| |
Air Academy Associates On the Road
June 19-21, 2007
Chicago, IL
WCBF 3rd Annual DFSS Conference
Dr. Mark Kiemele, President, Air Academy Associates, will facilitate a June 19 pre-conference workshop entitled DesIgNNOVATION: the Practical and Powerful Integration of Innovation and DFSS. On June 21, during the conference track sessions, Dr. Michael Slocum, Air Academy's Chief Innovation Officer, will speak on the topic The Third Wave: Systematic Innovation (TRIZ).
Use code AirAcad15 to register and receive a 15% discount.
--------------------------------
October 23-26, 2007
Las Vegas, NV
WCBF 2nd Annual Global Six Sigma Summit
Watch for details on this event in upcoming newsletters and in the News and Events section of Air Academy's homepage.
----------------------------------------------------
Colorado Springs Open Enrollment Sessions
Master Black Belt
April 30 - May 4, 2007
Design of Experiments
May 21-25, 2007
Lean Green Belt
Week 1: June 11-15, 2007
Week 2: July 9-13, 2007
Lean Black Belt
Week 1: June 11-15, 2007
Week 2: July 9-13, 2007
Week 3: August 13-17, 2007
Week 4: September 10-14, 2007
Design for Six Sigma
June 18-22, 2007
Six Sigma Master/Train the Trainer
July 24-August 1, 2007
----------------------------------------------------
Additional Open Enrollment Sessions
(with Air Academy Instructors)
Ashton College
Green Belt Training
Vancouver, B.C.
Week 1: (tentative) June 25 week
Week 2: (tentative) July 30 week
University of Texas Accelerated Black Belt Training
Austin, TX
Week 1: June 4-8, 2007
Week 2: July 16-20, 2007
----------------------------------------------------
Other Upcoming Events
3rd Annual WCBF Lean Six Sigma Summit
April 25-26, 2007
The Westin Chicago North Shore
Chicago, IL
-----------------------------------
ISSSP Leadership Conference
May 7-10, 2007
Hyatt Regency at Gainey Ranch
Scottsdale, AZ
--------------------------------------
News Briefs
Delphi Recognized at IQPC Summit
Delphi was the big winner in the Six Sigma Excellence Awards categories at the January IQPC Six Sigma Summit in Miami. These awards are regarded as the most prestigious problem-solving awards in Six Sigma.
Out of eight categories which named winners and honorable mentions, Delphi divisions were finalists in four: Most Innovative Six Sigma Project, Best Design for Six Sigma Project, Best Defect Elimination in Manufacturing Project, and Best Lean Sigma Project.
Delphi-Thermal engineering manager advanced reliability and ICIM, Dan Farley, said, "We're extremely proud to receive this industry recognition, and I congratulate all our award winners and finalists. These projects represent all of the good work being done at each Delphi division."
Delphi Thermal and other Delphi business units have for many years used the Air Academy Associates Master Black Belt/Train the Trainer program to enhance the skill sets of their Master Black Belts.
For photos from the IQPC awards ceremony, winning project topics, and a list of project team members, click here.
Xerox Receives Quality Launcher Awards
Pictured left to right: George Maszle, Mark Kiemele, Norm Fowler
Dr. Mark Kiemele recently presented Air Academy Associates Quality Launcher Awards to George Maszle, Director and Business Support for Lean Six Sigma Deployment, and Norm Fowler, Manager, Lean Six Sigma Strategic Initiatives, both of Xerox. The custom-made walnut Statapult® Catapults with affixed gold launch balls bore the inscription "Quality Launcher of the Year."
Air Academy periodically recognizes clients who demonstrate outstanding leadership and commitment in support of their company’s pursuit of competitive excellence. This special version of the Statapult® Catapult training aid, an integral element of Air Academy’s training programs since 1993, is used as the award design as its numerous applications reflect the versatility that is also required in those managers who champion a company’s continuous improvement efforts. Please join us in congratulating George Maszle and Norm Fowler as award recipients.
