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.

Error caused to produce bad product, once in 5 years, Root cause?

Discussion in 'ISO 9001:2015 - Quality Management Systems' started by Qualmx, Feb 20, 2021.

  1. Qualmx

    Qualmx Well-Known Member

    Joined:
    Oct 7, 2015
    Messages:
    464
    Likes Received:
    59
    Trophy Points:
    27
    Location:
    Mexico
    Hello everybody
    Recently one of our employees made a mistake and resulted in defective product, this product arrived
    at customer.
    Customer is asking us to do a CA.
    In the analysis, resulted that he had a bad night, was somewhat
    distracted, this employee never makes mistakes, he is a very responsible worker.
    It happened once in the last 5 years.
    At developing the root cause:
    Bad product-> employee distraction-->had a bad night--> noise around house->party of neighbors->>>>
    I think this 5W, stops in had a bad night.

    In this case if you are agree, root cause is had a bad night, if my assumption is fine...
    Questions:
    1-In order to eliminate the root cause, What Can I do?
    2-What action plans can I implement?
    3-Would this root cause will be sufficient to my customer, or will expect something better?
    a deep analysis, other data?

    Or if this was an common error any employee can cause everywhere, any time, could I just
    turn the page and go on?

    And how to close out his CA?
    Please help

    Thanks
     
  2. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    5,086
    Likes Received:
    2,553
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    This is not root cause.
     
  3. Qualmx

    Qualmx Well-Known Member

    Joined:
    Oct 7, 2015
    Messages:
    464
    Likes Received:
    59
    Trophy Points:
    27
    Location:
    Mexico
    I don´t have too much practice on this, I got only a little training on Root cause analysis, however, I don´t detect something different.
    I could address it in other way...
    Bad product-> employee distraction-->he didn´t stop the motor at the right time-->because it was sleepy--->had a bad night, didn´t sleep well--> noise around house->party of neighbors->>>
    again, had a bad night, It can´t go beyond that.
    I´m lost

    Your thoughts?
     
  4. Eric Twiname

    Eric Twiname Well-Known Member

    Joined:
    Jul 31, 2015
    Messages:
    329
    Likes Received:
    232
    Trophy Points:
    42
    Location:
    Northeast USA
    May I suggest that the root cause is the ability of your company to ship a bad part?

    Bad product-> employee didn´t stop the motor at the right time--> no safety interlock to stop the motor at the right time and insufficient QA to catch defective product before it went out the door--> ineffective test and quarantine protocol for out of spec parts---> insufficient QA/QC review of outgoing product.

    No one outside your company cares why it happened...your customer cares why you shipped it and made it their problem. 5-Why around that.

    The problem isn't that a bad part was made...the problem is that a bad part was SHIPPED.

    HTH
     
  5. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    5,086
    Likes Received:
    2,553
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    Once again, why is it "I am doing this"? Didn't your training tell you to put a team together?

    If stopping the machine is Critical to Quality and you have put it on being "awake", then the root cause isn't the operator, it's whomever designed the process and the controls for the process! This operator control should be error proofed. Expecting a process control to be like this, dependent on operators attention, is crazy.
     
  6. Qualmx

    Qualmx Well-Known Member

    Joined:
    Oct 7, 2015
    Messages:
    464
    Likes Received:
    59
    Trophy Points:
    27
    Location:
    Mexico
    In my specific case, this was the only time it happened in several years.
    Always have trusted in the operator ability, and the process was designed like that and had been ok, and didn´t detect
    the needed to automating it.
    Now I see I have to implement a control on the motor.
    On the other hand, the operator error, what is a common practice to address this issue?
    Just occasional distraction and do nothing for this case?
    Thanks
     
  7. Jennifer Kirley

    Jennifer Kirley Moderator Staff Member

    Joined:
    Jul 31, 2015
    Messages:
    1,071
    Likes Received:
    722
    Trophy Points:
    112
    Location:
    USA
    Qualmx,

    Andy and Eric are describing the need to design the process in a way that does not require the operator to pass a restful night in order to produce good product.

