Has anyone had an encounter with this "process improvement" program? I would like to hear your stories. http://www.unitedmedia.com/comics/dilbert/...70146612119.gif Thanks.
<div class='quotetop'>QUOTE(Alric @ Jan 19 2007, 09:22 AM) [snapback]377670[/snapback]</div> It seems to be largely forgotten management fad now. Was a big deal at my company around 10 years ago. Lots of meetings, required education sessions, process manuals, etc. etc., your basic corporate BS.
six-sigma is actually a pretty good concept... while i haven't had any personal experience with it, we did study it in one of my masters classes last year. Implementations of it seem to vary from location to location, with some being better than others. the real key to success is to take it to heart, and not just giving it lip service. And for what it's worth, the industry I'm in more or less expects that level of defects or better, and the FDA is right there to ensure we don't kill too many people...
<div class='quotetop'>QUOTE(Alric @ Jan 19 2007, 09:22 AM) [snapback]377670[/snapback]</div> http://www.businessweek.com/magazine/conte...49/b4012069.htm http://www.businessweek.com/magazine/conte...htm?chan=search No direct experience, only what BusinessWeek says about it: It may surprise those who have come to know Motorola (MOT ) for its cool cell phones, but the company's more lasting contribution to the world is something decidedly more wonkish: the quality-improvement process called Six Sigma. In 1986 an engineer named Bill Smith, who has since died, sold then-Chief Executive Robert Galvin on a plan to strive for error-free products 99.9997% of the time. By Six Sigma's 20th anniversary, the exacting, metrics-driven process has become corporate gospel, infiltrating functions from human resources to marketing, and industries from manufacturing to financial services. It's still paying off for Motorola Inc., too. Consider the hot-selling, super-slim Razr phone. A creative, innovative design, sure. Yet "Six Sigma's stamp is all over the Razr," says Michael S. Potosky, Motorola's corporate director of Six Sigma. Engineers, for instance, applied the process to the phone's antenna, helping keep it hidden while maintaining call clarity. With hits like the Razr, the Schaumburg (Ill.) company has climbed from a 15.4% market share in mobile phones to 22.4% over the past two years. Motorola netted $4.6 billion on $36.8 billion in revenues last year, helping it earn the No. 13 spot on the BusinessWeek 50 list of top corporate performers. The "sigma" refers to the Greek letter, which in statistics is used to measure how far something deviates from perfection. The "six" comes from the goal to be no more than six standard deviations away from that perfect measure.
Interesting. I didn't know 6-sigma came from Moto. I wouldn't say Motorola's cell phones are that hot (anymore). Sony-Ericsson, Nokia, and even Samsung are producing much hotter products. Moto is falling further and further behind.
<div class='quotetop'>QUOTE(eagle33199 @ Jan 19 2007, 09:32 AM) [snapback]377677[/snapback]</div> It is a great concept. The problem comes in when management decides it is the solution to every problem, or that every "process" can be improved by having a six sigma manual written for it.
<div class='quotetop'>QUOTE(Beryl Octet @ Jan 19 2007, 10:44 AM) [snapback]377683[/snapback]</div> Or that it can apply to processes other than manufacturing. I think it applies more to automated robotic systems like those at Toyota. Where Six Sigma really comes from.
<div class='quotetop'>QUOTE(Alric @ Jan 19 2007, 10:05 AM) [snapback]377688[/snapback]</div> Exactly what I was getting at.
Don't know anything about six-sigma, but anything that looks at a process and uses feedback to control the process can't be a bad thing. Keeps nuke plants from blowing up. Might be wrong, but if your IQ is 3-sigma, you are considered either "gifted" or "special." Seems like 6-sigma is reasonable goal to me if you are trying to make something (a person, cell phone, or process) "world class." I also wonder that if something is really going to be measured 6-sigma, how many samples do you need to statistically declare six-sigma. (not a stats guy either.) If you are looking at a business process with only a handful of samples, seems like the statistics won't work out for you, you need something with a bigger N to sample. I'm hypothesizing that some of the 6-sigma efforts gone awry are caused by people who don't appreciate the mathematics and try to apply on problems too small. Just wild hunches, maybe I should learn more about it....
