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.

Subjectivity of Occurrence and Detection ratings

Discussion in 'FMEA - Failure Modes and Effects Analysis' started by Aalekh Prajapati, Oct 17, 2019.

?

Do you think FMEA Occurrence and detection rating criteria has a lot of subjectivity ?

  1. Yes

    1 vote(s)
    50.0%
  2. No

    1 vote(s)
    50.0%
  1. Aalekh Prajapati

    Aalekh Prajapati Member

    Joined:
    Mar 15, 2019
    Messages:
    7
    Likes Received:
    0
    Trophy Points:
    1
    During creation of DFMEA, I feel there is a lot of subjectivity in Occurrence and Detection rating criteria given in FMEA 4th as well as newly released AIAG-VDA manual.

    example of subjectivity:
    Detection rating 4 & 5.


    4 - Moderately high
    : Ability of detection control to detection the failure cause of failure mode is moderately high based on verification or validation procedure, sample size, mission profile. etc.

    5 - Moderate: Ability of detection control to detection the failure cause of failure mode is moderate based on verification or validation procedure, sample size, mission profile. etc.

    How can someone categorize any specific failure/cause between these 2. How to choose if a failure mode's detection is Moderate or Moderately High.

    Also, definition of my moderate and some other person's moderate may differ. Similarly, occasional vs frequent, these 2 words may have different definition for each person.

    I would like to know the views/opinion of people here on the forum.

    I may be wrong. If you think I have an incorrect understanding, please feel free to correct me.

    I am open to have one-to-one discussion as well.

    Thanks
     
  2. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    3,710
    Likes Received:
    1,879
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    You are missing something - data.
     
  3. Aalekh Prajapati

    Aalekh Prajapati Member

    Joined:
    Mar 15, 2019
    Messages:
    7
    Likes Received:
    0
    Trophy Points:
    1
    Thanks for your response. Please elaborate.
    But, in order to arrange or categorize the data I have, I need to understand the difference between each. since, it is a DFMEA, I have the list of different tests at each stage of design flow, but which rating is applicable to which test, that is what I am willing to understand.

    FEA analysis, virtual validation tests, calculations, tests, studies and many more.which one should be kept in which rating block ?
     
  4. Bev D

    Bev D Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    472
    Likes Received:
    513
    Trophy Points:
    92
    Location:
    Maine
    You may find my paper on FMEA and the RPN helpful to begin your journey of understanding. It is in the Resources tab under practical quality engineering resources.

    Specifically the only way for you to have objectivity in detection and occurrence is to do the test. MSA for detection (you can see some of my MSA papers in the same area) and for occurrence you need to design an appropriate study (sample size and conditions) to determine the occurrence rate.

    THE fundamental flaw of the traditional FMEA is the subjectivity of the occurrence and detection values. NO ONE ever gets this right. (a quick read on this:
    The Failure of Risk Management and How to Fix It

    After this flaw the RPN ratings are ordinal data that violate almost all requirements necessary to 'do math'.

    I have changed my organizations use of FMEA to see severity as the only input and the only actionable number. Occurrence is an output (if you don't take the right preventive actions and mitigations your occurrence will be high; if you get it right your occurrence will be low). Detection is one of several mitigations. Detection is also critical to ensure that your V&V testing is adequate to catch any failures...
     
    tony s, Aalekh Prajapati and Miner like this.

Share This Page