    You did not describe the nature of the defect. Was it about holes being drilled in the wrong end of a part? If so, a fixture might be constructed that allows the piece to fit in the machine only one way. This is just an example. To understand what to do, I suggest reviewing the operations with the operator and identify a way that would make success no longer dependent on perfect attentiveness or peace of mind. After all, eventually it would be a different operator with possibly lesser skills. The process should be able to succeed regardless.
     
    tony s and Qualmx like this.
  8. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    5,086
    Likes Received:
    2,553
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    And you may indeed have been teetering on the edge during that whole period. If you trusted just one individual all the time, you got lucky they weren't sick, took vacation, or similar. Another person wouldn't have been so lucky. Plus you put the whole organization at risk and, furthermore, went against the principles of the QMS. I wonder what kind of mental state the person was in too, having such pressure put on them for so long...
     
    tony s, Qualmx and Jennifer Kirley like this.
  9. LiviuSt

    LiviuSt New Member

    Joined:
    Sep 27, 2016
    Messages:
    2
    Likes Received:
    2
    Trophy Points:
    2
    Hello all,
    I have been following this forum for some time and I have discovered quite a few interesting topics. I probably don't have the experience of many here, but now I feel the need to comment a little. I would see this as a good way to improve the process. We can apply 5 Why in several directions:
    1. Why did it happen (occurrence)? In this case, was the operator aware that he was tired? Could he notify the supervisor to be replaced? If the root cause is operator fatigue, isn't process automation justified?
    2. Why was the error not detected (detection)? Is the process controlled enough?
    3. Why was the error not prevented by designing the process (prevention)? Why was no control operation provided in the process? Do the costs due to this event justify or not the redesign of the process?
    Of course, some of these questions were answered in the comments above, but I wanted to exemplify how I would respond to a possible complaint like this. I have noticed that the answer to a complaint is never simple and involves several root causes (in occurrence, detection, prevention) and the most difficult to identify are the corrective actions that must deal with human actions.
     
    Qualmx likes this.
  10. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    5,086
    Likes Received:
    2,553
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    Welcome LiviuSt.

    How about:

    Why did the part get made wrong? Because the process isn't in robust and control and relies only on Operators to "catch" bad outputs

    Why isn't the process in control? Answer...
     
  11. LiviuSt

    LiviuSt New Member

    Joined:
    Sep 27, 2016
    Messages:
    2
    Likes Received:
    2
    Trophy Points:
    2
    Quick answer - poor process design. If develop there can be a lot of potential causes, which can be eliminated by analyzing this case.
    Were the outputs of process correctly defined? Are these outputs controlled? I think..."employee didn´t stop the motor at the right time" isn't enough to define the control.
    And we can continue with the method and/or resources and/or the staff involved (training, motivation, awareness, back-up).

    Of course, the correct answer can only be given by the team involved in the analysis of the complaint.
     
    Andy Nichols likes this.
  12. Eric Twiname

    Eric Twiname Well-Known Member

    Joined:
    Jul 31, 2015
    Messages:
    329
    Likes Received:
    232
    Trophy Points:
    42
    Location:
    Northeast USA
    Andy is giving good advice about the process.
    I would add (it's sorta in Andy's too) an after process catch for bad parts (in addition to your in-process ones).

    Bad parts get made...we do our best to limit how often, but we never get 100% yield.
    What is in place to identify the bad part(s) before they go out the door?
    Most places use a second pair of eyes/tests on critical features rather than just the operator who made the part...

    It is in your company's best interest to control the process to avoid avoidable errors...it is demanded by your customer that your errors (however few) don't show up at their site.
    Similar, but different, needs. HTH
     
    Qualmx likes this.
  13. Scott@jmnspecialties

    Scott@jmnspecialties Member

    Joined:
    Jun 1, 2020
    Messages:
    7
    Likes Received:
    2
    Trophy Points:
    2
    I would think that the "had a bad night" is not the issue, or an acceptable root cause. I say that, if there is not, there should be another QC before shipping out the finished product. The person making the product should not be the final QC. Production employees "NEVER" produce a defective product.
     
