Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Good and bad design

  1. Oct 9, 2005 #1
    What is the difference between a good and bad design? What characteristics does a good design need to possess?
  2. jcsd
  3. Oct 9, 2005 #2


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Just a start - a good design has to perform the task it was intended for and must do so for as long as possible without failure and at the lowest possible cost. It must be safe and meet all governing regulations.

    Generally, the best design is the most simple but that's often a matter of perception.
  4. Oct 9, 2005 #3


    User Avatar
    Science Advisor

    Good design = Fulfills all customer requirements

    Bad Design = Does not fulfill customer requirements
  5. Oct 9, 2005 #4
    The customer often doesn't know what their requirements are. They usually know what they want a design to do, but they hire a professional to achieve that purpose the most cost-efficient, energy-efficient, time-efficient, and code compliant way possible.

    There are usually many ways to design a project. For an example, let's look at heating, air-conditioning, and ventilating a school.

    1. You can use electric heating and cooling Unit Ventilators in each classroom. (Requires Unit ventilators, larger electric system, simpler controls)

    2. Unit Ventilators in each classroom with steam heating coils and seperate chilled water coils. (Requires Unit ventilators, Boiler, Chiller, 4-pipe, Chilled water pumps, condesate return pumps, boiler feed system, more complex controls).

    3. Unit Ventilators in each classroom with hot heating and chilled water coils. (Requires Unit ventilators, boiler room, Boiler, Chiller, 2-pipe with changover valve or 3-pipe with changeover valves or 4-pipe system, Chilled water pumps, condesate return pumps, boiler feed system, more complex controls).

    4. Variable air volume (VAV). (Requires mechanical room with Large air handling equipment, Ductwork, VAV boxes, diffusers, boiler room, Boiler, Chiller, 4-pipe, Chilled water pumps, condesate return pumps, boiler feed system, even more complex controls).

    And so on, there are many more options.

    The code says you must provide fresh air to each classroom. You can provide a set amount per-person (obtained from a chart in the code), you can use special filtering such as electro-static filters and cut that amount in half, or use CO2 sensors and vary the amount of outside air by monitoring the amount required to reduce CO2 in the room, you can use enrgy recovery units.

    So which of those ways is good design?

    It depends on the customer. There is good and bad in each. What is good design for one, may be bad design for another.
  6. Oct 9, 2005 #5


    User Avatar
    Science Advisor

    I have to disagree with you on that. The customer will always know on even a basic level what they want. Things like cost-efficient and energy-efficiency are basic knowns that anyone will know when stating a design need. Even if energy or cost efficiency is not a concern on the part of the customer, it will be stated somehow. When would anyone ever have a design where the subject of cost didn't come up? The only items I see the customer not having knowledge about would be in the areas of safety and specific codes that may dictate or impact the design. In that case it is the designer's responsibility to make sure those criteria are met.

    There has to be a basic framework from which to judge the design. That framework is the customer requirements. If it does what the customer wanted without loss of life or wiping out cultures or species then it is a good design.

    Now, if we are talking about judging a design from completely outsider's frame of reference, then the notion of "good" gets a lot more complicated.
  7. Oct 9, 2005 #6


    User Avatar
    Gold Member

    As somoene in the usability industry, I can tell you that it is not a given the users know what they want.

    - often have a myopic view of what they want to accomplish, missing the bigger picture
    - often consider only their own needs, not the needs of everyone who will be involved
    - have preconceptions about how - and what - things can and can't be done
    - etc.
  8. Oct 9, 2005 #7


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Fred, I rather like your definition myself. Short & to the point. Though the customer may not always know what his requirements are, you can be sure that in the end, a good design must meet them.

    To elaborate on what DaveC' said, how about maintainability? I've had to work on my own equipment and can tell you I've come to realize how important it is to be able to take it apart and put back together. You have to be able to adjust things and test them easily. There's a lot of items one needs to put an a "good design" checklist. In fact, there should be such a thing as a "good design checklist". I've seen some companies that use them.
  9. Oct 9, 2005 #8
    A customer knowing what they want in a design is not the same as knowing the requirements of that design. Or how to achieve those objectives. That is why they hire design professionals instead of designing the job themselves. To state they want a cost efficient design does not mean they know without being told which design is indeed most cost efficient to install, to maintain and to run. A large part of our business is doing studies to determine that very thing.
  10. Oct 10, 2005 #9


    User Avatar
    Science Advisor

    Again, how are you referencing your judgement of the design? From the customer's standpoint, if you have a proper working design that meets their needs and what they put forth in their requirements then it is a good design. If you judge based on someone who is not the customer then the rules change. I have done plenty of designs in which the customers did not care about the cost or the efficiency of the design. It simply needed to work the way they needed it to. The end result was much more important than the other factors. From the customers' standpoints, they were "good" designs. If you look at it from the viewpoint of someone outside of the transaction between designer/customer then I think most would agree that it would not be a good design.