Air Academy Provides Commentary
Dr. Michael Slocum, Chief Innovation Officer for Air Academy Associates, is a member of the editorial panel and provides regular commentary for two top internet information portals pertaining to TRIZ, competitive excellence, and systematic innovation.
Dr. Slocum contributes his expertise to CTQ Media's RealInnovation.com and The TRIZ Journal in a variety of forms to include discussion forums, articles, and case studies such as the following:
Integrating TRIZ Into Six Sigma
(with co-author Amir Kermani)
What Is TRIZ?
(with co-authors Katie Barry and Ellen Domb)
To access these information sources, participate in discussion forums, or to sign up for associated newsletters, click on the links above.
SSPG Website Launched
The website for Air Academy’s sister company, Six Sigma Products Group, Inc., was launched in mid-January 2007. Information regarding Air Academy’s books, software, and training aids is currently available on both the Six Sigma Products Group website and the Air Academy site. Product purchasers reviewing the Air Academy site should note that the order form links to an SSPG URL as product orders, payments, and shipping are handled by the SSPG Customer Service team.
The Six Sigma Products Group site includes several non-Air Academy continuous improvement/competitive excellence/TRIZ texts which are not reflected under the Products section of the Air Academy site. In the near future SSPG will be adding to its product line downloadable versions of several of the AAA/Digital Computations software packages and the extended download capability. Full e-commerce capability for the purchase of other products is also scheduled for the second quarter of 2007 though SSPG will continue to accept orders by fax, phone, and email as a convenience for our customers.
|
|
|
|
| | |
 Stagnation from Champion Apathy
by Dr. Mark Kiemele
Much has been said and written about why some deployments and implementations of Lean Six Sigma (LSS) are more successful than others. For any activity whose success will be evaluated, there are many factors that impact that appraisal. In LSS, one such factor is the ability, dedication, and passion of a LSS Champion to engage in LSS. For the purposes of this article, the term “champion” encompasses deployment champions, project champions or sponsors, and process owners.
In order to champion anything, people must first know what it is they are championing. It is easy for a LSS Champion to compartmentalize LSS by classifying it as a methodology or set of tools. This limited perspective of LSS will bound the effectiveness of a champion. LSS is indeed a methodology and a set of tools, but it is much, much more than that. It is also a business strategy, a philosophy, and even more, a “thought process.” That thought process must be transformed into actionable decision making on a daily basis. When we see a LSS champion appropriately using the tools independent of the DMAIC methodology to help them make better decisions, we know they have “arrived.” Unfortunately, many champions don’t know enough about the tools to even begin to use them in an independent fashion. Clearly, LSS is not only about the tools, but champions need to know enough about some of the tools in order to function effectively. The use of the tools is the means by which the thought process is transferred into action.
Inherent in the LSS thought process is the necessity to comprehend the difference between random variation and special cause variation. Many champions don’t even know what that means, and even if they did, many would not know how it applies to them. How, then, are we to expect champions, a.k.a. leaders and managers, to take action when it is appropriate to do so and not to take action if it is not warranted. The errors of over-reaction and under-reaction are much too commonplace in the deployment and implementation of LSS, and it is the champions’ responsibility to minimize these kinds of errors. But how can that be done? The first step is to educate, train, and allow them to practice, but that step alone will still be insufficient. The individual champion must be internally motivated to become a better decision maker, and we claim that this aspect is missing in much of what we call LSS champion training today.
Champion training must be much more than learning how to manage the bureaucracy of LSS. It must be motivational, giving the prospective champions something they can use immediately to demonstrate their commitment to and knowledge of LSS. These “ready-to-use” techniques must include the first line of defense against variation, questions that Champions need to answer and other questions they need to be asking, prioritization techniques, decision making vehicles like Pugh Concept Selection, quick tests for detecting significant shifts in average and standard deviation, and most certainly the understanding and application of process behavior charts (a.k.a. control charts). Despite being maligned, misunderstood, and misused, the control chart is still the best vehicle we have for knowing when to take action on a process and when not to.
