iSixSigma

Use of Six Sigma Tools with Discrete Attribute Data (Pass/Fail)/FMEA

Six Sigma – iSixSigma Forums General Forums Tools & Templates Use of Six Sigma Tools with Discrete Attribute Data (Pass/Fail)/FMEA

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
    Posts
  • #250919

    Marc68
    Participant

    Hello.

    I ‘m currently running a project within my company  (Pharmaceutical company, Research and development) which is the change of the Preventive Maintenance/Calibration intervals for analytical instruments.

    Usually the PM/calibration interval is 12 months for analytical instruments.

    Based on FMEA (Failure Mode and Effect Analysis) assessments, we will try to change to 18 or 24 months the PM/calibration intervals, based on the final risk defined (Probability x Risk x detectability). Low risk would lead to a 24 month interval.

    For the probability that a failure could happen, we checked the number of failures which happened during the last 4 years (2017 to 2020) on the same type of system (e.g : HPLC).

    The data is pass or fail and the question is how to determine the maximum number of failures to assess the probability level to “Low”.

    Per example, we discovered 3 failures over the last 80 calibrations (last 4 years for 20 systems).

    I ‘m a bit lost with statistics, it is not my domain of expertise, but maybe someone here could help me to clarify this.

    Thank you in advance.

    0
    #250922

    Chris Seider
    Participant

    Please use data for your FMEA ratings.

    Also, many pieces of equipment have recommended calibration frequencies that would be HARD to ignore.  However, have you considered use of SPC for monitoring the measurement systems.  This is a great tool to signal when calibrations are needed.

    0
    #250995

    Eric Maass
    Participant

    Hi Marc68

    You could consider Preventative Maintenance for analytical instruments in the pharmaceutical industry to be a tradeoff between Risk and Cost. The statistical analysis comes in to play as a way to balance Risk and Cost.

    Taking these to an extreme – if Risk was “high” and PM/Calibration was “free” (cost was “extremely low”), if three was high risk and PM / Calibration took zero time and cost no money, then you could do PM/Calibration incredibly frequently – daily, hourly, whatever.

    If Risk was pretty much negligible and PM/Calibration was incredibly expensive, then you might move the PM / Calibration frequency to years and years – perhaps the 2 years in your question.

    So, my suggestion would be to first create a financial model for the cost and time required. This should be fairly simple.

    Then, create a model for the risks – there are two types of risks, Alpha Risk (the “Producer’s Risk”, the risk of overreacting, in this case PM/Calibrating too often) and Beta Risk (the “Consumer’s Risk”, the risk of underreacting, in this case the risk of missing a shift in calibration that impacts your customers).

    Your MSA (Measurement System Analysis) could provide information for the risk model.

    Lastly, you could combine the financial model and the risk model to propose a frequency of PM/ Calibration.

    Best regards,

    Eric

     

    0
    #251046

    MBBinWI
    Participant

    @Marc68 – I’m going to take exception to @cseider and @poetengineer in their guidance on this issue.  When determining calibration frequency there are two items to evaluate – the bias and precision.  While an MSA can help, it also introduces issues related to the human element, but that is not what calibration is about.  Calibration is about the ability of the instrument itself (regardless of human interaction) being able to report a valid value and report with a level of precision required.  In order to evaluate this, you should periodically use a known value specimen to evaluate your measurement device.  Does it continue to report an accurate value to the level of precision required?

    Typically, calibration is a function of mechanical processes.  Wear and physical changes of the measurement mechanism.  In systems where there is little of this phenomenon, the calibration period should be set appropriately.  This is more a function of physics and mathematics than it is of risk (FMEA).  The setting of risk level acceptable can help to establish the values to be used, but not the calibration frequency.  That is a function of physics/math.

    Just my humble opinion.

    0
Viewing 4 posts - 1 through 4 (of 4 total)

You must be logged in to reply to this topic.