iSixSigma

Questions for the masses

Six Sigma – iSixSigma Forums Old Forums General Questions for the masses

Viewing 12 posts - 1 through 12 (of 12 total)
  • Author
    Posts
  • #38839

    Mark Ahlman
    Participant

    I have two major questions.
    1.  In calculating OEE.  I understand how the calculation works and what the variables are, but there is a severe amount of turmoil as to what can be considered “scheduled downtime”.   What we have is a problem with a lot of machines that are not constrained.  In other words, we can do changover without impacting our ability to meet demand, at all.  The way we have been doing it, is that we will have several hours between runs where there is nothing scheduled and our machines are in idle(scheduled shutdown).  During this “idle” classification our technicians will complete the changover well in advance of the scheduled start of the next run.  Now, do you penalize your OEE score for this changover, or do you not, that is the question.  And if so, why?  The intent of OEE is to capitalize on hidden capacity but when you don’t have anything to replace the capacity with, why would you waste resources on that machine to reduce a setup that does not impact your ability to deliver when there is so much else to do.  (For the record, I think we should count it but I don’t have a convincing arguement.)
     
    2. We are plastics manufacturer that installs IML’s in a blow mold operation.  We have had significant problems with scrap (>30%) comming out of the mold.  I am looking for anyone who might be interested in exchanging ideas for improvements in this area, preferably a non competitor, but to be honest I am more about solving the problem than worrying about trade secrets.
    Looking forward to the help. 
     

    0
    #116844

    Dayton
    Member

    Mark,
     
    I think that without clean separation between changeover and scheduled downtime you are blurring the lines for your capacity or utilization metrics.   OEE calculations should in my opinion require this distinction to be drawn in order for you to fully understand what your utilization and capacity metrics are.  You are in this situation because you have unutilized capacity and until you begin a clean separation between change over time and non-scheduled run time you don’t know what additional work you can effectively absorb should opportunity for additional work come your way.  Besides you can’t truly understand your operational constraints until you break the operation down into its base elements and measure and tweak them – again my opinion but from some years of engineering and operations research.   Look at the utility, visibility and operational transparency that a simple Markov Analysis can give you if done correctly – for which you need clean appropriately categorized data.
     
    By IML’s I’m assuming that you mean in-mold labels but can’t get an idea of what you are facing in regard to your excessive scrap rates (and yes 30% is excessive) with what little you have told us about your problem.   Please be more specific as to your problem if you want us masses to effectively respond.
     Vinny

    0
    #116858

    Mark Ahlman
    Participant

    Thanks for the response and I ask for your patience on this, but when I said the machines were well under their capacity, I was talking about machines which will not be running for at least 30% of the year.  Some will not run for more then 60% of the year.  Roughly speaking, we are at about 30%-40% capacity this year and we are changing over machines during scheduled “idle” downtime.  We do have a few cells that run at or near max capacity and thier changovers are viewed as unscheduled downtime and treated as such in the OEE calculation.  But the machines that I am most concerned about are the ones where the setup techs are taking advantage of the schedule and completing the setup well in advance of the production need.  Even if you say that any setup makes the machine unavailable, the return arguement is that you will not have the numerator to the equation in our system, because the machine is in idle.  And worse yet, if you did use the “idle” in the available time then for the 8 weeks that the machine is in “idle” you would be receiving 100% availability.
    As for the IML’s (jackets to insulated water coolers), and yes that is In Mold Label, we are having trouble with dropping the labels before they are applied.  We have also had difficulty with the labels during the mold process(scratching, folding in the mold process, misaligning, and various additional issues). 

    0
    #116863

    Dayton
    Member

    Mark,
     
    I’m not trying to be overly simplistic about it, but you do have several distinct conditions:
    1.) setup   2.) run   3.) down due to repair or PM but regardless not available and   4.) idle time when no work is scheduled but the equipment is available – you can parse these times out even finer but the conditions are what they are.  Don’t be afraid to call it like it is – if your equipment is idle and available it is non-utilized due to excess capacity/nonscheduled time for available equipment – it is what it is.  Convoluting your OEE calculations to obviate or obscure the fact that you have more available equipment than orders helps no one.  
     
    Regarding your IML’s, be careful acting on suggestions from an anonymous forum, when I read what you are saying I envision equipment, material and processes more than likely far different than you actually have.  I’ve seen some problems like you are describing that were attributed to misaligned pick and place machines and label location misalignment, excessive mold release, excessive silicone, poorly maintained molds, incompatibility between materials, sloppy handling, ejected hot parts falling too far to collection tub or conveyor, etc.   
     
