iSixSigma

James A

Forum Replies Created

Forum Replies Created

Viewing 100 posts - 1 through 100 (of 120 total)
  • Author
    Posts
  • #79589

    James A
    Participant

    Steve,
    5S was originally Japanese – it’s just a philosophy that many companies elsewhere in the world have adopted.  The English translations of each of the Japanese Ss are:- Sort, Set in Order, Shine, Standardise, and Sustain.
    If you are after the Japanese words for these are Seiri, Seiton, Seiso, Seiketsu, and Shitsuke.
    See this website for more information http://www.tpmonline.com/papakaizen/articls_on_lean_manufacturing_strategies/5s.htm
    Hope this helps – it’s an address my search engine came up with – haven’t read it all.
    James A

    0
    #79549

    James A
    Participant

    C,
    If my memory serves me right the ‘D’ in 8D stands for Discipline – as in the 8 Disciplines you need to follow in order to kill the problem.  These Ds have been in use since the early days of TOPS (Team Oriented Problem Solving).
    James A

    0
    #79535

    James A
    Participant

    That’s the one.  Thanks David.
    Amy, there are other postings around this – Eileen Beachell’s name is worth a further search.
    Good luck in your hunting.
    James A

    0
    #79512

    James A
    Participant

    Yes. 
    If you were to do a cap study on a machine using a CMM and your micrometer – the numbers would be slightly different (resolution) but the overall value will be very similar.
    With apologies to all who posted earlier for jumping in and answering this.
    James A

    0
    #79510

    James A
    Participant

    Amy,
    One of the people involved with the creation of the 8D has posted to this forum before now – I’m ashamed to say I can’t remember her name – but if you use the search facility on previous postings I think you’ll trawl more than enough information.
    Good luck
    James A

    0
    #79502

    James A
    Participant

    If you apply either a prioritisation technique (e.g. ranking), or just go for what the team feels to be the most vital effects in order to reduce the number of metrics you have, then you run the risk of intentionally ignoring a factor which may be the solution. 
    Presumably you are doing this 6S project because you do not yet know the answer to your problem?
    Before you take a huge leap of faith in your ability to detect solutions randomly, then I would suggest that you collect sufficient data to enable an informed decision to be made as to which metrics can safely be ignored.  Yes it can be expensive, but so can taking the wrong decision early in a project.
    Just my half-penny’s worth.
    James A

    0
    #79451

    James A
    Participant

    Broadly I agree with Girish, but if you have an area dedicated to calibration (and storage of ‘spare’ measurement equipment) then you can safely keep the “Not in Use” equipment in that area, and thus safely out of the way of people who may wish to use the equipment in its uncalibrated, albeit “Not in Use” state.
    If it’s out of the engineering lab, and under lock and key somewhere else, then it cannot be used.  Full stop.
    James A

    0
    #79450

    James A
    Participant

    Sarah,
    Trips to Hawaii?  Recognition dinners? Bonuses?  Wow! Who are you working for??  We don’t get incentives, we just get to keep our jobs!
    But seriously, I do agree with your general sentiment that aiming only for the big savings seems wrong – I don’t believe that you can peg a SS hat on a project on the basis of value alone – although it is true that most of the time if you have a problem that is creating business ‘pain’ then there is a cost associated with it.  Creating a rule that says “We only attack projects worth over $xxxxxxxx” was probably originally done to focus attention on the big hitters (good for the shareholders) – but has now become a bit of a red herring.
    Yes, the good quality, manufacturing and test engineers do a damn good job, day in, day out resolving day to day problems – but that’s what they are paid to do at the end of the day.  If they don’t get the recognition, then I would question if a Six Sigma roll out is going to work effectively in your place.
    Six Sigma is aimed at the long standing problems that have been ‘fixed’ many times before – only to return yet again and cause trouble.  These projects are difficult – that’s why they haven’t been fixed before, and that’s why you need the Six Sigma training to fix ’em.
    No, the project doesn’t have to be big, but yes, it does have to be difficult/longstanding/of customer concern.
    The day to day fixes by the normal team efforts are all good stuff, and all help reduce costs and increase efficiency – but it’s a whole different ball game.
    If you can fix a problem easily, then just do it – don’t get a Six Sigma team going if all you need to do is get maintenance to fix a leak.  There are other, lower cost, tools to do that.
    From your comments on ‘demeaning’ (and yes, you did spell it right) of colleagues, lack of recognition, and ‘cutting off at the knees’ I would hazard a guess that your Six Sigma deployment is not going to last long as yes, you are absolutely right, Six Sigma should be a CONTINUOUS effort at ALL LEVELS
    From those same comments, I would guess that your colleagues have not been included in the project teams yet and hence feel somewhat aggrieved – I wouldn’t blame them.
    Keep smiling – it’s Friday and it’ll all be over soon.
    James A

    0
    #79056

    James A
    Participant

    Amen to that.
    Thank you Mike for an excellent response.

    0
    #78976

    James A
    Participant

    Raymond – I’ve never yet used one of these, but have been able to track down the following references which may be of help (or they may not)  “Implementing Six Sigma” by Forrest W. Breyfogle III has a large-ish section on CUSUM charts.  MiniTab’s Help section also goes into some detail (with worked examples) if you have access to a copy of that.
    As I understand it your centre line should be a zero value, i.e. you are looking at the cumulative sums of the deviation of your samples from a target value.  No deviation from target, no problem, hence value = 0.
    UCL and LCL are +/- 3 sigma from this 0 value, and the sigma is estimated from MRbar/d2 (as per usual guesstimates of sigma) so you will need some historical (hysterical??) data to start with to get a current sigma value to work with.  The moving range is of length 2, since consecutive values have the greatest chance of being alike.
    None of the formulae I’ve looked at refer to ‘alpha’ but I’m guessing here that it may refer to an acceptable quality level – so is there an acceptable reject level as well in your formula? If this alpha is the case, then you are probably using the chart to determine incoming parts quality.
    Now I’ve stuck my neck on the block, is anyone else out there prepared to offer any further constructive comment to help Raymond out on CUSUM charts??
    James A

    0
    #78899

    James A
    Participant

    Patrick,
    Try here http://www.ncl.ac.uk/cpact/mspc_demo0_nf.html 
    Hope it’s of interest.
    James A

    0
    #78898

    James A
    Participant

    In a previous professional existence, when I set up and ran a Kanban system, we only used tokens at the machining operations – these were arranged in columns (according to part and op number) on boards.  As each box of parts was completed, the token was removed from the board and placed in the box with the components.
    As each machine was running around four parts for the same product, it was impossible to ’empty’ all four columns simultaneously – so as demand pulled on the machined components, tokens were removed and placed back in the relevant holder – eventually the stack reached a “re-order” point (based on safe shop stock levels) and that determined the next change-over after the job that was running.
    In this way, we placed control essentially in the hands of the assembly area, and the ‘pull’ worked back into production from there – the materials people didn’t really have any input.  It was only the assembly and machining areas that had control.
    Visual signals were used, as I said earlier, on the machines themselves, and in the window of the assembly area to call for more (specific) parts.
    Because of the general racket and noise we didn’t use sirens or flashing lights – only boards and signs.
    One thing to be aware of though, the Kanban system will rapidly highlight any production or part problems – in a way this is good, because you can then fix them, but if the company was previously of a ‘batch’ mentality (when the problems could be effectively hidden) then it’s a great excuse to declare that Kanban or JIT doesn’t work.
    That can be an unpleasant experience if you’re not prepared!!
    James A

