1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
Dismiss Notice
You must be a registered member in order to post messages and view/download attached files in this forum.
Click here to register.

Early Launch Containment plan

Discussion in 'APQP and PPAP' started by QMSmaster, Jan 20, 2016.

  1. QMSmaster

    QMSmaster Active Member

    Joined:
    Dec 29, 2015
    Messages:
    59
    Likes Received:
    9
    Trophy Points:
    7
    Does anyone have experience creating early launch containment plans? What is the intention of this document.

    I recently had a customer ask for this as part of their PPAP requirements. They sent a template but it seems like pointless document.
     
  2. Miner

    Miner Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    576
    Likes Received:
    492
    Trophy Points:
    62
    Location:
    Greater Milwaukee USA
    This sounds similar to what used to be called Prototype Control Plans. You typically would have enhanced controls in place such as increased frequencies of sampling for SPC or inspection, or even additional inspection points. Once you demonstrated the new process was stable, you would go to the Production Control Plan.
     
  3. Golfman25

    Golfman25 Well-Known Member

    Joined:
    Nov 6, 2015
    Messages:
    816
    Likes Received:
    402
    Trophy Points:
    62
    Basically what Miner said. What we did was agreed to a fixed number of parts. Say the entire year was 10,000 pcs. Then the first 1000 pcs. would go thru additional checks, inspections, etc. to ensure the process was good. After that, normal operating procedure would take over. Good luck.
     
  4. Emmyd

    Emmyd Member

    Joined:
    Jul 31, 2015
    Messages:
    38
    Likes Received:
    38
    Trophy Points:
    17
    Location:
    Tennessee
    We handle this as part of our control plan - we have 2 control plans that we include with our PPAP. One is a Pre-launch plan that includes the amount of pieces that will be 100% inspected, usually using a check gage. The Production plan is identical to the pre-launch, minus the 100% gage check. This is how our customer likes to see this.
     
  5. ncwalker

    ncwalker Well-Known Member

    Joined:
    Sep 21, 2015
    Messages:
    261
    Likes Received:
    168
    Trophy Points:
    42
    Location:
    North Carolina
    AHA!! Here is a topic I deal with regularly. So this is called many things by many people. If you look at a "canned" control plan, like the example in the blue books, there's a checkbox in the corner where you indicate "Pre-Launch" vs "Production". Honestly, this is a lot of darn work to make a "Pre-Launch" version (which is Early Containment) vs Production (which is normal operation). I have had on occasion one of my customers ask for a Pre-Launch control plan, and it implies that you adjust your production control plan during pre launch. For example: Let's say you have 5 operations you are doing and we will call them OP10 through OP50. Let's say that after OP30 you are going to require on the control plan that the operator does an intermediate check or two. And your intention is he is going to do this once an hour. Many will tell you that during Pre-Launch, or Early Containment, or whatever you want to call it, that this should be one part every 15 minutes - a heightened level of scrutiny. I believe this is the WRONG approach. Because during this early containment phase, BECAUSE you are performing these extra checks, you will get a false sense of security as to your processes ability to catch bad parts. You "pass" this early containment period, back off on your checks, and suddenly ship bad parts.

    The actual goal of this Early Containment process is to certify that what you intend to do long term is adequate. So here is how you do it:

    1) You identify what you want to monitor. At a minimum this will be the Key Characteristics as defined by your customer. You MAY want to add representative measurements of the process that are easy to measure. What I mean is - sometimes there are dimensions or features you can measure easily that will tell you how your process is behaving that may not have bearing on the final product. Add some of those. So... you have identified what you want to check.

    2) You run your process NORMALLY. Let 'er rip.

    3) You have an Early Containment area set up. And you check the final output of your parts. Typical requirements of the Early Containment area are:
    a) It is NOT LOCATED at the end of the line. It is NOT an end of the line check, which may be part of your intended process, it is separate. It is an audit of your process and should be in a different location. Honestly, it's functionally equivalent to a CS-I process without all the negative connotations.
    b) It should not be staffed by the operators running the line. This point gets some argument. On the one hand, if I use operators from a different line, they may be less forgiving. On the other hand, if I use operators from a different shift, but the same line, I get people who are familiar with the product (better checkers) AND they may learn (by discovering defects).

    Great - you are running your process NORMALLY and you have a kick butt Early Containment area. What next? Well, you funnel the process output through the Early Containment area. Now here is the KEY POINT. It is OK if the process makes bad parts, processes do. But you want the process to catch the bad parts and not ship them. So they should not reach the Early Containment area. (Think of the Early Containment area as a simulation of your customer).

    How many checks do you do? Answer: 100%
    Caveat - the ideal situation is to check 100%. But we do have checks, like hardness, that may be destructive. So 100% is the starting point. You should try and achieve this, but some checks require some thought. Let's look at hardness. If you are batch heat treating 1,000 parts in a batch and you do a good furnace study and determine you can check 10 from this batch and have a good idea of the batches hardness, the you map out how frequently a heat treat batch goes through the follow on process. And use that to set the frequency in the Early Containment area. Again, more is always better, but it OK to deviate from this in special circumstances. Just don't start using "special circumstances" as a crutch to back off on the amount of Early Containment checks. It's a customer requirement, but the point of it is to convince yourself your process is working (not shipping bad parts). Your customer may be reasonable about frequency of check, he may not.

