iSixSigma

DFSS

Viewing 11 posts - 1 through 11 (of 11 total)
  • Author
    Posts
  • #27728

    Whitehurst
    Participant

    I need a minute to regain my balance after reading that GE debate!!
    We’re in the process of developing a DFSS implementation plan.  As with the lag in DFSS literature on the web, there’s also precious little advice on the best way to roll out DFSS.
    Specifically, in a new product development oragnization, DFSS seems to be a specialized set of skills, and not a broad methodology that can be easily taught to the masses and implemented across the board.  Instead, we’re leaning toward a limited number of DFSS specialists with honed skills in transfer function derivation, DOE, and design optimization techniques, to serve as internal consultants to product development teams on a case-by-case basis.
    Any similar experiences?  We’re concerned that the skills within the DFSS methodology are difficult enough that they will require almost constant use.
    Thanks,
    Joe
     
     
     
     

    0
    #68214

    Hogg
    Participant

    DFSS can be successfully applied as a methodology, however it does require a little adjustment away from the ‘standard’ DMAIC approach. One common DFSS methodology is DMADV, but I prefer to use DCCDI – define, customer, concept, design, implement – which is perhaps better named. Remember too, such methodologies need to be contained within an overall New Product Introduction strategy and process, adopted by the owning business.
    The tools typically used in DFSS are large and complex, and since DFSS projects are often large and infrequent, the use of ‘long term’ experts as ‘guardians of the methodology’ is a good idea. Such people should carry the methodology and tool expertise, as well as a culture of creativity balanced against customer and business accountability. In particular, software simulation of temporal processes, using such tools as ProcessModel, takes a great deal of time and effort to set up and maintain, and the use of specialists in these fields is often essential.
    In balance, you will need to consider the needs and demands of such specialists – remember Six Sigma is about breaking down the ivory tower of the unique specialist and in placing the power of the tools and methodology in the hands of the many and the team, rather than the special few. Although DFSS is about the ‘new’ rather than about ‘improving the broken’, it is a good idea to promote your own DFSS methodology for all – good design should be something everyone contributes to on a daily basis! A good approach is to apply some of the DFSS specific tools as part of a reduced DFSS methodology tied back in to the (already adopted) DMAIC approach. This works well for the small number of DMAIC projects that require design, but without the full DFSS treatment. It thus allows some of the methodology and tool use from ‘heavy’ DFSS to filter through to the ‘masses’.
    The organisation that designs the ‘new’ for the living will benefit from taking a framework DFSS methodology and adopting and incorporating it with any existing ‘new product’ introduction approach. Much of the benefit from the complete DFSS approach comes from the parts that lie outside of the technical design stage! If you can adopt and adapt a framework DFSS methodology to your own organisation and needs, then you will be able to make it a formal part of each and every new product/service introduction.
    My book on Design For Six Sigma covers in detail the question of what is DFSS and how to make it work as such a framework methodology. I have written this book, not as a list of tools (although some are covered) but as a high level guide to introducing DFSS as a complete ‘methodology’ to any organisation. Regrettably the time lag in publishing is considerable, and although completed some months ago, the book will not be out in print until the end of the year. However, it is now available for pre-order and has already begun selling.
    http://www.gowerpub.com/TitleDetails.asp?sQueryISBN=0566084341
    http://www.amazon.co.uk/exec/obidos/ASIN/0566084341
    Geoff

    0
    #68222

    denton
    Participant

    My experience, for what it’s worth, is that DFSS is easier if you have a DMAIC program already functioning. 
    I disagree with the notion that it is a good idea to have a few highly trained specialists, and exclude the others.  Would you want a situation where only certain engineers were allowed to use Ohm’s Law, and everyone else had to go to them to get their calculations done?  Good grief! These are engineering tools!!  We want them baked into the company DNA, not sequestered.  As long as the designers remember what the tool will do, and roughly how it works, they can restore the rest from references, or with a visit to their friendly, local MBB.
    I say, train them all, but require use of the tools at your phase gate reviews, so DFSS is part of their regular job description.
    :::stepping down from soapbox:::
     

    0
    #68226

    Whitehurst
    Participant

    Geoff,Sorry to hear your book is taking so long to publish. Good luck when it’s out.We’re clearly moving in the direction you describe. Our organization is small enough that the “specialists” will actually be imbedded into the launch teams, not set off in a separate group. Furthermore, we’re seriously considering a division of responsibilities for our specialist; time not spent in design optimization will be used to lead a Black Belt “DMAIC” project. In other words, our specialist will be a trained Black Belt with complimentary DFSS training. This will link the DFSS and DMAIC efforts, as well as the product team with the operational segment of our division (we manufacture on site).Denton,RELAX – we’ve got you covered (nice speech though). We learned long ago that concurrent engineering is the only way. We’re not about to remove the tools from the users. However, in our industry, the program management aspects of new product launch are cumbersome enough to anticipate the use of the DFSS tools for only short intervals in the development process.Thanks for the feedback,
    Joe

    0
    #68240

    David M. Cronin
    Participant

    First, let me start with a disclosure. My company is a vendor of DFSS process management software.Your question is a common one we hear from almost all of our customers. We have done some extensive research into successful approaches and have offered no charge briefings to companies such as yours to duscuss ‘removing barriers to successful DFSS implementation’. We recently offered a workshop at the IQPC seminar in Chicago.If you would like to receive our slide set or talk more please feel free to call me: 781-271-9300 x241.