    0
    #59731

    James A
    Participant

    Jill,
    I’m not in the medical world – but I would suggest that you start by getting from all your key ‘Heads of Department’ exactly what it is that they would use as measures of success – i.e. determine their ‘Critical to Quality’ items in order of importance.
    What does ‘Admissions’ have as CTQs?  Does this compare with what those being admitted require?  What does ‘Surgery’ have as CTQs – what about the patients?  What does ‘Housekeeping’ have as CTQs – how does this compare with patient and staff requirements? etc. etc.
    Once you have that, then ‘all’ you need to do is design the systems that will support and verify those measures, coupled with ‘live’ planning documentation in support of new/refurb’d equipment, practices, methods, procedures (medical and paper-based).
    It shouldn’t really be any different to any other process – it just looks that way because you’re dealing with bodies (animate and not).
    Once you have their CTQs identified, then like any other customer, you will be able to provide what they want/need, and get your bill paid.  I would suggest that ‘Administration’ will provide areas rich with potential savings – both my brothers are medics, so that comment may a touch tinged with inside knowledge/bias, but you can see where I’m coming from.
    Hope this helps kick things off.
    James A

    0
    #78872

    James A
    Participant

    Jill,
    I’m not in the medical world – but I would suggest that you start by getting from all your key ‘Heads of Department’ exactly what it is that they would use as measures of success – i.e. determine their ‘Critical to Quality’ items in order of importance.
    What does ‘Admissions’ have as CTQs?  Does this compare with what those being admitted require?  What does ‘Surgery’ have as CTQs – what about the patients?  What does ‘Housekeeping’ have as CTQs – how does this compare with patient and staff requirements? etc. etc.
    Once you have that, then ‘all’ you need to do is design the systems that will support and verify those measures, coupled with ‘live’ planning documentation in support of new/refurb’d equipment, practices, methods, procedures (medical and paper-based).
    It shouldn’t really be any different to any other process – it just looks that way because you’re dealing with bodies (animate and not).
    Once you have their CTQs identified, then like any other customer, you will be able to provide what they want/need, and get your bill paid.  I would suggest that ‘Administration’ will provide areas rich with potential savings – both my brothers are medics, so that comment may a touch tinged with inside knowledge/bias, but you can see where I’m coming from.
    Hope this helps kick things off.
    James A

    0
    #78738

    James A
    Participant

    My pleasure.  Thanks for the explanation, it makes sense, now.
    Smile, it’s Monday!!!!
    James A

    0
    #78731

    James A
    Participant

    Marnie,
    If you already have an historical value for s then I would be tempted to use X bar moving R charts (or soft equivalent with a sample size of 1) that way you can use all the points from the previous month, and use the existing s-value (+/- 3sd for limits), if there has been no change to the process.  The only question I have is – What will you react to 30 days after it happened?  It must be a fairly stable process, so do you actually need to run SPC on it?  Or is it something like absenteeism and just for the record?
    Intrigued.  Anyone else got other suggestions?
    James A
     

    0
    #78611

    James A
    Participant

    Cat,
    There are a number of us in the UK who also post here – but we are usually only detectable by spelling and phraseology.  For various reasons I prefer to remain relatively anonymous beyond revealing my country of origin, but if you post your mail ID, I’ll contact you.
    James A

    0
    #78600

    James A
    Participant

    That depends on the culture within your company – where I am it can can be recognised that improvements are there to be had, but they may have constraints imposed upon them – therefore we could close the project as the BB input was complete.
    It may be different for you where you are – and I cannot tell you what to do as your company may well be very different to mine.  Certainly I would try to close the project off – what does your Champion suggest?
    Sorry for delay in reply.
    James A

    0
    #78596

    James A
    Participant

    If you have identified what needs to be done to gain the improvements, and all parties agree that this is the case – yet the constraints of the business prevent hiring of the necessary staff, then I would suggest that that is a business issue – not a 6S one, and is therefore outside the scope of your project. So close the project.
    If it is a case of not wanting to move staff around, then it’s probably a cultural issue – always a knotty problem to change.
    If it’s that the labour does not want to see the improvements realised, and is resisting change, then you will either need to challenge their assumptions and change their way of thinking, or move staff around until you have the team you need in that area.
    Either way, you seem to have proved already that the improvements are achievable – so your job is now done – it’s up to others to get the necessary motivation going.
    Does this help?
    James A

    0
    #78575

    James A
    Participant

    Kerem,
    At least one of them works in the textile industry – at least that’s what I read from the content of the discussion.
    James A

    0
    #78573

    James A
    Participant

    Kerem,
    Try this thread for clues – you may be able to get some of the contributors to contact you?
    https://www.isixsigma.com/forum/showmessage.asp?messageID=17413
    Good luck  Isn’t it quiet out here with the US on post Labor Day holidays?
    James A

    0
    #78519

    James A
    Participant

    Just goes to show that Co-Co the clown’s offspring have pervaded all areas of industry.  I would say that there is more than a cultural issue at stake with this individual.  Intensive aversion therapy using electric cattle prod may be required.  Extended education in common sense and what gauging is for may also prove helpful.
    Barmy.
    James A

    0
    #78490

    James A
    Participant

    James,
    I almost agree with everything you’ve just said, but with one exception – management is no more always right than the beloved customer – but hopefully if you have the right management then your project will still move the business in the right direction (management satisfied if surprised at incorrect assumptions) or satisfy the customer at the expense of a few myths or assumptions being dispersed.
    Either way, there needs to be 100% commitment to discovering the underlying problem, and eliminating it – even if this does require a degree of maturity on all sides! said maturity will determine the continuance of your salary – otherwise your projects and existence become that of a “Yes” man – and that does not help anyone.
    Just my $0.02
    James A

    0
    #78477

    James A
    Participant

    Hiya Billybob,
    Thanks for the link.
    As far as what goes where is concerned, I think it probably depends on who taught you in the first place – where I was it went “Scrap, Overproduction, Inventory, Motion, Processing, Transport, Waiting” – so all the key elements are in there – just not in the same order.  The important thing, I think you will agree, is that the wastes are identified and eliminated.
    Any other definitions/orders/variations are probably due to regional variations and semantics – neither of which are really worth a bean.
    Give the possums a scratch from me.
    James A

    0
    #78458

    James A
    Participant

    I can’t speak for everyone, but where I am, BBs are selected on the basis of:- being well-regarded within the company, being able to think laterally/vertically/horizontally/5th dimensionally/out-of-the-box etc., able to lead groups, provide inspiration for others, be objective, non-critical, be future leaders, etc.
    Current job function doesn’t come into it – it’s much more a case of identifying the required skills in the individual, and going from there.  This doesn’t answer your question well, I know, but hope it helps anyway.
    James A

    0
    #78457

    James A
    Participant

    For Motion – this refers to the movement of tooling etc within a machine.  e.g. during a CNC cycle it does not make sense for the tooling to go to 0,0,0 every time a tool is changed, and then start cutting.  It does make sense for tooling to have non-cutting travel reduced to the minimum required (without clouting the job!) 
    It could also refer to movement of materials/parts within the plant – e.g not having Op 10 and Op 20 next to each other, but separated by a long walk (some people may class this as Transport, but as long as you clearly define either transport within the plant or transport outside the plant, it’s OK)
    For Processing – this is re-work – i.e. not getting it right the first time, and having to do it again off-line.
    Hope this helps
    James A

    0
    #78442

    James A
    Participant

    There is a certain amount of training available from within MiniTab – look under the Help menu – Tutorials – and take it from there.  They are not a bad start – and also useful for refreshers.
    James A

    0
    #78272

    James A
    Participant

    I’d try using the arithmetical mean – if you think about it, it makes sense as it is the same as you would use in SPC – which also deals with ‘normal’ albeit skewed (on occasions) data.
    Statistician I’m not, so others may disagree, but that’s where I’d go.
    James A

    0
    #78270

    James A
    Participant

    Morning Ex-GB,
    I’d try using kurtosis to describe the skew – the NIST website is a great source of stats type data, and it’s free – the following link will take you to the area I think you need to look atfor more information:-
    http://www.itl.nist.gov/div898/handbook/eda/section3/eda35b.htm
    James A

