Jump to content

Six Sigma

From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by 67.177.33.6 (talk) at 21:22, 10 January 2006 (→‎Of the term: Six standard deviation). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

You must add a |reason= parameter to this Cleanup template – replace it with {{Cleanup|January 2006|reason=<Fill reason here>}}, or remove the Cleanup template.

For the band, see Sigma 6. For the toyline and animation series, see G.I. Joe: Sigma 6.

Six Sigma is a quality management program that measures and improves a company's operational performance of by identifying and correcting defects in its processes and products.

Originally, Six Sigma was defined as a process variation that would produce no more than 3.4 defects per million parts (or "opportunities"). Today, however, Six Sigma is applied to produce a product that satisfies the customer and minimizes supplier losses to the point at which it is not cost effective to pursue a higher quality.

Six Sigma was pioneered at Motorola in the mid-1980s by Bob Galvin, who succeeded his father, Motorola founder Paul Galvin, as head of the company, and by Motorola engineer Bill Smith. While Motorola still maintains the trademark, the methods associated with the term were picked up and followed by other large companies such as AlliedSignal (now known as Honeywell) and General Electric, which ultimately popularized the process. It has since spread to other large companies, including Ford, Caterpillar, Microsoft, Raytheon, Quest Diagnostics, Seagate Technology, Siemens and many more.

Although Six Sigma is usually applied to manufacturing companies, it can be applied wherever the control of variation is desired. In recent years, it has begun to branch out into the service industry, and in 2000, Fort Wayne, Indiana became the first city to implement the program in a city government. Some, claiming that Six Sigma's impact has not yet been fully realized, advocate an open source approach so that the principles of Six Sigma might be more widely adopted.

Basic methodologies

DMAIC

Basic methodology to improve existing processes

  • Define Formally define the goals of the design activity. What is being designed? Why? Use QFD or Analytic Hierarchical Process to assure that the goals are consistent with customer demands and enterprise strategy.
  • Measure to define baseline measurements on current process for future comparison
  • Analyze to verify relationship and causality of factors. What is the relationship? Are there other factors that have not been considered?
  • Improve optimize the process based upon the results of your analysis
  • Control continuously measure the process and institute control mechanisms to ensure that variances are corrected before they result in defects

DMADV

Basic methodology to develop new, customer-focused processes. Also see Design for Six Sigma quality.

  • Define Formally define the goals of the design activity. What is being designed? Why? Use QFD or Analytic Hierarchical Process to assure that the goals are consistent with customer demands and enterprise strategy..
  • Measure Determine Critical to Stakeholder metrics. Translate customer requirements into project goals..
  • Analyze to find and prove relationship between potential root causes and its effects (y=f(x)).
  • Design the process to meet customer needs.
  • Verify the design performance and ability to meet customer needs.

Six Sigma Training

There are two levels of training in the Six Sigma quality system, Black Belts and Green Belts.

Six Sigma Black Belts are basically the on-site Six Sigma implementation experts who will develop, coach and lead cross-functional teams, mentor and advise management on prioritising, planning and launching Six Sigma projects. In short, they are the ones who will be directly responsible for the execution of projects in a Six Sigma organization. They are expected to take on projects with projected savings of US$250K.

Six Sigma Green Belts are employees throughout the organization who execute Six Sigma as part of their overall jobs. They have less Six Sigma responsibility and their energies are focused on projects that tie directly to their day-to-day work. Green Belts have two primary tasks: first, to help deploy the success of Six Sigma techniques, and second, to lead small-scale improvement projects within their respective areas. Green Belts can do much of the legwork in gathering data and executing experiments in support of a Black Belt project.

There are other, equally valid definitions of Green Belt and Black Belt. In many organizations, all project leaders keep their existing jobs, and lead projects to improve the processes for which they are already responsible. Black Belts lead projects which are likely to require designed experiments. Green Belts lead projects which do not. Frequently, but not always, Green Belt projects focus on business processes. People who work on either type of team are simply referred to as team members. This structure avoids some fairly serious problems associated with the first listed structure, and produces excellent results.

One major manufacturing company found $2 in business process savings for each $1 they found in design or manufacturing. Their largest savings came from a business process project, led by a Green Belt, which provided $17,000,000 in first-year savings.

Master Black Belts should have a broad and deep understanding of Six Sigma principles. Usually, their tasks are to lead the program, teach the classes, and mentor the Black Belts and Green Belts as they work on their projects. Sometimes, a Master Black Belt will organize and coordinate a cluster of related projects.

Tools used in Six Sigma Projects

  • Failure Modes Effects Analysis
  • Cost Benefit Analysis
  • Customer Output Process Input Supplier Maps
  • Process Maps
  • Run Charts
  • ANOVA Gage R&R
  • Cause & Effects Diagram (aka Fishbone or Ishikawa Diagram)
  • Homogenity of Variance
  • ANOVA
  • Moods Median
  • Chi-Square Test of Independence and Fits
  • General Linear Model
  • Regression
  • Correlation
  • Design of Experiments
  • Taguchi
  • Control Charts

Six Sigma Successes

There have been many success stories with Six Sigma. This could be one reason why many companies and some governments have adopted the concepts. So while anyone can say anything they want about it, it seems to work for some unknown reason.