    Qualmx likes this.
  14. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    5,086
    Likes Received:
    2,553
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    Even that won't stop poor product leaving the building! 100% is ineffective and costly. Far better to error proof the process.
     
    Bev D likes this.
  15. TWhite

    TWhite Member

    Joined:
    Jan 12, 2021
    Messages:
    9
    Likes Received:
    6
    Trophy Points:
    2
    Don't dismiss the Human Error factor to quickly. Fatigue is something we are all susceptible to. The more fatigued we become the worse we are with concentration levels and decision making, we forget things and become easily distracted. A once in 5 years problem doesn't mean you have to spend a lot of money into automation or equipment to detect the problem. After all "Detection tolerates waste and Prevention avoids waste" You defiantly want to protect your customer from defective product but a thorough investigation may lead to alternative ways to accomplish the same goal. Spend some time with the operator and see what solutions he might have on how he could have prevented that problem with the situation he was in.

    Speaking from experience I know that not every company will be willing to spend money to upgrade equipment. Understanding Human Error and how to countermeasure it can be a useful tool. I would recommend taking some time to review the work of J.T. Reason and an article by Micheal Venner "Identify Human Factors - 1/14/21"
     
  16. Eric Twiname

    Eric Twiname Well-Known Member

    Joined:
    Jul 31, 2015
    Messages:
    329
    Likes Received:
    232
    Trophy Points:
    42
    Location:
    Northeast USA
    Howdy TWhite,

    I don't think anyone is dismissing Human Error as a factor...simply dismissing it as a Root Cause.
    You are absolutely correct that when humans are involved, human factors (including fatigue and many others) are always at play.

    A robust system will account for these things and (preferred) not let the Human Error happen by prohibiting it physically or (partially effective) identifying bad parts created and quarantining them from going out the door.
    I don't think anyone is talking about dismissing them, quite the opposite...knowing that they are there and important factors to consider, how does the system prevent them from interfering with the process?
     
    RoxaneB likes this.
  17. TWhite

    TWhite Member

    Joined:
    Jan 12, 2021
    Messages:
    9
    Likes Received:
    6
    Trophy Points:
    2
    In the scenario stated by Qualmx the employee had not made this defect in 5 years. I would consider that a robust process. However, due to fatigue (lack of sleep) the incident occurred. ISO 9001 7.1.4 Environment for the operation of process: b) has the e.g. as stress reducing (or fatigue). This is showing that fatigue is recognized as a potential root cause (...achieve conformity of products and services). Now in this scenario the fatigue was not caused at work but fatigue is fatigue. Then in section 8.5.1 Control of production and service provision: g) the implementation of actions to prevent human error. The standard does show that human error is a potential root cause. If it were not, then why implement actions to prevent it?
    Again, after studying Human Error for several years Human Error Factors can be a root cause. Not the human but the factors causing the human to make the error. This is just my opinion and I appreciate your comments above.
     
    Eric Twiname and tony s like this.
  18. Eric Twiname

    Eric Twiname Well-Known Member

    Joined:
    Jul 31, 2015
    Messages:
    329
    Likes Received:
    232
    Trophy Points:
    42
    Location:
    Northeast USA
    Not the human but the factors causing the human to make the error.
    Fair enough, and your opinion is valid, just not commonly agreed.

    Restating your writing with a single word change would make it something I heartily agreed with...

    Not the human but the factors allowing the human to make the error.

    The reason for my belief this way is that it makes it practically actionable by an organization.
    a. No organization can/will address the fatigue of a new parent who was up all night with a newborn.
    b. Any organization can put guards on the dicing machine so the new parent doesn't lose a finger.
    There's no way I'm putting "fatigue" on a causality report for workman's comp, or filing that with a customer.

    Both handle the same situation...one handles it within the organization.
     
    tony s and RoxaneB like this.