<div class='quotetop'>QUOTE(bigdaddy @ Jan 19 2007, 10:40 AM) [snapback]377702[/snapback]</div> You know your stats, but basically the goal of the six sigma as used here is to have a single defect be a six sigma occurrence. In other words, to be six sigma compliant, if you are making a million widgets, then you can only have 3.4 defects.
<div class='quotetop'>QUOTE(Beryl Octet @ Jan 19 2007, 10:48 AM) [snapback]377704[/snapback]</div> That's what I was thinking. That sounds world-class to me. So, does six-sigma have a feedback component to it as well? I was under the impression that the measurements somehow fed back into the system, like a process controller.
<div class='quotetop'>QUOTE(bigdaddy @ Jan 19 2007, 10:56 AM) [snapback]377709[/snapback]</div> Almost certainly. Probably tons of web sites on it. It left a sour taste in my mouth, since it was basically being applied to things that couldn't possibly be six-stigmatized.
<div class='quotetop'>QUOTE(Beryl Octet @ Jan 19 2007, 12:07 PM) [snapback]377714[/snapback]</div> Yes. How can you search for less than a six sigma deviation, when the components of the system vary by more than six sigma over time, like people. You can only do it when your components do not vary over time. Cheers!
<div class='quotetop'>QUOTE(bigdaddy @ Jan 19 2007, 09:56 AM) [snapback]377709[/snapback]</div> The problem is, six-sigma is basically an idea: drive your processes to reach a goal in the number of defects. Once you reach that point, you're considered "six sigma compliant". however, the implementation to get there is rather involved and differs in every location I've read about. The general overall themes, however, were fairly consistent. The things you can do to improve quality are rather limited: fix your processes and fix your design. organizations have turned to automation to help improve the consistency and reliability of processes (since, after all, a machine is less error prone than a person in doing a simple repetitive task) and DFMA (design for manufacturing and assembly) to improve their design. Thought about simply, if you can have a machine do 90% of the work, and design it so that the work the person has to do is mind-numbingly simple (insert one screw here and you're done!) they won't make a mistake... on the other hand, if they are responsible for something more complicated (for example, putting together a bookshelf or a workbench for parts), the odds are greater that they'll screw up and forget a screw. But to answer your question, six sigma itself doesn't have a feedback component (at least, not really). It's a goal, a milestone, something you can claim. but it in and of itself is not a process. The process you use to achieve six sigma will most likely have feedback incorporated into it.
Don't use Motorola as a perfect example. Sure they made tons of $ of the Razr. But look at their losses this year! Im a Six Sigma Green Belt. Most of you have it down. It's more a way of thinking and approaching process, than anything else.
<div class='quotetop'>QUOTE(Screwdriver @ Jan 22 2007, 02:11 PM) [snapback]379070[/snapback]</div> As Slim would say, what in the Wide World of Sports is a 6 sigma green belt? Are there different color belts?
I use Six Sigma at work. Currently, I am managing four different projects using Six Sigma strategies. Here's the neat part (as far as I'm concerned): I don't manufacture anything nor does anyone in my department. I work in Corporate Education. It's true that the goal is to reduce the number of defects, but that's a tricky word. I choose a similar word: deviations. For example, right now, I'm heading a project to reduce the amount of money we spend on lunch for our internal students, measured in cost-per-student. Through research, we identified an achievable target while tracking the current amount we spend per student. Charts were made with boundaries. The goal is to stay within those boundaries. When one of our classes varies from those boundaries (in this case, too low is not a bad thing) we investigate the cause and provide feedback to the offending instructor. Once we prove that we can maintain the target cost without compromising the quality of the lunches, we will look to see if we can lower it again. It's an ongoing process. I am also managing the cost of materials per course, the cost of producing our catalogs and cost of shipping. Those are not "traditional" Six Sigma applications, but by simply applying the system, it works.