    0
    #78247

    James A
    Participant

    John,
    I think so, yes.  There is a manufacturer who produces various gauging and measurements sytems – one of which was designed to measure thickness (using either laser or light from memory – i.e. non-contact).  I think it was designed to measure paint thickness originally.
    By using a label or labels of known thickness (i.e. master parts) then an msa could be done – I believe the resolution was pretty good on the system I read about.
    All the usual disclaimers apply – I don’t work for them, promote their product for reward, or have any connection with them, but try a search on the web for Keyence – it might help.
    James A

    0
    #78177

    James A
    Participant

    I’ve just caught up with this comment in another thread from Anna O’C (the now free, former Annonymous) and I think it sums things up pretty well:-
    “Organizations have “homeostasis” just like organisms do.  So they try to return to their previous state.  That’s why change is hard unless the organization or system is deliberately desdigned and trained to accept change.  That’s why eternal vigilance is the price of real improvement / process control.”
    Maybe that approach is what we should all be aiming for in systems design and 6S projects – thanks for that Anna – and yes, eternal vigilance I totally agree with.
    James A

    0
    #78176

    James A
    Participant

    Firstly,
    Thanks to all who responded – like I said, I’m sure we don’t have these problems here – but one must be ever vigilant – if we can’t cope with truth in engineering, then a lot of people are going to be disappointed in their product!
    Skew and/or kurtosis seem to be worth having a look at some day on some ‘genuinely pencil-whipped’ examples.  Like most of you alluded – changing the environment seems to be a key in this as well – and as has been shown many times on this forum, that’s the hard part.
    And yes, Mike, I freely confess to using non-PC phraseology – but at the end of the day, it doesn’t matter whether you refer to it as being RIF’d, downsized, re-allocated out, transferred, moved sideways, or plain fired – it’s basically all the same thing – and we know it – no matter what country we’re in.
    Thanks again for the input – all of you.
    James A

    0
    #78140

    James A
    Participant

    Hi Marc,
    Thanks for your response – but let’s assume we’ve done all the GR&Rs, we’ve done the work studies, we know the FTF times, and we know the capbility of the machines, and we are depressingly familiar with the good words from aiag, aiob, IATF, QS, TS and Uncle Tom Cobbly and all.
    Let’s also assume that there is no criticism for out of control points, that continuous process improvement is encouraged, and that the leadership promotes ‘Authority to Stop’ where needed.
    Like I asked – is there a tool within MiniTab that would assist in identifying – with a reasonable degree of certainty whether or not ‘pencil whipping’ as you put it, has been or is occurring?
    Thanks for your response – and sorry for the delay in replying – I’ve been off getting my hands dirty – honestly, who’d want to be an engineer (only joking).
    James A

    0
    #78132

    James A
    Participant

    Shandon,
    This sounds like an amazing piece of confectionary – but what happens if you eat a whole bag?  The coefficient of thermal expansion vs. the elastic limit of the stomach?

    0
    #78126

    James A
    Participant

    Eddie,
    I confess that I am not working on any safety type stuff, but if I may offer some thoughts for guidance:
    Firstly – what does your ‘customer’ expect as an outcome of your project?  This answer may start to define it straight away.  If your customer doesn’t know, then try the following:-
    What are the main safety issues you have?  If you’re in a manufacturing environment, perhaps you could pareto the injuries logged over the last year – e.g. fingers 25, toes 10, legs 7, arms 3, head 2, and then examine the data for fingers and toes to see if there are any further pareto(e)s you can do to isolate causes of injury – e.g. cuts, pinching, dropping parts, manual handling errors, forklift accidents etc.
    Doing this will help to highlight where you need to start your project, and will help to define it more clearly – e.g. Finger injuries due to forklift errors.  Then you can start to investigate how the injuries occured, and develop ideas how to prevent recurrence.
    Project definition is the difficult stage – for everyone – but it gets easier with practice.  As some famous golfer once said in response to a question why he was so lucky in winning tournaments: “It’s strange but the more I practice, the luckier I get”
    Hope this helps.

    0
    #78094

    James A
    Participant

    Travis,
    In a way three of my colleagues and I are in a similar position to you – in that we, the BBs, are currently leading the drive to get 6S successfully introduced/implemented into the company – I believe (but I’m not yet totally convinced) that the management/champion support is there – it’s just that they’re mostly too busy to commit to supporting the program at present and cannot attend the relevant training for the aforementioned reason.
    The approach we are adopting is to be very careful what projects we select (high probability of initial success) and to create islands of excellence within the company during this process – the thing about success is that everyone likes to bask in its glory – even if they had nothing to do with creating it.  In a way we are ‘puppy selling’ the techniques – if this phrase does not make sense let me know.
    Also, we are careful not to let projects ‘die of old age’ – either through lack of activity by team members, or through lack of management support.  Lack of activity can be addressed by “selective attitudinal readjustment” – they took away our guns here in the UK, but electric cattle prods seem to work just fine.  Lack of management support is more difficult and relies entirely on the success factor of projects to swing the meter in your direction.
    It’s a tough one – but if you really want to do it you will – but it won’t happen fast, and it won’t be easy – but that’s the joy of being a BB.  If you can develop the right attitude (bulletproof and a huge sense of humour) you can also get a great deal of fun out of it.  If not – well I suppose you can always update your resumé.
    James A
     

    0
    #78050

    James A
    Participant

    Carl,
    I was going to post a reply to this earlier, but was hoping that a more intelligent life form would have a go first.  I worked in a tannery for my year out during my degree and during that time the shoe manufacturers were trying to minimise waste from patterns cut from hides – so maybe that industry could provide some leads/info?
    Also, a similar problem exists for clothing manufacturers in getting the most material patterns out of cloth lengths/widths and minimising wastage there – so again maybe there is some existing bespoke software you could trial from that industry.
    Stamping operations must also have to consider similar issues?
    The other thought that crossed my mind is that since you are faced with a reasonably ‘random’ series of requests from customers, then you are in the same position as some retailers and so need to maintain a ‘stock’ (yes, I know stock is a bad word but bear with me) of pre-cut widths/lengths which will cover ‘normal’ demand and only leave you with the reduced ‘odds’ to cope with – I don’t know your industry so I could very well be talking garbage – but even so, something in my ramblings may spark an idea in someone else’s head?
    James A

    0
    #78049

    James A
    Participant

    Billybob you old (but currently accurate) cynic!
    OK, this one’s nothing to do with Quality, but I love it nonetheless.
    Another Day ended,
    All tasks completed,
    All targets met,
    All systems fully established, up, and running,
    All staff happy, fulfilled and enthusiastic,
    All pigs fed and ready to fly
     
    Now isn’t that great?  If you don’t like pigs, you can always substitute possums.  Who wrote it – I don’t know, but they must have been in manufacturing at some time.
    James A

    0
    #78018

    James A
    Participant

    Carol,
    I’m truly surprised at you – you’ve made a number of assumptions as to the customer requirements, and not based your response on statements from the customer – shame on you  ;-)  (wicked grin – it’s Monday after all)
    James A

    0
    #78015

    James A
    Participant

    That would depend on what the request was – can you give me a clue?
    James A

    0
    #78013

    James A
    Participant

    Thank you Robert for the Petronius quote – it’s a while since I’ve seen that one.  The more things change . . . . . .
    A lecturer once observed that in manufacturing engineering “One should always be on the lookout for tight tolerances, loosely enforced”.
    It appealed to my cynical British sense of humour.
    James A

    0
    #77601

    James A
    Participant

    Does it matter to you if deliveries are earlier than planned?  If not, just use + days after due date, and then use standard notation of (mean delivery time minus 0) divided by 3 standard deviations – as for a unilateral tolerance to give you a CpK figure for smaller-is-better.
    Not thought through fully – just typed and posted – but it might be another option.
    James A