    0
    #68241

    Rick Otero
    Member

    Joe,
    This is fairly new area for us as well, but your thoughts about a few experts are probably on track (I say this as I am writing the 3rd revision of DfSS training…so take ith with that grain of salt).
    The essence of DfSS is really the DMADV methodology.  It seems that any BB/MBB/Champion – at a minimum needs to be exposed to DMADV framework and enough training to know WHEN to apply DMADV.  The underlying concepts around Voice of the Customer (i.e. with Kano Analysis), Customer Driven CTQ’s (using techniques such as QFD), tolerancing, etc. all are derived from prior design methods.  So having a “generalist” training approach for all would at least let them know when to diagnose a DMADV implementation. 
    Then, as you suggest, supplementing with a few specialists – when you need the “die-hard” statistics for capability flow-down, tolerancing, monte carlo, etc. and use of more advanced software such as crystal ball – seems like a the most cost effective option.
    While you mention there is not much on the web.  I have learned a lot about the underpinnings of DMADV from Glen Mazur’s site http://www.mazur.net.  Glenn is a QFD expert and has some excellent case studies on application of Kano and QFD.  Additionally, while several folks offer DMADV training (we use GE)… I have noted a very innovative training session on DfSS has been developed by Rath and Strong of Lexington, MA  781-861-1700.
    We are applying DfSS to top line growth and revenue generation – while you are approaching new product development. There are some other on-line resources which are very complimentary to DfSS for new product development.  You might try Gene Lieb’s series of workbooks about new product development at :
    http://www.lieb.com/WORKBKS/guidebk.htm
    Good Luck!
     
    Rick Otero
    DuPont – Director, TLG and 6s MBB
     
     

    0
    #68266

    Ruddy
    Participant

    I’m going to make a few enemies by doing so but I agree with your premise that you should develop a few good PRACTICIONERS of DFSS.  If you will remember, in the development of Six Sigma it was discovered very early that black belts have to have a practical training as well as an academic training.  This could not be more true than with DFSS.  To a one, all of the major DFSS schemia (DMADV, DMEDI, DCCIME, etc) are all flawed.  Each makes assumptions and one or more of these don’t apply to your product, service, or even industry.  So what can your do?
    Well, you have to practice Six Sigma Design, in context.  Just as the traditional design schools always expect you to get some practical industry experience, so must DFSS and this doesn’t lend itself well to mass training.  You can then extrapolate that knowledge knowing where the pitfalls are. 
    For example, in 1997 I was working with GE Plastics during their implimentation of DFSS and they were considering the use of Berryman scorecards which are extremely useful in descrete component manufacturing (esp. electronics) but quite misleading in the chemical industry where individual “parts” are rarely independent.  Is there a theoretical basis for rejecting these?  Yes, but can you really expect the typical black belt to have that depth of understanding.  I don’t think so unless you mentor in the knowledge.
    Successful DFSS should build upon your existing DMAIC and other proven design methods (many of which are industry and even company specific).  Teach the general concepts and tools (VOC -> CTQ, QFD, FMEA, Scorecards, variation modeling, etc) then apprentice your black belts to an existing design team as both student and teacher.  You may end up with something that others outside your company don’t recognize as DFSS but it will work and at the end of the day, you are only responsible to your customers and shareholders not us.
    Michael CarverDirector of Six Sigma Union Pacific Railroad

    0
    #186605

    Deshazer
    Member

    Joe,
    In reading your comments, I would like to add that I have successfully influenced four large companies in the advancements of DFSS.
    I have worked it into the following company life cycle processes:
    NPD/I
    PCP(Product creation process)
    APQP
    PDP(Product development Process)
    The success is down to tailoring the tools to the situation to get a win win status, this has then proceeded to more buy in and greater understanding and demand.
    I would also like to add that I’m a GE certified six sigma black belt and DFSS black belt practioner for the past twelve years and have no hesitation in recommending a six sigma blackbelt trained at GDA ltd between the years of 1997 to 2002.

    0
    #186606

    Mikel
    Member

    GE? You guys are kind of slow aren’t you. 8 years to answer?You’ve influenced 4 large companies? Wow.Who is GDA? Never heard of them.

    0
    #186621

    BTDT
    Participant

    Stan:Probably General Domestic Appliances, a GE company in the UK. I hadn’t heard of them either. I found a reference back in 2000. Pretty topical reference for a 2001 post.Cheers, Alastair

    0
    #186633

    GB
    Participant

    Uh gee, Roy, thanks (?) for tooting your own horn…
    I guess?

    0
Viewing 11 posts - 1 through 11 (of 11 total)

The forum ‘General’ is closed to new topics and replies.