    That is absolutely correct. However, the statement "To state they want a cost efficient design..." is EXACTLY a customer requirement. That is how the customer will relay to the designer what they want. They are not going to usually tell you that they want an exact efficiency number. They're going to tell you what they want in vague statements like that. Once they state that THEN you have another design requirement and THEN it is a requirement to be considered a "good" design.
  11. Oct 10, 2005 #10
    FredGarvin, I think we are in the same book, but we can't find the same page at the same time. First of all,
    I agree with your first statement:

    My point simply adds that while a knowledgeable customer knows what they want the final product to do, they don't know the best way to get to that point. If they did know, they wouldn't need engineers to do the work for them, we would never be asked to do cost estimates or life cycle cost analysis because they would not need these tools to help make the decisions.

    Ultimately, however, your statement is exactly true. If the customer does not get their requirements met (whether they determined them or we did), it is a bad design.
  12. Oct 10, 2005 #11
    I don't know why, but my feel is that most of the discussion, to now, involves "official" customers", such as Government or corporate entities. Do these considerations also hold for the "man in the street" buying consumer products? What are "customer requirements" in this arena? Here there are a couple of interesting questions:

    What part does "planned obsolescence" play? Do these supposedly deliberate tactics (admittedly more managerial than engineering) contribute to good design or bad design --- if they bring the buyer back to buy again? There is strong belief that this was practiced in the US auto industry from the mid-thirties through the mid-seventies. (This seemed to work well for them during much of that period, when they had monopolies, however now, even though they have become among the most reliable and cost-efficient in the world, their reputations among many are still in the dumper.)

    How about much of Today's software industry. Here, the belief is that those producers constantly revise their products and at the same time drop support for older versions while making sure that those older products are not forward compatible, in order to force people to keep buying. Is this good design, if it enhances profit?

  13. Oct 10, 2005 #12


    User Avatar
    Gold Member

    Hi guys;
    My approach to such things is on a tiny scale compared to those of you who do major projects for high-level clients. Once in a while an acquaintance (or in a couple of cases an employer) asks for some gizmo or process to do some silly-ass little thing like automate some series of tasks or perform a formerly manual task. I don't suppose that a corporate structure such as your employers would want to 'waste' the time on it, but I always sit down with the guy and make sure that he understands every possible aspect of the design, ie: initial cost, upkeep, impact upon other systems or procedures, legality (electrical codes and whatnot), and even what sort of aesthetics might be required. The end result is often something a wee bit more complex than he originally expected, but cheaper in the long run than sticking to the basic idea and then having to correct things later.
  14. Oct 11, 2005 #13


    User Avatar
    Homework Helper
    Gold Member

    Leon, you might find this site interesting and useful.
    http://www.baddesigns.com/examples.html" [Broken]
    Last edited by a moderator: May 2, 2017
  15. Oct 11, 2005 #14

    Having been involved in DOE contracts, automotive electronics, consumer electronics, and even 1:1 consulting with the man on the street, I'd say most of the issues discussed are the same. No one knows really what they want, or even the definition of a correct outcome. Its an ambiguous mess that has to be sorted through, and is usually a function of many costly iterations.

    Planned obsolescene is alive and well. In the 90's when the MTBF criteria was reduced for the automakers, I am certain that planned obsolencence did play a role, although I was not privey to such meetings. Otoh, to neglect its potential role in the decision process would be like burying ones hand in the sand.

    As far as software goes, recurrent purchases are the life blood of most businesses. Their is also the cost/benefit ratio of supporting the old designs. Once a product reaches end of life, the infrastructure costs needed to support it start to climb multifold. The EOL may be dictacted by internal or external factors as well, irrespective of planning.

    As far as planned obsolence goes and being good or bad design. If we look at a big picture view, things change. For example, if WIN31 had not been replaced, no doubt Apple, or someone else would have come out with a superior operating system... and put WIN31 out of the running. At some point, the financial capability to support such a product would no longer exist, and as a result WIN31 would be obsolete either way.

    Otoh, there do get to be ethical issues, for example, designing a system with a calender function which kills it is probably going too far, but then again, its not all that uncommon in higher end software where a yearly license fee is required. I guess if its known upfront, maybe its not too bad, but to have your highdef TV shutdown 1 day after the warranty period without prior notice would be pretty poor design.

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook