As was foretold, we've added advertisements to the forums! If you have questions, or if you encounter any bugs, please visit this thread: https://forums.penny-arcade.com/discussion/240191/forum-advertisement-faq-and-reports-thread/

Six Sigma and other business models

BlackDoveBlackDove Registered User regular
edited October 2009 in Debate and/or Discourse
Okay, one, English is not my primary language, and two, I am downright horrible when it comes to mathematics unless you draw it for me like I'm a two year-old.
Wikipedia wrote:
The term "six sigma process" comes from the notion that if one has six standard deviations between the process mean and the nearest specification limit, as shown in the graph, practically no items will fail to meet specifications.[8] This is based on the calculation method employed in process capability studies.

Capability studies measure the number of standard deviations between the process mean and the nearest specification limit in sigma units. As process standard deviation goes up, or the mean of the process moves away from the center of the tolerance, fewer standard deviations will fit between the mean and the nearest specification limit, decreasing the sigma number and increasing the likelihood of items outside specification.[8]

what

I didn't study Chinese.

Can someone do me a favor and dumb that down for me? Apples and oranges preferably included in the explanation.

Also, which business models or management strategies do you like/think are superior?

BlackDove on
«1

Posts

  • TheMarshalTheMarshal Registered User regular
    edited October 2009
    I never actually read any Six-Sigma stuff, although when I was interning at Boeing I kept hearing the name thrown about. The only business-management books I've read are The Goal and Critical Chain, both written by a guy named Eli Goldratt. I found them both to be pretty interesting.

    The Goal speaks about the "Theory of Constraints", and how attempting to achieve local optimums within the scope of a production pipeline actually leads to a terribly inefficient system overall.

    Critical Chain looks at things more from a project-management perspective, and talks about how to best schedule dependent tasks. It looks at the phenomena that people automatically give themselves a "buffer" when estimating how long it will take them to complete a task, and how to organize tasks and deadlines so as to best deal with that.

    TheMarshal on
  • DocDoc Registered User, ClubPA regular
    edited October 2009
  • GoumindongGoumindong Registered User regular
    edited October 2009
    Basically it means this.

    If there are 6 standard deviations difference[see definition above] from "what you do" to "what you need" you will never have a product that fails to meet "what you need"

    What its really saying is "do everything much better than you need to and even random chance will not create situations where you have done something that doesn't do what you needed it to"

    E.G. lets say we were building a rocket to the moon. And we needed to land within 100 feet of some target. If we had an average deviation from the target of 50 feet and a standard deviation of 25 feet, we do not meet the requirements to say we're going to land in the circle. Such, we need to get our standard deviation down to 8.3 feet before we can reasonably say "we won't land outside of the circle"

    Goumindong on
    wbBv3fj.png
  • SaammielSaammiel Registered User regular
    edited October 2009
    Six Sigma is a methodology aimed at refining processes until there are less than 3.4 defects per million opportunities using the wonderful power of statistical analysis. It does this through by running through the DMAIC cycle, which is a set of 5 distinct project steps that are used to try and wring out ineffeciencies and errors in a process. So basically you say 'hey, x sucks we should try to fix it' and then you run x through the DMAIC cycle (stands for Define, Measure, Analyze, Improve, Control IIRC). Each step has its own tools and goal and by the end X is supposed to suck less.

    GE and Motorola originated most of this back in the olden days of the mid 80's. Then in the 90's every god damn large company on earth decided they would 'do Six Sigma'. Cue a giant clusterfuck of ridiculousness that most people have come to hate. So the core ideas are ok, if not all that radical and a bit arbitrary. The implementation on the other hand is generally god awful.

    Saammiel on
  • YarYar Registered User regular
    edited October 2009
    I'm not a "six-sigma black belt," as they are called, but I've been in business for a while, and as I understand it, six-sigma is just a state in which you are practically error-free in your business operations, because you've added process after process after process to account for every mishap and exception and possibility until you're basically in a situation where like 99.99999% of what happens will be accounted for and handled in a manner that is within your defined goals.