Examples of Fort Wayne Indiana’s Six Sigma trained black belt program: (Reference: http://www.brookings.edu/metro/speeches/20050210_wingspread.pdf)

  • Water main costs were reduced from $61.00 per foot to $50.00 per foot.
  • Numbers of days to get Improvement Location Permits were reduced from 51 days to 12 days.
  • 50% reduction for transportation engineering projects change orders.
  • Numbers of fire code re-inspections were increased by 23% while reducing the number of days for re-inspection from 51 to 34 days.
  • Missed trash pick-ups reduced by 50%
  • Response for pothole complaints reduced from 21 to 3 hours.

Even Real Estate investment firms have seen a noted improvement by implementing Six Sigma theories that have a reported savings range from $250,000 to $450,000 … http://www.us.am.joneslanglasalle.com/NR/rdonlyres/349C99EC-90B4-4857-8636-CC9B6C866EC7/6465/SixSigmaInsert.pdf

In Engineering and Construction, Bechtel, on the Channel Tunnel Rail Link project in the UK, the project team uncovered a way to save hundreds of job hours on one of the tunnelling jobs. http://www.bechtel.com/sixsigma.htm

In healthcare, North Carolina Baptist Hospital says, "The Six Sigma process improvement deployment at North Carolina Baptist Hospital is starting to show the kind of results that convert skeptics to believers." and "A Six Sigma process improvement team charged with getting heart attack patients from the Emergency Department into the cardiac catheterization lab for treatment faster slashed 41 minutes off the hospital’s mean time" http://www1.wfubmc.edu/articles/Six+Sigma

Dow reduced severe ergonomics injuries by 90% http://www.osha.gov/SLTC/ergonomics/dow_casestudy.html

Six Sigma-Based Methodology A Motorola/3M Case Study http://www.future-fab.com/documents.asp?d_ID=2308

According to the The Institute of Quality Assurance … Reference: http://www.iqa.org/publication/c4-1-38.shtml

  • ‘Wipro reports successes in its first year. "First of all, we now have a common language across our divisions." … "Defects are steadily falling in cylinder manufacturing,"
  • "Although Motorola has made huge reductions in defect rates, it has not yet achieved Six Sigma overall. Motorola now considers itself a 5.7 sigma company. While Six Sigma is a noble goal, the rate of improvement is what is important. It has saved Motorola billions of dollars in costs (in terms of scrap and re-work)"
  • "Six sigma was appealing because it is pretty straightforward," says James Bailey, executive vice president and corporate quality officer for Citibank.

These are but a few documented examples of where Six Sigma worked, and how well it actually worked.

Properly applied, Six Sigma not only works, it routinely works splendidly, across a broad range of problems. It is common to see 90% improvements in cycle times, scrap rates, etc. It is uncommon to see improvements less than 50%.

The power of Six Sigma is not in the tools. The power of Six Sigma is in the problem solving approach, and in the commitment to constant improvement.

Criticisms of Six Sigma

Of its origin

While the successes enjoyed by the named companies are strong arguments for the adoption of Six Sigma, and Robert Galvin and Bill Smith deserve recognition, a number of people in the Quality community would argue that the methodologies and results have been available since the 1920s and were developed by luminaries like Shewhart, Deming, Juran, Ishikawa, Ohno, Shingo, Taguchi and Shainin, and that Galvin and Smith merely had the good sense to adopt the techniques.

The use of "Black Belts" as itinerant change agents is controversial as it has created a cottage industry of training and certification which arguably relieves management of accountability for change; "Old School" implementations simply used the technical talent at hand — Design, Manufacturing and Quality Engineers, Toolmakers, Maintenance and Production workers — to optimize the processes.

Of the term: Six standard deviation

According to the graph of the standard normal distribution, only two billionths of the normal curve falls beyond six standard deviations, in contrast to the value of 3.4 millionths publicized by Six Sigma promoters. In the real world, processes do not follow normal distributions, making much of the Six Sigma theory unrealistic. Confusingly, that value corresponds to precision within 4.5 standard deviations, reflecting an allowance for a 1.5 standard deviation "drift" in the manufacturing or service process mean value. Introduced by Mikey Harry around 1980, its magnitude was based on an error in applying a theory on tolerancing to continuous processes.

The +/-1.5 sigma drift implies that if we are manufacturing a product that is 100 +/- 3 cm (97 - 103cm), over time, it may drift down to 98.5 — 104.5 or up to 104.5-101.5. This is something that might be of concern to customers. So where does the "+/-1.5" come from?