Project selection, supposedly the “Achilles heel” of LSS, must also be addressed in champion training. Unfortunately, we have found the project selection issue to be a convenient mask for many champions’ lack of knowledge about process and people. Any good leader or manager should be able to spew out 10 potential improvement areas in his or her area of responsibility within just a couple of minutes. If he can’t, I have to wonder about what this person actually does in his position as a leader or manager and/or if he is oblivious to what is happening around him. Mining projects should not be difficult, but the selection and scoping of projects can be. Prioritization techniques (once again, tools) can be used to help select and scope projects, but the experiential factor is important here, too. There is no substitute for knowledge about people and process, which is a key ingredient for the success of any champion.
We have found that, by giving champions something they find useful and that they can apply easily and immediately, their motivational needle is bumped. They become more excited about the prospect of LSS actually helping them do their jobs more effectively and efficiently, rather than thinking this is just another flavor-of-the-month initiative imposed upon them from above. Dedication and passion are key ingredients to a champion’s and, consequently, an organization’s success in LSS and must be addressed and ignited during champion training.
If generating intrinsic motivation is not a primary objective of your champion training, then you can expect a bumpy road on your LSS journey, one wrought with repeated aborts and restarts. Take action now by injecting the proper ingredients into your champion training. Get people excited. Show how the training applies to them and get them moving and using the tools on a daily basis, and you will see an infusion of energy into your LSS initiative. It is extremely important for champions to walk the talk, because many others, including the belts, are watching and evaluating them all the time. The effect of a motivated and passionate champion can be contagious. Having both the will and the skill is a winning combination, and we need to work on both. After all, turning the word “champion” from a noun into a verb is what it is all about.
 Comparing AAA Software to Other Statistical Packages
by Dr. Stephen Hoyle
A very common question that we hear is: Why do we prefer Air Academy Six Sigma software to other statistical packages? Our answer is that it supports the Air Academy Keep It Simple Statistically teaching philosophy the best and is most beneficial to our clients.
Students spend a lot of time learning how to use features in many statistical packages that will never be used in practice and which initially take valuable time to learn. The AAA software has user-friendly drill down menu choices so that the user can focus on what is needed and what will actually be used. Our software is very easy to learn for those new to Six Sigma and easy to remember how to use. Our in-class time is spent on learning Six Sigma tools and techniques and not software. As the AAA software packages are Excel add-ins, most people immediately feel comfortable with the software. Also, the results are in normal Excel files that can be viewed and modified by anyone with Excel. (AAA software is not needed for this type of file review.)
A less common question is: Do numerical results differ between AAA software and other packages? The short answer is no. For example, there is only one formula to calculate a standard deviation or to build a least squares regression fit. The main differences between AAA's software and other software are in the selection of included features and ease of use. There is an explanation for any apparent numerical difference.
As an example, there was a question recently about why DOE Pro and JMP gave (apparently) different answers to a full factorial model with one three-level quantitative input factor and one three-level qualitative input factor. Again, the short answer is that the numerical answers from AAA and JMP are both correct. However, they are performing different calculations and present the answers differently. For someone experienced in only one of these packages, the results of the other package may look odd. Here is a short explanation of some critical differences.
JMP builds an ANOVA table and then a “model” that includes only the main terms and the two-way interaction term. This means that it does not have to convert the qualitative term into numbers and that the model values for these terms are not affected by the order of the three settings in the qualitative factor. JMP also only presents the uncoded values – which means the confounding pattern between the terms is not clear. (If the software presented the underlying orthogonal design, the user would understand the effect of each term independently.) The R-squared value is based only on the included terms.
DOE Pro (or DOE KISS) builds a regression model that includes the main effects, the two-way interaction, and the higher order terms available with this full factorial model. This means that it does have to convert the qualitative term into numbers and the model values of the terms are affected by the order that is chosen for the qualitative factor. However, the model is correct regardless of the order chosen for the three settings of the qualitative input factor. DOE Pro presents coded values – which means the effect of each term can be determined independently. Gaining this type of knowledge about the process is central to AAA philosophy. (DOE Pro will also present the uncoded regression model.) The R-squared value for any model with all available terms, regardless of the chosen order for the qualitative term, is exactly the same. (The R-squared values will start to differ slightly for the six possible orders of the qualitative factor, as terms are removed.) Both JMP and DOE Pro get the identical SS Total and SS Pure Error values, regardless of the order of the qualitative factor. The SS and df for Regression and Lack of Fit differ because JMP is doing ANOVA and DOE Pro is building a regression model.