    Yar on
  • BlackDoveBlackDove Registered User regular
    edited October 2009
    TheMarshal wrote: »
    I never actually read any Six-Sigma stuff, although when I was interning at Boeing I kept hearing the name thrown about. The only business-management books I've read are The Goal and Critical Chain, both written by a guy named Eli Goldratt. I found them both to be pretty interesting.

    The Goal speaks about the "Theory of Constraints", and how attempting to achieve local optimums within the scope of a production pipeline actually leads to a terribly inefficient system overall.

    Critical Chain looks at things more from a project-management perspective, and talks about how to best schedule dependent tasks. It looks at the phenomena that people automatically give themselves a "buffer" when estimating how long it will take them to complete a task, and how to organize tasks and deadlines so as to best deal with that.

    Okay I think I get it. Tolerances for various deadlines, and the more accurate the optimal production, the more inefficient (and I assume, more expensive) the system is.

    Neat.
    Doc wrote: »

    Thanks, I already tried that. It so didn't work.

    I get the principle, woohoo, we found the deviation, I get how it applies, but when we come back to the original concept, as to what the practical value of that standard deviation is, I am at a loss.

    What does the result tell me? How does it help me? How am I better for knowing the standard deviation?

    BlackDove on
  • mrdobalinamrdobalina Registered User regular
    edited October 2009
    How much information do you want?

    Six Sigma as a business designation is either the second coming or over-hyped managese, depending on who you ask. If you're interesting in the math portion, Doc linked to the basic idea. If you're looking for the management portion, I can provide a very basic conceptualization.

    For managers involved in most anything that can be broken down to a process (from manufacturing to processing tech support calls to treating patients in a hospital), six sigma is a method that supports the processing function by eliminating errors. It's a big deal for many companies, as it teaches the methods used to identify errors (usually measured on a per transaction basis) and is supposed to eliminate re-work and increase quality. I was taught the DMADV method, which is:

    Define design goals that are consistent with customer demands and the enterprise strategy.
    Measure and identify CTQs (characteristics that are Critical To Quality), product capabilities, production process capability, and risks.
    Analyze to develop and design alternatives, create a high-level design and evaluate design capability to select the best design.
    Design details, optimize the design, and plan for design verification. This phase may require simulations.
    Verify the design, set up pilot runs, implement the production process and hand it over to the process owners.

    Is it a good process? Yes. But it can also be over-thought and poorly implemented.

    Are there alternatives? I guess. Management is full of de jour methods and buzz words. Finding the ones that work really depends on what your industry is and what exactly you're trying to accomplish. In your case that's......?

    mrdobalina on
  • BlackDoveBlackDove Registered User regular
    edited October 2009
    Okay, three new posts while I was being baffled by two others.

    I guess I kinda sorta get the moon landing. I definitely get the "adding process after process until everything can be accounted for".

    I guess it is easier to understand it a bit now.

    BlackDove on
  • QuetzatcoatlQuetzatcoatl Registered User regular
    edited October 2009
    Quality control in general is something that didn't really exist until the late 80's or so. Six sigma is a benchmark for how many mistakes or defects you are having and completely depends on how you define it.

    It gets thrown about as kind of a business model or whatever because it sounds snappy but it really isn't. Quality control methods are usually all about tracking errors and finding problem spots and fixing them. A lot of statistical analysis and stuff to find out causes to problems, then fixing what you recognize as your biggest causes, then starting all over again until you get to the point where you are satisfied with the errors.

    I don't really know many of the details, most of what I've learned is from ISO 9001 stuff.

    Quetzatcoatl on
  • BlackDoveBlackDove Registered User regular
    edited October 2009
    mrdobalina wrote: »
    In your case that's......?

    Being employed here.

    BlackDove on
  • SpeakerSpeaker Registered User regular
    edited October 2009
    Standard Deviation is the average of the distance between the numbers of a set and the Average of those numbers.

    So, like, say you are manufacturing something. Say you are bagging jelly beans and you have a batch of five bags that weigh:

    2 oz.
    2.5 oz.
    2.3 oz.
    1.7 oz.
    1.5 oz.

    The average weight of the bags is 2 oz.

    The standard deviation is a way of measuring how much irregularity in bag size you have. So you take the distance each weight is from the average weight. 0 + .5 + .3 + .3 + .5 = 1.6 / 5 = .32

    So if you are supervising people who bag jelly beans you can match the standard deviation numbers for the different batchs and work out how much error you are willing to live with. Or something.

    But if you are looking at a bell curve created by totaling up all the jelly bean bag weights your company made last year you can use standard deviation a different way. They make a bell curve because the greatest number of bags will be around 2 oz. due to your vengeful supervision, while bags that are 1. 8 or 2 .2 oz will be less common and bags that are 1 or 3 oz. will be very rare.

    It turns out that if you make a vertical line through the peak of the bell curve, one standard deviation to the left and to the right of that line will cover 64% of all your jelly bean bags. So in our case 64% of your jelly bean bags will be between 1.68 and 2.32 oz.

    Two standard deviations out will cover 95% of your bags. Six is going to cover virtually everything except maybe a one in a bazillion freak accident.

    Your example is talking about making something so precisely that the difference between the average product produced and a specification limit is six standard deviations.

    So standard deviation as a number tells you about the irregularity of what you are making. Two or three or six standard deviations tells you about how far irregular products are getting from the average.
    The term "six sigma process" comes from the notion that if one has six standard deviations between the process mean and the nearest specification limit, as shown in the graph, practically no items will fail to meet specifications.

    So in our jelly bean business if you really want to avoid 1 and 3 oz. bags, you need to get the average difference between the bags you are producing and the 2 oz. average down to .16 oz.

    If your records show that you are stuffing bags that precisely, our six sigma formula tells us that you can rest easy that you aren't producing any bags of jelly beans that are 1 or 3 oz. or less/more. That kind of obsession makes sense when you are manufacturing jet engine parts and an irregular part might cause people to die in a fiery ball of fire, but individually measuring each part would bancrupt your company. So you take periodic samples and make sure the standard deviation is within the limit you've set.

    Speaker on
  • BlackDoveBlackDove Registered User regular
    edited October 2009
    Speaker wrote: »
    Standard Deviation is the average of the distance between the numbers of a set and the Average of those numbers.

    So, like, say you are manufacturing something. Say you are bagging jelly beans and you have a batch of five bags that weigh:

    2 oz.
    2.5 oz.
    2.3 oz.
    1.7 oz.
    1.5 oz.

    The average weight of the bags is 2 oz.

    The standard deviation is a way of measuring how much irregularity in bag size you have. So you take the distance each weight is from the average weight. 0 + .5 + .3 + .3 + .5 = 1.6 / 5 = .32

    So if you are supervising people who bag jelly beans you can match the standard deviation numbers for the different batchs and work out how much error you are willing to live with. Or something.

    But if you are looking at a bell curve created by totaling up all the jelly bean bag weights your company made last year you can use standard deviation a different way. They make a bell curve because the greatest number of bags will be around 2 oz. due to your vengeful supervision, while bags that are 1. 8 or 2 .2 oz will be less common and bags that are 1 or 3 oz. will be very rare.

    It turns out that if you make a vertical line through the peak of the bell curve, one standard deviation to the left and to the right of that line will cover 64% of all your jelly bean bags. So in our case 64% of your jelly bean bags will be between 1.68 and 2.32 oz.

    Two standard deviations out will cover 95% of your bags. Six is going to cover virtually everything except maybe a one in a bazillion freak accident.

    Your example is talking about making something so precisely that the difference between the average product produced and a specification limit is six standard deviations.

    So standard deviation as a number tells you about the irregularity of what you are making. Two or three or six standard deviations tells you about how far irregular products are getting from the average.
    The term "six sigma process" comes from the notion that if one has six standard deviations between the process mean and the nearest specification limit, as shown in the graph, practically no items will fail to meet specifications.

    So in our jelly bean business if you really want to avoid 1 and 3 oz. bags, you need to get the average difference between the bags you are producing and the 2 oz. average down to .16 oz.

    If your records show that you are stuffing bags that precisely, our six sigma formula tells us that you can rest easy that you aren't producing any bags of jelly beans that are 1 or 3 oz. or less/more. That kind of obsession makes sense when you are manufacturing jet engine parts and an irregular part might cause people to die in a fiery ball of fire, but individually measuring each part would bancrupt your company. So you take periodic samples and make sure the standard deviation is within the limit you've set.

    Oooooookay.

    That explanation makes a lot more sense.

    Thanks for that.

    BlackDove on
  • SpeakerSpeaker Registered User regular
    edited October 2009
    $1000 spent on statistics class - now officially not wasted.

    Unless someone who is actually good at math comes by in a moment and explains how very very wrong I am.

    Speaker on
  • mrdobalinamrdobalina Registered User regular
    edited October 2009
    BlackDove wrote: »
    mrdobalina wrote: »
    In your case that's......?

    Being employed here.

    Depending on where you are in the company (line manufacturing? sales? service? accounting? clerical management?) the "Management Strategy" will vary.

    Manufacturing has two widely known and generally well-regarded "theories". The first is six-sigma, wherein you focus on making sure that you can make 1 item perfectly, and duplicate it to the point where statistically speaking, they are all perfect. The second is "Lean". I highly recommend Jeffry Liker's book, The Toyota Way.

    If you're in some other form of Management, you'll likely be focusing more on the people aspect of things, which is much harder to categorize. I like Marcus Buckingham's books, others swear by Dale Carnegie.

    You might also want to look at some that don't have easy books, but are worth being trained in. My favorite is the Situational Leadership model.

    Your mileage may vary.

    mrdobalina on
  • EggyToastEggyToast Jersey CityRegistered User regular
    edited October 2009
    The layman's way of thinking about six sigma is that you make a range of products, some of which are perfect (or above perfect). So, how do you make all the shitty stuff more like the perfect stuff?

    And once you do that, how do you make it even better?

    The guys above handle the standard deviation and statistical definition stuff the same or better than I would, so I'll just point out that, as they say, it's typically used in manufacturing. The idea is generally to analyze what you do, how to improve it (with actual benchmarks), and then to actually improve it. And then analyze against your benchmarks and improve it again.

    It ultimately has to do with creating and maintaining a competitive advantage -- a reason why your stuff is better, and why people should buy your stuff. The Toyota stuff dobalina mentions, as well as the rise of Japanese consumer electronics, is the perfect case -- look how many people consider American cars (especially from the 80s and early 90s) to be overpriced junk compared to foreign cars.

    The other point that isn't mentioned above is that, in general, the process of improvement (and reducing the range of your standard deviation) is rather continuous, so you don't just suddenly become super efficient. That means that a company that has a significant head start, and continues to improve their business model, is likely to maintain that advantage over another company that is "late to the party." Again, this is why some argue why Ford is simply licensing Toyota's hybrid technology, as an example.

    EggyToast on
    || Flickr — || PSN: EggyToast
  • OliverOliver Registered User regular
    edited October 2009
    BlackDove wrote: »
    mrdobalina wrote: »
    In your case that's......?

    Being employed here.

    Oh hey we buy your bearings (in very small numbers).

    I've got a friend who is a 'black belt' in six sigma however I've never dug into it too much.

    *end non-contribution to thread*

    Oliver on
  • TheMarshalTheMarshal Registered User regular
    edited October 2009
    BlackDove wrote: »
    TheMarshal wrote: »
    I never actually read any Six-Sigma stuff, although when I was interning at Boeing I kept hearing the name thrown about. The only business-management books I've read are The Goal and Critical Chain, both written by a guy named Eli Goldratt. I found them both to be pretty interesting.

    The Goal speaks about the "Theory of Constraints", and how attempting to achieve local optimums within the scope of a production pipeline actually leads to a terribly inefficient system overall.

    Critical Chain looks at things more from a project-management perspective, and talks about how to best schedule dependent tasks. It looks at the phenomena that people automatically give themselves a "buffer" when estimating how long it will take them to complete a task, and how to organize tasks and deadlines so as to best deal with that.

