Corrective Actions

Six Sigma – iSixSigma Forums Old Forums General Corrective Actions

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
  • #36856

    Trish G

    I am looking for information on what warrants a corrective action request?  We are working on an improvement project to reduce the DPPM level with outside suppliers.  We are custom manufacturing company and at times push our suppliers really hard with short lead-times.  Has anyone developed any guidelines or are there any industry standards that depict a corrective action based on types of severity?


    Mike Carnell

    Most of this stuff is an independent event which means an average time/level is pretty meaningless. Motorola set us up on 10 fold every two years (68% reduction per year) but that was different type of measurement.
    When you have suppliers they should be able to offer containment immediately. They may not be able to offer CA on the cause but they should protect you as a customer. If you look at the Ford 8D program and how it was intended to be used (depending on the situation it can be used punitively) it is pretty good start. There is a seldom used tool called a Paynter Chart that is also good for tracking the corrective action.
    The key to good corrective action is to make it”closed loop corrective action” (bless you Jim Blanden where ever you are). Don’t stop with the containment. The only way to prevent escapes is to not build them. That means you need a follow up program for vendors (do not let your SQE’s tell you they are “working with the supplier” – have them define what that means and hold them accountable). The paradox there is that based on the quality of your SQE’s they may need to be trained in more than SPC – a good supplier program means you have well defined specs that flow through your product to their product so that they understand what they need to produce (actually it can go several tiers into the supply base). In most cases they just drive excess cost by insisting on a control chart on everything that moves. At one point we had over 80 control charts on a production line when there were maybe 7 that actually had an effect on production. Had to keep the SQE happy. Control charts and Windsor (or Ft. Erie) the keys to a successful quality survey.
    For in house defects you also need containment but it needs to be chased back to the design cycle. DMAIC is rework. It is reactionary in most cases to a defective design process. Just as before you need to not make the defect to avoid an escape. Close the loop inside your own facility plan the transition of the focus of a program such as BB’s to a DFSS program.
    Those are assignable causes or whatever name makes people feel good. A program that runs without the special cause injection of adrenaline is your best solution. Demand a program from suppliers that shows discipline and results without the special cause.
    Probably the most sensitive topic. “We are custom manufacturing company” Sorry but that really doesn’t mean much. It is the environment you chose to work in so you also need to make a decision to be better than anyone else at it.Your products may be custom but the same process is used repeatedly to make the product. This is about processes and making them predictable. If you are in the custom business then the focus of your program should be understanding the x’s so well that whatever they throw at you you know the process well enough to make it without a ton of setup scrap.As far as the response time – go to Lean. There is no reason why there shouldn’t be a program that constantly drives reductions in cycle times (check out SMED). If you pay extra for fast turn around times there is no incentive to shorten them and they are obviously capable of doing it. Why can’t they do it every day?
    Just my opinion.
    Good luck.



    Find an old copy of Mil-Std-1520 it was the basis for most corrective actions designs.

Viewing 3 posts - 1 through 3 (of 3 total)

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