Although several key steps were described in the literature as contributing to good design practice, I think that a combination of many of the different steps would also lead to and “efficient” design process provided there is still slightly more concentration in the more important steps. This may be similar to Atman’s statement about paying adequate attention to each of the steps. 17-83 exhibits this notion, where several steps are being performed throughout the process, and often simultaneously (in combination with other steps).

04-26 seems to be an “inefficient” design process based on the criteria described in the literature. The designer did not do much information gathering, or perform many of the other steps besides modeling. Although the designer did perform problem definition, I wonder how thorough the problem definition was without gathering information. I also wonder how good the models were if not much additional information was determined.

little-t truth said...

The process diagram used by the design instructor clearly lays out the tasks associated with design in a fashion that can assist new learning designers to grasp the various stages to be utilized in the design process. This provides a reasonable assessment; each designer has their own opinion on the time required for various stages in the process, and about how much time should be spent associated with each task in the overall cradle-to-grave design project. The problem with this Gantt chart of the design process stems from the failure to clearly identify the iterative nature of the design process and the interrelated nature of various stages of the design process.

The readings from the week give some hints about what would tend to be”effective” and “ineffective” processes. The processes should be iterative, iterations between multiple design steps, and a point in time where the designer embraces the problem definition they have created in order to proffer a solution to the problem within given time constraints. Using these guides, I look at Exp 05-30, 03-14, and 01-01 as candidates for "effective" processes in their design. All three are iterative throughout the process, engage multiple design stages in their iterative process, and while manipulating the problem definition, seem to reach, by time indication, a clear sense of problem definition to be solved within the design time. Exp 04-26, 11-49, and 20-99 is candidates to me for being "ineffective" processes of design. Exp 04-26 lacks numerous iterative activities within their design process. This raises concern about engaging the individual engaging the design process with enough considerations to truly solve the process. Exp 11-49 had limited iteration during the problem definition stage, which may, or may not, be of concern in regards to the ct of defining the problem. Exp 20-99 has iterative behavior across multiple steps, although the individual seems to spend a lot of time at the end of the project allotted time working on the project definition, which could mean many things, however, my concern would be they are still redefining the problem.

I find that I am re-evaluating the design process as I start engaging this material. Previously, I only thought about the macro-level aspect of a process going from cradle-to-grave (i.e. the stages to be engaged in order to reach a solution to a given ill-structured problem). Now I am thinking more about how the process actually occurs. Iteration, for example, is an idea that I recognize as part of a "good process" and believe I have engaged in during several of my own design process opportunities, but I do not think about being iterative as I go about the design process. The other factor, which I am not sure how I engage, is the problem definition phase. I recognize the value of establishing a clearly defined problem for solution emerging from the proposed ill-structured problem; yet not sure how well I engage this stage.