Product News - SPC XL 2007 Software Changes
 SPC XL 2007, currently in beta, contains the following new features:
- Excel 2007 Ribbon Interface
- Support for Unified Menus
- Addition of a Paired t Test
- The independence test output has been reformatted to
be easier to read
- t Test, paired t Test, and F test include summary stats
- Pareto chart can work with percentage data
- ANOVA MSA calculations for one operator or one
replicate
This SPC XL update is compatible with Excel 2000, 2002, 2003 (XP), and 2007. In order to run SPC XL on Microsoft Excel 2007, you will need to download and install SPC XL 2007. This version of the software, still in beta, is available at no cost to those who already hold valid licenses for SPC XL. This link will take you to a web page containing information on how to obtain SPC XL 2007 beta. Further questions regarding obtaining the beta version may be addressed to Air Academy Associates or Six Sigma Products Group Customer Service function.
 Questions Corner
Question: A participant in a Six Sigma class claims he had run the distribution and that 6-sigma was 4.5 defects. Does someone have the explanation for how 3.4 was calculated?
Answer: I'm thinking that perhaps the misconception comes from a z value of 4.5 rather than dpm of 4.5. Recall, that for a sigma level of 6, this is equivalent to looking up a z value of 6 in a standard normal table. This produces a tail area (dpm) of something like 2 defects per billion! So people get confused and wonder where the 3.4 dpm comes from, if a z value of 6 means 2 defects per billion. Enter the 1.5 sigma shift! Recall that Motorola found that over time, the process mean tends to shift and drift due to "noise". They found that this shift tends to be around1.5 sigma. So, if you take a z value of 6 and subtract 1.5 (due to moving the mean 1.5 sigma either right or left), you're left with a z value of 4.5. Now, when you go into a standard normal table and look up a z value of 4.5, you get a tail area of .0000034 or 3.4 dpm. This shift in the mean is equivalent to the difference between short term and long term variation. Motorola used shifting the mean of a short term distribution as an approximation to what really happens in the long term.
There are web pages, such as this link located on CTQ Media's iSixSigma site, which discuss the 1.5 sigma shift.
BOTTOM LINE: Whether you like the idea of short term or long term, and the whole sigma shift, industry in general uses the Motorola "Richter scale" for quality which says that 3.4 dpm (long term) is equivalent to a 6 sigma capable process. If you have short term data, then technically you need to subtract 1.5 from the sigma capability numbers found in a standard conversion table, because these numbers already have the 1.5 sigma shift built in.
BOTTOM, BOTTOM LINE: The FOCUS is and must be on improvement. Arguing about what scale to use to measure "sigma capability", the shift, and all the rest of that just subtracts from that focus. That's why we prefer to use dpm, Cp, and Cpk and just forget about sigma capability altogether (unless we're trying to benchmark products or processes with dissimilar complexity). We need to understand our ability to meet the customer requirements (in dpm or % of time we don't meet a requirement) and then work to improve it by reducing non value added steps, variation, defects, etc. Whether we achieve 6 or not is not the true focus. If we're at 200,000 dpm today and can get to 50,000 dpm, will our customers see and feel the difference? Will we internally feel the difference in terms of reduced cycle times, etc? We think so. We should measure and track dpm over time and work to constantly reduce defects.
Meet the Staff
 Sue Darby, Vice President, Marketing and Advertising, joined Air Academy Associates in January, 1995. After serving in the Business Development role for 10 years, she is now engaged in the branding and marketing of Air Academy to include electronic and magazine advertising and management of Air Academy's website.
In addition to being a member of the Air Academy Strategic Marketing Committee, Sue is also president of Six Sigma Products Group, Air Academy's sister company and product fulfillment group. Sue may be reached at sdarby@airacad.com.
|
|
|
|
|