    Okay I think I get it. Tolerances for various deadlines, and the more accurate the optimal production, the more inefficient (and I assume, more expensive) the system is.

    Neat.

    There's a little more to it than that.

    The Theory of Constraints is based around the idea that any system is going to have a bottleneck, and that bottleneck happens because there are dependent events and statistical fluctuations.

    For example:
    You have six cups (labeled A-F).

    You also have a number of marbles which you want to "process".

    You do this by passing marbles into cup A, then from A to B, B to C, and so on, until they get to cup F. (This represents dependent events)

    However, you're only allowed to take 1d6 marbles from the previous position at a time (representing statistical fluctuations).

    If you were to estimate how many marbles per round could make it from A-F, you'd probably say '3', but you'd be wrong.

    So you start out and A rolls a '5', so there are now 5 marbles in cup A.
    B rolls and gets a '3', so you take 3 marbles from A and put them into B.
    C is lucky and rolls a '6', but there are only 3 marbles in B. All of them get put into C.
    D and E each roll a '2', so 2 marbles get passed from C up to E.
    F rolls a '5', but there are only 2 marbles in cup E, and both of them get put into F.
    Round 1:
    A - 2
    B - 0
    C - 1
    D - 0
    E - 0
    F - 2
    

    In the next round, A rolls a '6' right off the bat. So A now has 8 marbles.
    B rolls a '4', and takes 4 marbles from A. B now has 4 marbles
    C also rolls a '4', and takes 4 marbles from B. C now has 5 marbles
    D rolls a '5', and all 5 marbles in C are transferred. D now has 5 marbles.
    E rolls a '3' and gets 3 of D's 5 marbles. E has 3 marbles
    F rolls a '5', but only is able to take E's 3 marbles. F now has 5 marbles.
    Round 2:
    A - 4
    B - 0
    C - 0
    D - 2
    E - 0
    F - 5
    

    A few more rounds without the song and dance (numbers in parenthesis are the roll results):
    Round 3:
    A - 5  (3)
    B - 0  (2)
    C - 2  (3)
    D - 0  (1)
    E - 3  (5)
    F - 8  (5)
    
    Round 4:
    A - 4  (1)
    B - 0  (2)
    C - 3  (2)
    D - 0  (1)
    E - 2  (4)
    F - 10 (2)
    

    What you'd end up finding is that the later cups are almost always going to be starved for 'work', because of a bottleneck upstream.

    This is a small example with some non-realistic settings (fluctuations are rarely so uniform so as to have all 'cups' have a possible range of 1-6, for example). There are a LOT more details that I'm probably forgetting, but if it interests you, I recommend picking up The Goal. It's written in novel form, so it's actually a very easy read.

    TheMarshal on
  • mrt144mrt144 King of the Numbernames Registered User regular
    edited October 2009
    Doc wrote: »

    :lol: Excellent off topic post.

    Edit: To explain myself, I am quite cynical of Six Sigma being implimented in businesses because it's mostly a half assed HR directive that pops up every year and any effort to impliment the measuring part requires an investment that usually outweighs the return from Six Sigma.

    mrt144 on
  • AiouaAioua Ora Occidens Ora OptimaRegistered User regular
    edited October 2009
    Since everyone's been talking about Six Sigma, I though I'd mention a different busniness philosophy. At my work they're all big on the Deming system... there's not really a good wiki page for it except for the one on the man himself: http://en.wikipedia.org/wiki/W._Edwards_Deming
    You can see the idea behind Deming management where it lists the 14 points, but the jist of it is getting the management to actualy lead their employees though actual leadership, not just driving them with fear and quotas. Also generally treating the employees as humans intead of parts of the machine, opening up communication between employees and mangement (because who would know best how to improve a process other than the employee who does that process every day), and really to make everyone on the same side, working together to improve.

    I don't know much about six sigma, but other than as a method of measuring precison, is it supposed to apply to how a company is managed and ran?