    0
    #77598

    James A
    Participant

    I agree with Gabriel – especially the part about LSL=0 and USL=10 being confusing if you don’t really mean it.  If you get run over by a bus or visiting alien spacecraft then your successor needs to be able to see your logic easily in any of your calculations.
    If you have unilateral tolerancing, stick with (USL-Mean)/3sd or (Mean-LSL)/3sd depending on your case.  Stick with convention – it’s good practice.
    Just a suggestion.
    James A

    0
    #77587

    James A
    Participant

    Reinaldo,
    If you post a message with “To iSixSigma Community” in the subject line, and the question in the body of the message, then someone with site responsibility might pick up your request and give you the details how to email the paper to them for posting to the site.
    I don’t know for sure, but it might work.  Just a suggestion.
    James A

    0
    #77579

    James A
    Participant

    Hans,
    There have been a couple of good threads on this in the past – I think if you were to use the ‘Search’ facility you will be able to track them down fairly easily – I think the titles were something like “Six Sigma Average”, or “Company Six Sigma Value”.
    From memory I think opinions were divided on whether a company average was really that useful a figure  . . . . . . .
    James A

    0
    #77547

    James A
    Participant

    Ovidiu,
    No disagreements with you there – I can blame the Babel fish for this one, I think.  My misunderstanding.
    James A

    0
    #77545

    James A
    Participant

    Ovidiu,
    I know how you are thinking – I’ve been there.  What if your demand suddenly changes?  Where will your bottleneck be?  The key is in the understanding that there is always a (Pareto) of bottlenecks for every process – depending on demand you may not directly feel the effects as process pain, but they will be there.  Consequently if you acknowledge their presence and deal with them before they become painful, then the existing process will also benefit.
    Just me and my shovel.
    James A

    0
    #77537

    James A
    Participant

    Sorry to butt in on this thread, but to support what Ron has said – we do the same thing.  In effect your stores area beomes like a bonded warehouse – you only pay for the parts as they are withdrawn from the consignment stock. 
    Ron is right, it’s a great idea, and it works really well – you just need to get your suppliers to agree to it – and that shouldn’t be too hard as they get the same sales in the end, anyway.
    James A

    0
    #77188

    James A
    Participant

    Hello Mel,
    I don’t think there is a stated rule or standard for how often any gauge or measuring device is calibrated.  Generally, gauging of any sort is calibrated according to how much use it gets, and how much it wears between calibrations. 
    E.G.  if a gauge’s history shows it wears appreciably between calibrations, then it should be calibrated more often – but a similar gauge used in another environment may not wear so fast, and can be calibrated less frequently.
    Because this can involve hundreds of gauges across a business, many businesses use an automated gauge tracking system, so that the standards engineers are warned of a gauge needing calibration.
    So, to answer your question:- calibrate gauging as often as it needs it.  If you need information on how to do a Gauge Repeatability & Reliability (GR&R) study, any text book that covers MSE (Measurement Systems Error) should give you the necessary detail.
    James A.

    0
    #77151

    James A
    Participant

    OK Billybob, let me suggest the following:-
    1) Long term, you need to simplify the system – so that the data gets entered once into a data ‘pool’ and can be accessed by all interested parties direct from there – so reducing errors.  Your systems/IT people should be able to come up with a few good(?) ways of doing this – a relational/SQL type database seems like a good place to start
    2) You may have to resort to manually checking all the data yourself initially – you know, if you want a job doing right, DIYFS.  Once you have a level of confidence in the data collection and distribution, then you can at least prove you’ve considered the problem, and acted accordingly. If you find it’s a problem, then you need to come up with a solution to that issue first, and then go back to the real ‘project’ work.
    Don’t worry about seemingly going round in circles – iteration will always occur in projects, and does help (tho’ it is a bit frustrating).
    3) If the person immediately behind you in this long chain of repetitive data entry currently puts the data into anything vaguely Microsofty, then you can just copy and paste the numbers direct into Minitab (saving hours of typing, and allowing you to go home on time to feed the pining possums).
    Does this help?
    James A

    0
    #77130

    James A
    Participant

    Norrin,
    I think Gabriel and Lee have both answered this during their daylight hours.  I think Gabriel is correct in that a copy of the AIAG manual would be very useful – it will help to give you confidence that you are heading in the right direction.
    Good luck.
    James A

    0
    #77107

    James A
    Participant

    Norrin,
    I had to look this one up – it’s been a while.
    1. For each set of results, calculate the proportion defective, for your data e.g. (21/462), (11/309) etc.
    2. Add all the above numbers together, and then divide by the total number of parts inspected – this will give you p-bar.
    The UCLp = (p-bar + 3*sqrt of p-bar*(1-p-bar))  / sqrt (n)
    the LCLp is the same as above, only p-bar – 3*sqrt etc.   ‘n’ is the constant sample size used every time.
    If you don’t use a constant sample size each time, it gets a bit more complex – as the UCL and LCL have to be shifted each time.  See the AIAG SPC Reference manual for the full works.
    Hope this helps a bit.
    James A

    0
    #77097

    James A
    Participant

    Abhinav,
    What is in the section headed “Six Sigma Basics” ?
    (left and below in the blue column – it might help)
    James A

    0
    #76881

    James A
    Participant

    YourDictionary(dot)com supplies the following definitions:
    Main Entry: in.no.va.tionPronunciation: “i-n&-‘vA-sh&nFunction: nounDate: 15th century1 : the introduction of something new2 : a new idea, method, or device : NOVELTY- in.no.va.tion.al /-shn&l, -sh&-n[^&]l/ adjective

    And for variation (not all listed):-
    Main Entry: var.i.a.tionPronunciation: “ver-E-‘A-sh&n, “var-Function: nounDate: 14th century1 a : the act or process of varying : the state or fact of being varied b : an instance of varying c : the extent to which or the range in which a thing varies
    I would hazard a guess that in order to have variation, the innovation must already have happened, been defined, drawn up, modelled, and specified.  In other words, the “Out of the Box” type thinking has been and gone (a good thing if you rely on continuous innovation).
    Six Sigma is surely about reducing unwanted variation to an acceptable (near zero) level (also a good thing if you rely on repeat business).
    Regards
    James A

    0
    #76870

    James A
    Participant

    Annonymous,
    Do please use the example if it helps – I thought about this overnight (sad, isn’t it) because when I was typing the story I was unsure if the score was seven or eight – so seeing your response this morning confirms in my own mind that I was probably wrong in saying 8, and that it should have been 7.
    Either way, it’s a low-budget game that I thought really brought the point home, and yes it applies to Lean extremely well.
    The company I was referring to was into Kaizen, Lean, 5S and rudimentary 6S years ahead of other engineering firms in the UK.  Unfortunately the culture just wasn’t there, so yours truly as a then Kaizen engineer, and the colleague I referred to both ended up looking elswhere for gainful employment.  Sadly, he was so hacked off that he left engineering permanently.
    Enjoy.
    James A.

    0
    #76848

    James A
    Participant

    Marty,
    I remember my DOE trainer saying that averaging data loses you good information, and the more you average, the more you lose.  I think it was good advice.
    I’d stick to your guns.
    Just my tuppence worth.
    Regards
    James A.

    0
    #76845

    James A
    Participant

    Gabriel and Mike,
    Good morning/afternoon/thing to you both.  I think I heard the sound of a hair splitting, but I’m not sure . . . . .
    I love it when this happens – we started off discussing whether or not the input from a previous process should be included in an FMEA.
    The end-state is that we are agreed teams need to talk to each other, and be cross functional (in the true sense of the word), and that inputs from previous processes need to be considered, but need not be included on an FMEA. 
    But according to the FMEA training given by the major OEMs, a Process FMEA should always assume the part is received from the previous process in spec. and in tolerance.
    It is time for us all to retire before we get into a violent agreement.
    Have a good one.
    James A.  :-)
    P.S. Who was/is Dennis Miller?
     