The +/-1.5 shift was introduced by Mikel Harry. Where did he get it? Harry refers to a paper written in 1975 by Evans, "Statistical Tolerancing: The State of the Art. Part 3. Shifts and Drifts". The paper is about tolerancing. That is how the overall error in an assembly is affected by the errors in components. Evans refers to a paper by Bender in 1962, "Benderizing Tolerances — A Simple Practical Probability Method for Handling Tolerances for Limit Stack Ups". He looked at the classical situation with a stack of disks and how the overall error in the size of the stack, relates to errors in the individual disks. Based on "probability, approximations and experience", he suggests:


What has this got to do with monitoring the myriad processes that people are concerned about? Very little. Harry then takes things a step further. Imagine a process where 5 samples are taken every half hour and plotted on a control chart. Harry considered the "instantaneous" initial 5 samples as being "short term" (Harry’s n=5) and the samples throughout the day as being "long term" (Harry’s g=50 points). Because of random variation in the first 5 points, the mean of the initial sample is different to the overall mean. Harry derived a relationship between the short term and long term capability, using the equation above, to produce a capability shift or "Z shift" of 1.5 ! Over time, the original meaning of "short term" and "long term" has been changed to result in "long term" drifting means.

Harry has clung tenaciously to the "1.5" but over the years, its derivation has been modified. In a recent note from Harry "We employed the value of 1.5 since no other empirical information was available at the time of reporting." In other words, 1.5 has now become an empirical rather than theoretical value. A further softening from Harry: "… the 1.5 constant would not be needed as an approximation".

Despite this, industry has fixed on the idea that it is impossible to keep processes on target. No matter what is done, process means will drift by +/-1.5 sigma. In other words, suppose a process has a target value of 10.0, and control limits work out to be, say, 13.0 and 7.0. "Long term" the mean will drift to 11.5 (or 8.5), with control limits changing to 14.5 and 8.5. This is nonsense.

The simple truth is that any process where the mean changes by 1.5 sigma, or any other amount, is not in statistical control. Such a change can often be detected by a trend on a control chart. A process that is not in control is not predictable. It may begin to produce defects, no matter where specification limits have been set.

World Class Quality means "On target with minimum variation" .

The 1.5 sigma shift assumption has many critics. Donald J. Wheeler, a respected quality professional, labels it "goofy".

The other common objection is that the choice of a shift of 1.5 sigmas is too arbitrary and probably inaccurate. Some suggest that the 1.5 sigma shift was implemented for marketing reasons, so that the program could be named Six Sigma instead of "4.5 Sigma" without setting the unrealistic goal of two defects per billion. However, according to original training material used at Motorola in 1985, the point at which a shift became detectable with a sample size of 4 was 1.5 standard deviations, suggesting that the number was not arbitrarily selected.

In practice, the principle of six standard deviations of quality between the upper and lower specification limits is often not applied with mathematical rigor. Instead, Six Sigma is seen as a methodology or mindset with the goal of minimizing defects. It is used in this way in non-manufacturing environments, where it serves as an analogy to manufacturing processes and is not used for statistical distributions. Similarly, the frequent misuse of the 1.5 shift assumption in manufacturing processes is a reflection of a similar attitude in industrial applications as well.

While the foregoing is quite eloquent, and mostly correct, it has at least two serious errors: 1) An Xbar and R chart will indeed detect a 1.5 sigma shift with n=4. Therefore, what? In what way does that support the 1.5 sigma shift? It does not. It is nothing more than mere coincidence, and is irrelevant. n=2, 3, 5, 6, and so on are equally valid, and give entirely different results. 2) Not only is the magnitude of the shift completely arbitrary, the sign is reversed. The convention is to call a true 4.5 sigma process a 6 sigma process, i.e., to take an unearned 1.5 sigmas of "credit." If you are really trying to account for the fact that long-term variation is greater than short term variation, you must do the opposite, i.e., my short term data indicate that I have a 4.5 sigma process, so, in the long term, I can really expect 3 sigma performance.

Of statistics

Six Sigma is controversial with the statistics profession. Some teachers of statistics are critical of the standard of statistical teaching found in Six Sigma materials. Others object to the idea that a single universal standard can be appropriate across all domains of application. They argue that quality standards should be set on a case-by-case basis using decision theory or cost-benefit analysis.

There are, indeed, things commonly taught in Six Sigma that are flat wrong: 1) The whole 1.5 sigma shift theory is, to the best of my knowledge, almost completely unaccepted by the best sources in statistics. 2) The notion that you can successfully estimate defects down into the tens or hundreds of parts per million, based on a sample of several dozen, is mathematically unsupportable.

In addition, there are things that are taught that are annoying, such as clinging to the outdated model of "attribute" and "variable" data, rather than the much more widely accepted "nominal", "ordinal", "interval", and "ratio" model. There is also the problem that the widely used Capability Study drags an alarmingly high level of uncertainty into its calculations, and is often given credit for more than it can usually do.

The quality of teaching is very uneven. Some is outstanding. Some is very poor. Of course, you can raise those same objections in almost any field. The requirement to have a PhD to teach at a university level certainly has not solved the same problem there.

Still, if you do Six Sigma right, it routinely works again and again. Having coached many hundreds of projects, I have yet to see a case where Six Sigma properly applied produced anything but excellent results, in fields as diverse as design, manufacturing, marketing, sales, HR, finance, and legal departments.

Of methods

Others suggest that Six Sigma, rather than being a true methodology, is more often implemented to start an unending cycle of improvement and use of better tools on the industry day to day practices rather than to use advanced statistical theories that cannot be daily applied.