    Aioua on
    life's a game that you're bound to lose / like using a hammer to pound in screws
    fuck up once and you break your thumb / if you're happy at all then you're god damn dumb
    that's right we're on a fucked up cruise / God is dead but at least we have booze
    bad things happen, no one knows why / the sun burns out and everyone dies
  • Irond WillIrond Will WARNING: NO HURTFUL COMMENTS, PLEASE!!!!! Cambridge. MAModerator mod
    edited October 2009
    Six Sigma is supposedly useful and effective in manufacturing. In engineering and systems, it's a useless and total fucking waste of time.

    Irond Will on
    Wqdwp8l.png
  • mrdobalinamrdobalina Registered User regular
    edited October 2009
    Aioua wrote: »

    I don't know much about six sigma, but other than as a method of measuring precison, is it supposed to apply to how a company is managed and ran?

    It's a method of process improvement, that's about it.

    I wouldn't try to apply it to humans, except in such a way as to how they do the process you're working on.

    mrdobalina on
  • japanjapan Registered User regular
    edited October 2009
    Irond Will wrote: »
    Six Sigma is supposedly useful and effective in manufacturing. In engineering and systems, it's a useless and total fucking waste of time.

    Applied to financial services it basically loses all meaning.

    EDIT: Especially those aspects of financial services that relate to customer services. I can kind of (without having thought about it too hard) see it making sense in the risk assessment or underwriting.

    japan on
  • mrt144mrt144 King of the Numbernames Registered User regular
    edited October 2009
    japan wrote: »
    Irond Will wrote: »
    Six Sigma is supposedly useful and effective in manufacturing. In engineering and systems, it's a useless and total fucking waste of time.

    Applied to financial services it basically loses all meaning.

    EDIT: Especially those aspects of financial services that relate to customer services. I can kind of (without having thought about it too hard) see it making sense in the risk assessment or underwriting.

    Even then it's not useful.

    mrt144 on
  • japanjapan Registered User regular
    edited October 2009
    mrt144 wrote: »
    japan wrote: »
    Irond Will wrote: »
    Six Sigma is supposedly useful and effective in manufacturing. In engineering and systems, it's a useless and total fucking waste of time.

    Applied to financial services it basically loses all meaning.

    EDIT: Especially those aspects of financial services that relate to customer services. I can kind of (without having thought about it too hard) see it making sense in the risk assessment or underwriting.

    Even then it's not useful.

    Fair enough then. I know I have been subjected to Six Sigma being contorted into a shape that looks like it might fit into a process like disputing liability, or credit hire notification. Turns out it doesn't work so well when most of the major variables are outside your control.

    japan on
  • ColdredColdred Registered User regular
    edited October 2009
    As with most of these things it's basically common sense wrapped up in a branded package. It wouldn't be so bad either if they didn't label Six Sigma practitioners as Green Belts, Black Belts, Master Black Belts, etc. These people need to stop trying to make Systems Engineering "cool".

    Coldred on
    sig1-1.jpg
  • JebusUDJebusUD Adventure! Candy IslandRegistered User regular
    edited October 2009
    I've heard of six sigma applied to non-manufacturing jobs from my dad. Basically they took it and tried to apply it to everything because they are idiots.

    edit: the company is. :P

    JebusUD on
    I write you a story
    But it loses its thread
  • KalkinoKalkino Buttons Londres Registered User regular
    edited October 2009
    I was lucky enough to avoid Six Sigma, but I did encounter Itil, which seemed to be more popular at the time.
    It is really popular in complicated procurement or service delivery situations. It isn't a bad idea to get into either, if you are planning to work in IT or Telco industries - because quite often a company will have to provide a detailed report or audit on their compliance as part of the engagement process - so anyone who has anything they can cite, or trot out in sales meetings will be well liked.

    Kalkino on
    Freedom for the Northern Isles!
  • japanjapan Registered User regular
    edited October 2009
    I stems from the thinking that every outcome is measurable (technically true, but the issue is if it's usefully measurable) and every outcome is the product of a process that can be mapped.

    japan on
  • mrt144mrt144 King of the Numbernames Registered User regular
    edited October 2009
    japan wrote: »
    mrt144 wrote: »
    japan wrote: »
    Irond Will wrote: »
    Six Sigma is supposedly useful and effective in manufacturing. In engineering and systems, it's a useless and total fucking waste of time.