    0
    #76828

    James A
    Participant

    Gabriel (and Mike),
    Maybe I’m being pedantic here, and I don’t wish to start WW III, but to take Gabriel’s points in order:-
    1) If incoming parts are no good, then that is the fault of the previous FMEA team – they didn’t do the job properly (used a ‘template’ FMEA, or carry-over FMEA, one man wrote it, or they just didn’t think).
    After the root cause(s) has/have been identified they need to revisit their control plans, FMEAs and any other relevant paperwork to update them (as per APQP).  You wouldn’t find Ford (for example) assuming responsibility for low incoming parts quality, would you?  Oh, no.  That’s a supplier issue, boss – talk to that tyre company.
    2) Use of a truly x-functional team would have avoided this – Why do engineering teams communicate so badly??
    If it was critical to you that a feature from your supplier was right, what was purchasing doing? Did the engineers ever talk to one another? Did you (or they) do a capability study? If so, on what features? Is the supplier on your ‘Approved List’? Are they quality assured?  Do they get audited by a third party?
    3) What’s your point, exactly?
    As far as tap manufacturing and casting goes, I’ve never made any taps (or indeed faucets!!), but porosity is a common enough problem.  This is a simple incoming parts quality problem – it has nothing to do with your process as such.  It’s a product issue and only highlights a problem from a previous process – see (1) above and use an 8D. 
    If your supplier will not change his process to improve his product, and suit you, then perhaps you should include a functional test if the failure impacts customer satisfaction, or perhaps you should consider re-sourcing – both of which cover the ‘legal’ aspects you mentioned.
    Everyone makes a big deal out of FMEAs (Process or Design) – they are not complicated (unless you are considering an engine management system – UGH), they are one of the great, really useful, simple tools – but everyone gets hung up on them.  They are part of the big Quality picture – not the whole story – think the big picture, and it all falls into place.
    Regards
    James A  :-)
     

    0
    #76791

    James A
    Participant

    Let me guess, the hesitant nay-saying stakeholders are also the ones who feel they will be blamed, right?
    I haven’t worked on an identical project, but the problem you have here is the really tough one of changing attitudes – this is the one you don’t get a bunch of tools for, and for which dear old Minitab just won’t help.
    May I suggest that (although it runs counter to 6S, and you shouldn’t suggest ideas) you present first the findings, and then some ‘solutions’ which do not focus any ‘blame’ on anyone.  This may help encourage ideas to flow.
    Alternatively, a neat trick I saw used by a previous colleague consisted of a ‘process’ each of six people round a table had to run.  These six people were usually managers, who temporarily had to masquerade as operators.  The ‘process’ was a closed box containing two dice, and each manager had to shake the box, and tip out the dice.  If the score was >8 then the product was ‘good’, <8 was scrap (I think it was that way round).  At the end of three runs, the ones who produced the most 'scrap' got a verbal warning, and the process started again, three times – and the ones who 'improved' were rewarded by verbal praise, and the ones who didn't got 'encouragement'/'attitude adjustment'/'training' and ultimately 'the sack'.
    The thing is that statistically this process is biased towards failure – so most of the managers – who worked in a culture of ‘blame the operator’ – ended up getting chewed.  The funny thing was, they never worked out that the process was based on chance, and some of them got really steamed up about ‘getting the sack’.
    Then you explain that in fact, it is the process at fault, and not them – (watch the lights go on at this point).  Now although this was used to demonstrate natural variation, it can also be applied to your current situation, to drive home that it is the process that is the problem, not the users of the system.
    Hope this helps.  (By the way, exercise care in who you use for the above exercise – in a deeply entrenched blame culture it can be career limiting as my talented but unfortunate colleague discovered!  The company subsequently went bust)
    Regards
    James A

    0
    #76788

    James A
    Participant

    Amit,
    Yes you restrict yourself to the process only.  At each stage of the process you have to assume that (1) Incoming product from either the previous process, or (2) the raw material is AOK.
    So to answer your second papragraph, characteristics related to the input for that process can not be taken as a cause (as it should have been considered in the previous stage of the process).
    Regards
    James A

    0
    #76580

    James A
    Participant

    Good question Mary,
    A few thoughts if I may . . . .
    For the first part, I have worked in companies where there is a centralised ‘Corporate’ function.  The advantages of this are that the ‘consultants’ called in can operate outside the petty vested interest bickering often found internally – and can just get on with the job.  They are not clouded by the smaller issues (we think it’s caused by ‘x’, it’s a ‘supplier issue’ etc.), and they are also not influenced by the inherent process knowledge of the internal team.
    BUT they can also come unstuck and propose apparently excellent solutions based on data only, which do not work in the real world.  For example, one team I know of used floor space reduction, manpower utilisation, and machine utilisation as key measures.  Net result, they removed an extensive automated, robotised transfer line, and put people in instead on the basis that floor area was reduced by 40%, machine utilisation was better, and manpower was also better utilised.
    Crazy, but the numbers worked in their advantage.  It didn’t do the product, the margin, or the customer any good though.
    The second group (locally based BBs) have the advantage that they are on site and can swing into action at the drop of a hat – but they need to be careful to remain unbiased and influenced only by the facts – not the fictions so often perpetrated in the manufacturing environment.
    No doubt others will have differing views – but this has been my experience.  Hope it helps.
    Regards
    James A.

    0
    #76415

    James A
    Participant

    Venkat,
    I have just given myself a quick refresher on Chi sqrd tests – I was wrong in my views about expected not being greater than observed. But I’ve (re-) learned something, so that must be good.
    Having cleared my conscience, I can now go home.
    James A

    0
    #76411

    James A
    Participant

    Venkat,
    I am a bit worried that your expected number of counts is more than your initial observed number of counts in Column 1.  This should not happen if you have calculated Expected Frequency using (Row Total x Column Total) / Grand Total.
    Before you draw any conclusions, check your data for sanity.  However, on the basis of your question, on the face of it there is a significant difference between the two populations.
    But check your numbers before proceeding further.  It’s the end of our week now, so I’m logging off, and am out of town next week.  Have a good weekend all.
    Can anyone else help?
    James A

    0
    #76410

    James A
    Participant

    Woei,
    I will try to explain this as a non-statistician, and if I’m wrong then I am sure that someone who hasn’t tried to explain it yet will point out my mistakes . . . . .
    Alpha risk (also called producer’s risk) and Beta risk (aka consumer’s risk) can have values ranging from 0.01 (99% confidence ((or more))) to 0.99 depending on how certain you want to be on your outcome’s statistical significance.
    In any one hypothesis test, and no matter what your value of Alpha is, you always run the risk of accepting a null hypothesis when you should reject it (Type 1 error), or rejecting it when you should accept it (Type 2 error). e.g. Shooting an innocent man or letting a guilty man go free.
    The probability of making a type 1 error is controlled by using a maximum allowable value of Alpha (usually 0.05 for 95% confidence that you are making the correct decision – of course, if you are talking about something serious, like pharmaceuticals, you may feel that 95% just isn’t good enough).  The probability of making a Type 2 error is controlled by using an appropriate sample size.
    So, staying with the example of shooting (well, they took all our guns away!!) assume that we have a busy court room and we get 500 people through in a week.
    The Alpha risk will be = (Number of people found guilty who were innocent / 500)
    and the Beta risk will be = (Number of guilty people not found guilty / 500)
    So, given your example of Alpha = 0.05 and Beta = 0.1, someone is suggesting that you want to be 95% confident that you only find those who are guilty, guilty, but you are happy to accept that you will free 10% of those who are truly guilty but get the other 90%.
    Now the question should be “Are these acceptable levels for what I am about to test?”
    Does this help?
    James A