    It does sound like you have the makings of a good little Six Sigma project on your hands though.
     Vinny

    0
    #116865

    Sinnicks
    Participant

    Vinny,
    First of all, I realy appreciate the help.  This is the first time I’ve worked with either one of these issues. 
    So, let me through this back at you, Even if we are setting up during idle time, we would still want to include that as part of our unscheduled downtime and take the OEE hit?
    As for the IML thing, I have about an ounce of experience with plastics and I am looking for someone to converse with on this issue as I move through it. 
    Again, thanks for the help.

    0
    #116867

    Dayton
    Member

    Mark,
     
    No.  I am suggesting that you take a pristine and non-combinatorial approach.   Your setup time is setup time – record it as such.   If after you have setup your equipment you have additional idle time due to lack of equipment utilization – nothing scheduled to run – record it as such.   Setup and PM time should not be unscheduled downtime – they are scheduled downtime.    Equipment failure or breakage that needs repair, unexpected lack or materials, power failures, short staffing, jerking the production schedule back and forth, etc., can all be sources of unscheduled downtime.   If you have downtime due to not having orders to process – you need to take the OEE metrics based hit – your business is adversely impacted due to excess capacity/lack of orders, either way you want to look at it. 
     
    I’ve worked in and around IML molding operations and problem solved for some but certainly would not consider myself an expert.  Others in this forum are though and perhaps one or more will step up and lend expertise.   If you need help, you need experienced capable help.  And I hate to say it, but you need experienced help on site looking at your materials, equipment, facility, utilities, process, maintenance, staffing, training, and (gasp) leadership. 
     Vinny

    0
    #116916

    Sinnicks
    Participant

    Thanks Vinny,
    I hadn’t even thought about taking the hit for idle time, but I like your train of thought. 
    Thanks again.

    0
    #116919

    Cone
    Participant

    what would your change over people be doing if they werent doing the changeover?  Something else id imagine.  Which is why i would calculate it like normal.  Waste is waste whether it effects your ability to meet capacity or not. 
    What happens when your demand goes through the roof and all the sudden your changeover becomes a problem?  Everybody is gonna wonder why you arent hitting “the numbers” when all this time your OEE was looking “good”.
    Do it correctly.  Anything short of that is asking for problems and or sweetening the numbers in my opinion.
    gary
     

    0
    #117225

    Jeff
    Participant

    If you would like I can send you a presentation on OEE with the specific characteristics according to Taguchi.
     
    [email protected]

    0
    #117318

    Abdelaziz Moustafa
    Participant

    Hello AllI had this problem in calculating the OEE for our machines,
    1- the problem is that the shift is 8 hours per day and it doesn’t change for the whole year..
    2- another problem is that the machines are not working for the whole 8 hours per day…
    the solution proposed is to know that planned hours and the calculate the worked hours from planned hours…for example your planned hours for a certain day is 5 hours and the machine only worked for 3 hours and the rest 2 hours are wasted for setup,breakdown of m/c,etc..
    This can help in improving the machine to achieve the number of hours it should work per day…The question will be it a machine was planned to work 5 hours and it didn’t work due to breakdown of a certain kind…will the machine be consider to be 0% available or what????Hope this help and hope that you can help on my questionAbdelaziz

    0
    #117320

    Chris Seider
    Participant

    I have followed the string of posts and am unsure where the confusion is. 
    Simply put, the machine’s status should be assigned at the hour it is needed.
    If the machine is not planned to be down due to lack of orders or manning, then it is either running or down for another reason (e.g. setup, repairing, short stop/adjustment, cleaning, planned maintenance, and whatever other categories the organization uses.
    To answer this specific post to Abdelaziz…. If the machine was only manned/planned for 5 hours and operators were sitting around doing no production because of the breakdown, then the availability is 0% because 0/5 manned hours it ran.  The other 3 hours were unplanned.  Some organizations take unplanned hours out of their OEE manned hours and some show it as a management loss (lack of orders).

    0
    #128823

    Edo
    Participant

    As a matter of fact, the objective of OEE is not directly getting a high OEE but moreover an insight in losses. A high  OEE you can get by setting other definitions. After all, the definitions you make your self.
    Take a look at http://www.oeesoftware.info  With the OEE Toolkit you ‘ll get all the information.

    0
Viewing 12 posts - 1 through 12 (of 12 total)

The forum ‘General’ is closed to new topics and replies.