    How long do you do this for? Answer: 90 days (3 months)
    Caveat - Again, this depends on your process, and your customer. 90 days is typical. If you produce much faster than your customer (example: you make a stamping producing 600 parts a minute, but your customer is an assembly plant consuming 100 an hour) you may change the parameters (in a discussion with your customer) to be a number of parts instead of a time period. Or, if he insists on the time period you may be able to convince him to back off on the frequency. Like "We will 100% check the first days production, but after that sample for the remaining 90 days."

    Again - don't forget the goal - the goal is to validate that the process doesn't let bad parts ship to the customer when the process runs normally. Which is why I personally do NOT like to see "extra checks" within the process itself. I want to see the process run as intended.

    What happens if Early Containment catches a bad part? Answer: The clock resets. What I mean is if you are in day 60 of EC and you catch a defect in the EC area, you are back to day 1 of EC and have to go another 90 days.
    Caveat - It's usually acceptable to only continue it ON THAT FEATURE. If you are checking 10 items, and 9 of the 10 go 90 days without issue, but one diameter is a problem, it is usually OK to stop checking the 9 and only continue on the 1 offending problem. Again, have the conversation with your customer. The catching of the bad part SHOULD initiate a root cause investigation as to how it escaped the line. Remember, EC is simulating your customer. If your customer gets a bad part, he's going to want root cause and an 8D. Early Containment should have the same response from the normal production line. But again, apply some sense to it. The first week of EC may yield a lot of defects which you can immediately fix. You may not have to do "full battle rattle" problem solving and let your subject matter experts handle it. This is the shaky period of launch where you are stretching the process. But 30 days in? It should be stable at that point. So an EC breach SHOULD elicit the full 8D/Problem Solving effort. Also, if you do a good root cause with a demonstrable corrective action, your customer may waiver the "reset the clock" requirement so it is in your best interest to do it.

    Phew. That' basically what it's about. Checks done away from the process while the process is new to test it and make sure defects don't ship.
     
    Last edited: Jan 21, 2016
  6. ncwalker

    ncwalker Well-Known Member

    Joined:
    Sep 21, 2015
    Messages:
    261
    Likes Received:
    168
    Trophy Points:
    42
    Location:
    North Carolina
    YAAAAH! I didn't answer the question. :) OK, I am a customer of a LOT of suppliers. And what impresses me the most regarding Early Containment is this:

    1) I DON'T want a "special" control plan, I want you to run your NORMAL control plan.

    I DO want to see you understand what Early Containment is. So I want to see:
    2) A well lit Early Containment area that looks like you take it seriously. Organized, so I am comfortable that you can manage the WIP through it and you won't lose track of which box has been checked and which has not.
    3) A clear instruction that is visual for the people who run the EC area so I am comfortable THEY know what they are supposed to be checking. (Again, I don't need or want an entire control plan for this, a quality plan is fine).
    4) What you are checking with is adequate, this may require a Gage R&R to prove it.
    5) That you are RECORDING your early containment results for each feature that you are checking in an organized way. (I may ask you send them to me daily.)
    6) That you generate action items when you catch a defect and that you follow through on the actions generated.
    7) You SEND ME your Early Containment plan with your PPAP before I get there and not make me ASK YOU FOR IT. Hand me the damn thing when I show up.

    If you give me the sense that you understand Early Containment and it is taken seriously at your location, I most likely will let you go on about your business. But if it looks sketchy, or if I don't get the warm an fuzzy from your PRODUCTION guys and your PLANT MANAGER that you don't consider it value added, I am going to park myself in your conference room and make you miserable. I do not care what your Quality Engineer says. Most Quality Engineers get it. And your quality engineer is the guy I talk to all the time, I'm not there to hang out with him. I am there to see the operation. I am going to talk to your lead production guy and your plant manager. Probably your operators, too. If they give me the impression that Early Containment is a pain in the ass thing the quality engineer makes them do to keep their TS Cert, you bet your butt I am parking myself at your facility until you get it, because you are NOT going to use MY process as YOUR gage. I cannot stress this point enough. Most of "us" that go out there and evaluate "you" fully know how the game works. We have been in countless plants and have seen all the "tricks" you try and pull. We know we can be giant pains in the butts and you'd rather us go away and let you do your jobs. Here is the secret formula to make us go away and leave you alone - when I show up, do NOT pawn me off on your Quality Engineer to babysit me. I am going to be more interested in your production folks and them demonstrating they have a quality culture. If you pawn me off on your Quality Engineer, my first impression will be "your plant sees quality as a necessary evil." If I show up and your PRODUCTION folks have a modicum of interest and working knowledge of Early Containment, and SPC (it doesn't have to be as high as the Quality Engineer's, but it damn sure needs to be more than "blank stares" on the topics). If your PRODUCTION folks take me around your plant and show me this stuff with a little pride, you're going to get 5 stars from me and I will leave you alone. If you don't do this, I am bringing the pain. I don't care how many quality awards and certifications you have hanging on the wall. :)

    Another trick - I may be showing up to see my process right at the beginning, and as such, your Early Containment may just be getting started so there's not a lot to see. If you said "Here's the plan for this area, the process is just getting started. But let me take you over here and show you one that has been in operation for a month....." I would be TOTALLY impressed. Even if the other EC area wasn't my part. It would show me you get it and that's why I am there.

    (What I ought to do is make a guide. How to keep supplier quality off your back :p )
     
    Last edited: Jan 21, 2016
    QAengineer13, hogheavenfarm and Emmyd like this.
  7. QMSmaster

    QMSmaster Active Member

    Joined:
    Dec 29, 2015
    Messages:
    59
    Likes Received:
    9
    Trophy Points:
    7
    Thanks for the replies. It helps.

    SPOOOONNNN!
     
    Bazinga likes this.