    0
    #76409

    James A
    Participant

    You are not alone – stats is not my forte, either.  The purist(s) will probably disagree, but I find it easier to view the ‘p’ as the probability of your hypothesis being right or wrong – this then must mean that as one probability goes up, the other reduces, as your total probability has to remain ‘1’ to account for 100% of everything you may expect to see.
    So yes, for your example of 90% confidence, you can accept H0 if p>0.1
    Hope this helps and does not result in my being burnt at the stake.
    James A

    0
    #76408

    James A
    Participant

    Billybob,
    Enjoy the course and wind ’em up a bit – see if their pips squeak.
    I’ll raise a warm one to your success . . . . . we don’t like like it cold and fizzy over here!!
    James A
     

    0
    #76381

    James A
    Participant

    To Err is Human,  to Forgive is divine.
    To Umm and Err is downright inexcusable.
    :-)

    0
    #76379

    James A
    Participant

    And continues here . . .
    https://www.isixsigma.com/forum/showmessage.asp?messageID=14708
    I STILL need more coffee!
    James A

    0
    #76378

    James A
    Participant

    Morning Mike,
    I think it starts here . . . . .
    https://www.isixsigma.com/forum/showmessage.asp?messageID=9598
    regards
    James A

    0
    #76368

    James A
    Participant

    Sorry James,
    p should be > than 0.05 (or 0.10 for your example) NOT < 0.05 if the two machines are similar.
    I need more coffee!!
    James A

    0
    #76367

    James A
    Participant

    James,
    I too will have a crack at this – the way I would approach the problem is as follows (assuming that you only have one process feeding these two test stations) and that the test stations each handle approximately equal volumes of ‘traffic’.
    a) Don’t bother creating a sample plan – if you are only interested in pass/fail, then use a simple process monitor at each of the two machines for total throughput per day and no. rejected per day.  Tally the numbers rejected (if you get a reason or value for rejection, hoover that data up as well as it may be useful later for a Pareto or subsequent analysis)
    b) If you’ve got Minitab, create two columns of sequential data for the numbers of parts rejected by machines A or B.  Then run an ANOVA to see if you get a normal-ish response (i.e. straightish line, no skew or pronounced ‘S’ shape) for each machine, and then do a test for equal variances on machines A and B with the number of bits rejected as the response.
    If you can draw a straight vertical line joining the results of the test between the two sets of charted data (and if p<0.05, or in your case 0.1, change the confidence level when you do the test) then there is no statistically significant difference between the two machines.
    It doesn’t mean they are the same, however.  Just that at 90% confidence, there is no significant difference.  Minitab uses both Bartlett’s and Levene’s tests – it’s a bit belt and braces, but handy as it gives you the results next to the chart.
    I could be wrong on this, but I think it would be a good path to start on.
    Regards
    James A

    0
    #76316

    James A
    Participant

    Woey,
    I think you may have pointed yourself in the right direction in this thread from way back.
    https://www.isixsigma.com/forum/showmessage.asp?messageID=6551
    Good luck
    James A

    0
    #76310

    James A
    Participant

    Steve,
    We used to complain a lot about our intranet server – you know the sort of thing, “Slow”, “Unreliable”, “Crashes” etc.
    Then we discovered that every time someone logged onto our network, the (200+) desktop machines would poll each other to decide who was the server that time.  Sometimes we were all lucky and we got a fast machine, but other times we got some old tired 200MHz thing with an octogenarian hamster powering the hard disk – and this would then tell the main server (dedicated or so we thought) what to do. . . . . . .Was it slow or what!! 
    However, accessing each machine in turn, and switching off the polling facility, we got a useful improvement in apparent server speed, and the complaints dropped.
    I’m not a software type techie at all – so don’t ask me where this gizmo is hidden, but pass it on to your tech. team – we only found it by happy accident.
    Good hunting
    James A

    0
    #76300

    James A
    Participant

    Norrin,
    We use just one ‘page’ for all the data entry – in this case it was for 8Ds raised across the various business units – so if you are familiar with 8Ds, picturing the form should be easy.  Each field on the form can be made ‘required’ if necessary so that key data must be input for the form to be saved initially.
    The fun part (did I really say that? I need to get out more) of Access is the Queries that can be set up – using these you can then filter out, automatically (say by items past their due date) which records on the database require, shall we say, “encouragement to proceed”.  Similar queries can be used to generate listings of which business unit(s) are slowest at clearing 8Ds successfully, or which personnel, part numbers, customers, suppliers or teams, or problem categories, can be Pareto’d to identify the areas to ‘hit’ first.  The queries can feed their data into reports generated by Access.
    Operating with dates in Access needs to be done carefully – hint use someone who can write in Visual Basic, but if you have even limited experience of Access a system can be set up.  Pretty it isn’t – in dbase terms, but if it works . . . . . .
    The next step is to take the system from a file residing on a centrally accessible drive to the corporate intranet – and at that point (using the logins and passwords) you know who exactly is using the system, when, why, for what, and whether they need/ought to be there.  It also prevents people from claiming that M. Mouse (that well known employee) raised the form.
    It is a little Big Brother-ish, but it works, and once everyone knows that they are being monitored, things start to move along nicely.
    Does this help clarify the solution we’ve used?  A screen dump won’t really help, to be honest, as it doesn’t show the structure of the database.  We have structured ours the way it works best for us – that does not mean that it will work best for you – I am only trying to give you a feel for how it might work for you.
    Regards
    James A

    0
    #76256

    James A
    Participant

    Mike,
    As ever, I enjoyed reading some of the postings generated over the last week and I must admit, like others have said, I tended to read those with your name on first – there is usually a good thread around/within it.
    I also like the story you presented, but there is another angle to the age vs experience route.  I cannot remember the name of the theory, but it says that at any one time, a number of people around the world may be working in isolation on a very similar problem – each without the knowledge of the others. 
    The ‘Big Bang’ happens after each party has hit the proverbial brick wall and stopped – only later do they discover and learn about the efforts others have made, and when that happens some of the ‘other’ knowledge provides the key to the heart of the problem and enables further/final development and a solution.
    This is one of the values a forum like this has – it is a valuable resource in itself, but it can also (indirectly) provide us all with a fresh angle on apparently ‘old’ problems, and in the process keep all our thinking ‘fresh’.
    Life is only boring when there is nothing left to learn.
    Regards
    James A

    0
    #76241

    James A
    Participant

    Norrin,
    We too had the same problem of Excel-based spreadsheets being used to track progress – we got round it for one process by developing an Access-based system which was accessible/editable by all necessary parties, and viewable by anyone.
    It has a number of distinct advantages – namely using drop downs gives you consistent data, using required fields means you get complete data, and going the SQL route means that you only have one ‘pool’ of common data to dip into for the dropdowns.  There are other ‘systems’ benefits as well.
    Using a series of ‘handoff’ boxes (for date of handover by email by whom and to whom) can then be used to filter out roadblocks (departmental or personal) for resolution/progression of the task.  The queries can also be set up to simplify reporting/provide evidence for auditors.
    The drawback is that if you have multiple tracking sheets, you need multiple databases – but if the return is worthwhile then it’s worth the development effort.  If the return is not worthwhile, then question the need for the tracker to exist!
    Any help?  Apologies for the late posting on this, but we’ve been on shutdown here for a week.  Something to do with the local monarch’s 50 year celebrations.
    James A

    0
    #75956

    James A
    Participant

    Breizh,
    Sorry, just read your posting – it works better than mine.
    James A