    Applied to financial services it basically loses all meaning.

    EDIT: Especially those aspects of financial services that relate to customer services. I can kind of (without having thought about it too hard) see it making sense in the risk assessment or underwriting.

    Even then it's not useful.

    Fair enough then. I know I have been subjected to Six Sigma being contorted into a shape that looks like it might fit into a process like disputing liability, or credit hire notification. Turns out it doesn't work so well when most of the major variables are outside your control.

    It's one of the most egregious problems with finance and economics; Trying to shoehorn in things from various fields of science to serve as a functional model with little adaptation. Some major investment banks and quant departments within them propose that stock prices follow a Brownian motion but they mean it quite literally to the point that they approach it purely as physical system at the exclusion of everything else.

    On a more business management related level, the process of implementing and measuring data is a chore.

    mrt144 on
  • DeebaserDeebaser on my way to work in a suit and a tie Ahhhh...come on fucking guyRegistered User regular
    edited October 2009
    I once worked with someone that put a sign on his door indicating that he was a six sigma black belt. (tool)

    The dude in the office next to him retaliated by putting up a sign indicating that he was a Six Sigma Super Saiyan. (awesome)

    Deebaser on
  • EggyToastEggyToast Jersey CityRegistered User regular
    edited October 2009
    The thing is, there's a lot of people in business who don't really understand how to make a business better. It's not hard to order shit on time, buy a machine to make something, and hire competent people to make everything work. And if you're in the black every year, you don't even have to really know anything about accounting.

    It's when your business starts to do well, or when you're looking to expand, and you suddenly realize that you don't really know anything at all about business that these things start to look interesting. I'm in an MBA program now and it's pretty amazing how many people in my classes have bosses (including some professors) who really have no idea how to run a business. Stuff like "What are the costs of this particular project," which you'd think would be something a business would have a handle on before starting the project, are completely foreign ideas to some managers -- especially in the small business world, where you can succeed by finding a niche rather than having actual business savvy.

    So stuff like Six Sigma and TQM and other buzzphrases pop up and, once there's big seminars and HR initiatives, start to lose their original meaning. They all have a basis, of course, and ultimately it's about "Pay attention to what you do, for god's sake, and always be on the lookout for ways to make it better."

    So many people in business barely pay attention to what they do. You can tell the ones who do because when you encounter them you think "Wow this person is good."

    EggyToast on
    || Flickr — || PSN: EggyToast
  • enc0reenc0re Registered User regular
    edited October 2009
    Speaker wrote: »
    Standard Deviation is the average of the distance between the numbers of a set and the Average of those numbers.

    So, like, say you are manufacturing something. Say you are bagging jelly beans and you have a batch of five bags that weigh:

    2 oz.
    2.5 oz.
    2.3 oz.
    1.7 oz.
    1.5 oz.

    The average weight of the bags is 2 oz.

    The standard deviation is a way of measuring how much irregularity in bag size you have. So you take the distance each weight is from the average weight. 0 + .5 + .3 + .3 + .5 = 1.6 / 5 = .32
    Speaker wrote: »
    $1000 spent on statistics class - now officially not wasted.

    Unless someone who is actually good at math comes by in a moment and explains how very very wrong I am.

    You are so very wrong. Standard deviation is the square root of the average of the squares of how far each deviates from the mean.

    So in your example:
    (((2-2)^2+
    (2.5-2)^2+
    (2.3-2)^2+
    (1.7-2)^2+
    (1.5-2)^2
    )/5)^0.5=0.37

    EDIT: Technically we should be making a small sample degrees of freedom adjustment by dividing by 4 instead of 5, giving 0.41. But let's not get into that.

    enc0re on
  • goatboygoatboy Registered User regular
    edited October 2009
    mrt144 wrote: »
    Doc wrote: »

    :lol: Excellent off topic post.

    Edit: To explain myself, I am quite cynical of Six Sigma being implimented in businesses because it's mostly a half assed HR directive that pops up every year and any effort to impliment the measuring part requires an investment that usually outweighs the return from Six Sigma.

    Heh. A while back my company (in manufacturing) decided to go with Six Sigma. All of the profit gains (and then some) were eaten up by the consulting firm they hired to implement Six Sigma.

    Their solution to this was to hire another consulting firm to implement Theory of Constraints.

    ...

    goatboy on
  • Irond WillIrond Will WARNING: NO HURTFUL COMMENTS, PLEASE!!!!! Cambridge. MAModerator mod
    edited October 2009
    Raytheon is currently really invested in Six Sigma. It's just a chore and I've never seen anything useful come out of it. Basically it's a labor-intensive process improvement practice that involves gathering tons of statistics and trying to quantify the value of changing a process. Which probably makes a lot of sense in manufacturing.

    But in Defense Contracting Systems Engineering, most of the time your "product" is "shitloads of mandatory paperwork" and any actual engineering is ad-hoc, project-specific and a one-off, thus obviating any opportunity to "improve" a process.

    God do I hate 6Sigma

    Irond Will on
    Wqdwp8l.png
  • SpeakerSpeaker Registered User regular
    edited October 2009
    enc0re wrote: »
    You are so very wrong.

    Ah yes, the square roots. I remember it now.

    Time plus shaky math skills have put me to public shame.

    I am undone.

    Speaker on
  • DocDoc Registered User, ClubPA regular
    edited October 2009
    Ugh, this reminds me of how many software teams I've worked on that have tried to implement the scrum model.

    A lot. How many did it in an effective manner? Zero. The meetings are supposed to be "Alright, who is blocked? Everyone knows what they are working on? Good. Let's go to it."

    Instead, every morning they go "alright, time for the standup," and everyone goes into a room and sits down. Then people keep on getting sidetracked on things that really ought to be discussed outside of the meeting, causing everyone to waste 45 minutes of their day.

    Doc on
  • ShadowfireShadowfire Vermont, in the middle of nowhereRegistered User regular
    edited October 2009
    JebusUD wrote: »
    I've heard of six sigma applied to non-manufacturing jobs from my dad. Basically they took it and tried to apply it to everything because they are idiots.

    edit: the company is. :P

    Our company has been attempting to use 6Sigma for a while... grocery stores. :|

    Also, 5S can burn in a fucking fire.

    Shadowfire on
    WiiU: Windrunner ; Guild Wars 2: Shadowfire.3940 ; PSN: Bradcopter
  • DaedalusDaedalus Registered User regular
    edited October 2009
    Where I work, we're "CMMI Level 5".

    CMMI is basically a bunch of horseshit.

    Daedalus on
  • PantsBPantsB Fake Thomas Jefferson Registered User regular
    edited October 2009
    japan wrote: »
    Irond Will wrote: »
    Six Sigma is supposedly useful and effective in manufacturing. In engineering and systems, it's a useless and total fucking waste of time.

    Applied to financial services it basically loses all meaning.

    EDIT: Especially those aspects of financial services that relate to customer services. I can kind of (without having thought about it too hard) see it making sense in the risk assessment or underwriting.

    Also medical information systems. Luckily that's one bit of bullshit my particular company has avoided, possibly to save room for other bullshit like features that make the product worse and ridiculous versioning systems.
    Daedalus wrote: »
    Where I work, we're "CMMI Level 5".

    CMMI is basically a bunch of horseshit.

    Pretty much any business model quality standard is bullshit. ISO 9000 is another one.

    But management needs something to do all day. Otherwise its just going to be more meetings and layoffs

    PantsB on
    11793-1.png
    day9gosu.png
    QEDMF xbl: PantsB G+
  • KalkinoKalkino Buttons Londres Registered User regular
    edited October 2009
    I like the incompetence theory explanation - so many of us have this to some degree, combine that with laziness and the such and I can imagine some managers/shareholders are always desperately searching for some sort of Holy Grail to save them

    Kalkino on
    Freedom for the Northern Isles!
Sign In or Register to comment.