    0
    #75935

    James A
    Participant

    Nick and Yvonne,
    I’m probably going to invite all sorts of trouble here, but what the heck – sometimes it just has to be done – if only to liven life up a bit.
    Both your postings rang a bell – for different reasons.  Firstly if your sponsor is ‘hard headed’ and ‘knows the solution’ then why hasn’t he/she already fixed the problem, and why do they need you to do it?  Maybe they want to see the project fail . . . . obscure I know, but I’ve been there – they get to look good basking in the flames of your failure.
    I have also been involved in a project to solve a high warranty problem, and the team was being railroaded down a particular path on an 8D as the solution was ‘already known’ and had ‘been seen before’ and ‘didn’t need a team to fix it anyway’.  To cut a long story short, it wasn’t the problem, it hadn’t been seen before, and no-one had ever thought about it – BUT the ‘railroader’ realised that the tool would have been used incorrectly, and will never make the same mistake again, I hope.
    Take courage in your convictions, don’t be afraid of failing (at least it shows you tried) and try to coax the blinkered few in the right direction.
    Translating a potential root cause into a cost will only help if that possible cause is the real root cause (or a contributing factor in a relationship) otherwise I would suggest that it is of no help at all and will if anything slow down fixing the problem. 
    E.g. Your car won’t start, so you cost the problems it may have, and decide that on the basis it’s the most expensive problem, you will change the engine.  This you do, but it still won’t start – reason, the battery was flat. Silly example I know, but it’s what being asked of you.
    Good luck
    James A

    0
    #75891

    James A
    Participant

    Carol,
    I’ve only just caught up with this thread’s latest developments, and it’s not often I really wish I’d said something already stated by someone else.  But I this time I do.  Thank you.
    James A

    0
    #75741

    James A
    Participant

    Nancy,
    To follow on from Mike, yes, it’s a good life on the whole – you get some real challenges that no-one else has been able to fix, and if you’re lucky the resources with which to fix them!
    But, apart from the groups of people who really care about what they do, there are also the ones who don’t, won’t, shan’t or can’t care.  This can be unbelievably frustrating and about as much fun as wrestling with a pig in the scrap bin of a nail factory.
    The stats and the analysis show you where to go – and sometimes its very tempting to tell certain employees where to go, too!  Change management, or change empowerment is also a big part of 6S – so be prepared for some pathetic but loud arguments from those who just don’t give a fig.  The attitude of “I just turn up, do a day’s work and go home” is still alive and well – who said Neaderthals died out thousands of years ago?
    BUT – if you succeed in getting these types on your side it is a huge rush – nearly as good as adrenalin, but not quite.

    0
    #75644

    James A
    Participant

    Husker,
    Try looking up ANOVA, or Analysis of Variation. It’s the same animal as an F-Test.
    Use an F-Test (or ANOVA) for more than two groups of data, use a T-Test for two groups only.
    Regards
    James A

    0
    #75462

    James A
    Participant

    TMac,
    I’m unable to answer your question directly because I don’t have a copy of the book you are referring to – HOWEVER – I have made an assumption that the d2 you are referring to is the same ‘fudge factor’ d2 used in SPC to obtain an estimate of the standard deviation of the process, i.e. define the process variation.
    If this is the case, and you have the time to calculate the real SD using the sigma xn -1 button, then the example you have should either point you in the direction of the correct value of d2, or clarify your problem, at least.
    Hope this helps.
    James A.

    0
    #75429

    James A
    Participant

    Gabriel,
    I appreciate someone with a sense of the absurd – it’s good to know that some of us at least can still have fun – even when we are working.  In fact, it can even improve productivity (at low cost, too!!  How’s that for an intangible).
    However, to answer some of your question, “Product” is I think a tangible object, whereas “Service” is an intangible.  I would suggest that “Goods” can refer to either “Product” or “Service” equally, and is therefore referring to things which can be either tangible or intangible.
    An ability to “Deliver the goods” (or not, as the case may be) could therefore refer to anything.
    The generation of meaningless phrases was until recently the domain of management consultants (no offence intended to any consultants here on this site) who needed to baffle bemused management with hitherto unheard of applications of once well understood words in the English (or American) language, and in the process make themselves appear to be more knowledgeable.
    Have fun.  It’s time I went home now :-)
    James A.

    0
    #75391

    James A
    Participant

    Lisa, you are not alone.
    From memory, a Monte Carlo is a simulation method based on use of randomly generated numbers (usually between 1 and 100) to indicate process outcomes.
    These strings of numbers, and the outcomes, can then be used to make reasonably informed decisions about actions to be taken.  I assume it’s called Monte Carlo because it’s a bit of a lottery.
    To be honest, I’ve never used one since college – and then it was an example to prove the advisability of changing all fluorescent tubes in a factory at one time, rather than as they failed.  Useful Huh?
    I think software can do the same job these days much faster and easier – e.g. Witness.
    Can anyone else throw some light on this??
    Regards
    James A

    0
    #75319

    James A
    Participant

    Not wishing this to appear rude Marc, but I do know what the symbol stands for, and I believe I made that clear in my response – I suspect that the inclusion of the phrase “is not an acronym yet . . . . .” caused confusion in the translation software, and was mis-interpreted.
    The correct translation should have been “does anyone out there who has not had a sense of humour bypass wish to come up with alternative suggestions as to what Sigma might stand for if it were an acronym”
    I hope this clears up the confusion – what’s that old saying about two countries with a common language separated by two hundred years?
    :-)
    Regards
    James A.

    0
    #75314

    James A
    Participant

    I don’t know Sang, but how about about
    Statistical Information Gathering Means Advancement ?
    Any other ideas anyone?   (Printable ones Billybob!!)
    To be fair, I don’t think Sigma is an acronym – yet. . . . . . . .
    Regards
    James A

    0
    #75016

    James A
    Participant

    Oh come on Billybob, you should know the answer to that one . . .
    Connect a chaff cutter to the back of the tractor, and set the PTO speed to 540 rpm.  Feed the possums into the chaff cutter one at a time (wear gloves), and measure how far they are thrown out of the machine.  The possum that goes furthest is the lightest.
    While doing this, suck on the fish scale.
    Later, wash the possum chunks, and enjoy the delicious stew.
    regards
    James A

    0
    #74989

    James A
    Participant

    Ally,
    In possible answer to your question, I walked into a hardware shop the other day, and asked if they had a particular product.  I got a blank look and the answer:
    “No, but we get a lot of people coming in and asking for it”
    Regards
    James A

    0
    #74529

    James A
    Participant

    Withheld,
    I say, steady on old chap.  Comments such as this:-
    “Think of the $000,000,000’s that could be saved if (enlightened) design engineers would sit down with (enlightened) manufacturing engineers during the design phase!”
    almost amount to an incitement to revolt.  Much more of this heretical thinking might lead to engineers talking to each other – which would I think we all agree, be a shocking state of affairs!!  No, no, far better we all stay in our little cubes and design in isolation – it’s much more fun that way.
    :-)
    James A.

    0
    #74409

    James A
    Participant

    Dann,
    I suspect that you are interchanging your sigmas rather too freely.
    Sigma (the 6S one we are all aiming for) is NOT the same as sigma the calculated one a.k.a standard deviation.  This is a frequent discussion topic on this forum – and frankly I wish someone would rename the 6 sigma sigma something sensible like ‘Fred’ or ‘Erik’, or ‘BillyBob’ or ‘Possum’.
    The 1.5 Sigma (6S) shift is based on the assumption that over time, and with a sufficiently large number of samples, a realistic (6S) sigma value is 1.5 (6S) sigma less than that calculated to show the success of your project.
    (This is probably not unreasonable if you bear in mind that a machine may have to achieve a CpK of 1.66 to get a pass off, but after that will produce an amazing quantity of non-conforming parts when not being watched over and run by a team of thousands!!  But remember that the sigma implicit in the CpK calculation is not the same as the 6S sigma)
    If you look at a set of Sigma (6S) calculation tables, your figures of 93% success equate to a (6S) Sigma value of 1.5 (long term) or a short term value of 3.
    A 99.73% success rate gives you a long term (6S) sigma value of about 2.8, or a short term value of 4.3.
    Just think of the 6S sigma value as a measure, and don’t confuse it with the ‘real’ sigma that we are all used to.
    Regards
    James A

    0
    #74323

    James A
    Participant

    Hi Mike,
    Sorry about the delay in replying – I literally ‘switch off’ over the weekend – and due to the time difference, I suspect that by the time you posted this, I had already started my weekend!!
    OK, why don’t I believe in RPN as a sole indicator – quite simply, because if you take three numbers, any three numbers, between 1 and 10, e.g. 9, 2, and 7, and multiply them together in any order, you will get the same RPN. So you have 3! possibilities on the FMEA which could all have the same RPN value, but markedly different effects on the next process or the ‘customer’.
    As an indicator of which problem to attack first RPN alone means nothing – unless you have the FMEA as well.  (Many companies regard their FMEAs as commercially sensitive, so won’t share them externally anyway – even with the customer)  If you have conducted the FMEA properly, with the right team, to the correct level of detail, and understand the implications of each line item fully, the additional task of calculating an RPN is non value adding.
    An FMEA, in my view, is a tool for ensuring that a logical sequence is followed by any team (Design or Production) to assess the inherent risks, and reduce them as far as is economically or practically feasible.  It is not a magic wand that, when completed, means that you have a perfect solution, and it is not a mathematical model of predicted behaviour.
    If you have properly considered the system, sub-system, sub-sub-system, . . . . . . . and perhaps even the component levels of your design or process factors, then you have a robust solution.  If you then consider the system to system interactions, and the noise factors, so much the better.
    Does this make sense, or am I in danger of needing a bigger soapbox to stand on?  Maybe I’ve been brainwashed – perhaps one of my previous employers was wrong – but then, they pretty much invented FMEA, so I would expect them to have thought it through pretty carefully.
    My understanding is that the whole issue of RPN reduction came about because a senior exec in an automotive OEM (who didn’t really understand FMEAs, or Quality that well) came up with the idea that all RPNs over 40 had to be fixed.  He didn’t bother checking this idea with anyone, and just announced it to the world at large.  His organisation was then in the position of (a) Recanting this statement and in the process publicly making their exec. look like he didn’t know what he was doing, or (b) accepting the statement, and re-working all the FMEAs.  Guess which option they took?
    The rest as they say, is history.  If I’m wrong about this, please put me right – if I’m missing something, I’m always ready to learn.
    Regards
    James A

    0
    #74271

    James A
    Participant

    José,
    I see no-one responded to your posting, so I will try.
    At one company I came across the attempted use of the Kano model to predict future customer satisfaction based on possible improvements to a product .
    At best I would have described its use as ‘interesting’ – it was visual, different and new, but it did not give really clear direction on where the ‘Best Bang for the Buck’ would come from in terms of shifting customer perception, and regaining best in class position.
    We tried an alternative method, based on market research data, comparing satisfaction with specific vehicle features directly across similar vehicle types, e.g. ‘B’ class vehicles, specific competitors, and then used this to point in the direction where best improvement might be made. It wasn’t the really precise rocket-sciencey approach we wanted, but it helped add strength to the decisions already made by the engineering teams.  There is a thread on this from last year in this forum, somewhere.
    As I see it, the Kano model is not really a mathematical model, it’s more of a sketch with a felt tip pen on an envelope.
    Sorry I can’t be more positive about it but maybe it’s just the experience I had with it.
    Regards
    James A

    0
    #74270

    James A
    Participant

    Didik,
    See this thread three lines below your posting
    https://www.isixsigma.com/forum/showmessage.asp?messageID=12175
    James A

    0
    #74268

    James A
    Participant

    Hi Opey,
    Well, in answer to your question, no I don’t think there is an accepted way in which to “pool” occurences and end up with a magic number, and yes improvisation is probably the way to go – although improvisation is a bit too loose a word for the rigours of a good, well constructed FMEA.
    At the end of the day, an FMEA is really only all about trying to make sure that your team considers as many potential problems as possible, in a logical manner, and weighs up all potential solutions so as to minimise the inherent risks to the customer/end user/next process step.  It should not really be elevated to the point where it spits out a series of magic numbers to save your team thinking about what they should do.
    I agree with Annonymous’ comments, with the possible exception of the very last paragraph, because of the above held views.
    Just for a laugh/thought starter, one of the worst FMEAs I’ve ever considered related to the software content of an engine management system – this was a prime example of Annon’s words relating to the effects of one sub/sub/sub system on another neighbouring system.  I don’t envy anyone doing one of these!
    At the end of the day, if the FMEA disciplines work for you (and your customer) I suppose it doesn’t really matter if you have to sacrifice the odd chicken (or possum) in order to get a good result – perhaps that’s one of the reasons why FMEAs – when done properly – are such a good tool.  If properly approached by a good (cross functional) team, to the appropriate level of detail, they work.
    Over and out.
    James A.

    0
    #74189

    James A
    Participant

    Mike,
    We use two alternating colours on for example, lifting and material handling equipment to indicate that the equipment has been certificated for use – so this year’s colour may be Green, and last year’s colour, say Red, shows that the equipment is either not certificated for use this year, or has failed (in either case, use of the equipment with the wrong colour leads to a disciplinary matter because of the health and safety implications).
    For gauging, and fixtures we do not feel a paint spot is good enough – most of our gauges need calibrating more than once a year!  Therefore we use self adhesive calibration labels to indicate date for next cal. due, and the gauge is marked with a unique reference number – because we have to have traceability, and it makes sense from a logistical point of view as well. 
    In addition to the calibration sticker, a separate computer-based system flags calibration due warnings for gauging and gives the gauge reference number mentioned above. So if the sticker falls off – and they do – we still catch the gauge in the system.  Both the sticker and the PC based systems allow for damaged gauges to be repaired and re-calibrated.
    Does this help?
    James A.

    0
    #74121

    James A
    Participant

    Vinny,
    No the system isn’t broken (John) – it just takes a while to translate Bean-Counterese into English.  I for wun are a enjinear, so finanshul stuf iz triccy.
    We are currently doing a similar exercise using all the ‘Lean’ type tools and trying to get around the problem of supplier issues (late, lousy quality, wrong parts etc.)  One thing I have heard of is using ‘Consignment’ stock where the (qualified) parts stay in a bonded area and are withdrawn only when needed – ONLY THEN do we pay for them – until that point they are owned by the supplier(s) – this has obvious benficial effects for stock value held on the books!!!
    Just an idea.
    James A.

    0
    #74099

    James A
    Participant

    Reza,
    1. See previous reply re 6 sigma.
    2. APQP is Advanced Part Quality Planning – it’s used extensively by the ‘Big Three’ (Ford, GM and DaimlerChrysler) in the automotive industry and is a structured approach to product development aiming for best practice.
    You will get more info on this subject from the Automotive Industry Action Group (AIAG) – try their website – I’m afraid I can’t remember the address.
    Regards
    James A

    0
    #74097

    James A
    Participant

    Reza,
    To save re-writing what has been said before – use the “Search by Keyword” button to the right of this message – all the info you could need will be in there if you use the right keywords searches.
    Regards
    James A

    0
    #73219

    James A
    Participant

    Max,
    If you can understand the cause of the variation leading to the outliers, then their presence may force you into thinking out of the box to remove them.
    Just a thought.
    James    (another Brit, but about to log off for the day – have a good one the other side of the “pond”)

    0
    #73195

    James A
    Participant

    Gemma,
    Sorry, I’ve only just caught up with your posting – being the other side of the planet, some postings get lost in the time/space continuum.  Although manufacturing based, it happens that my current project is transactional.
    I am sure that RR will be able to provide a wealth of info as he does this kind of project all the time, but if you want another head to bounce ideas off (albeit a UK-based head) then please do email me at [email protected]
    Regards
    James

    0
Viewing 100 posts - 1 through 100 (of 120 total)