1. Hello and Welcome to The Quality Forum Online...Continuing in the spirit of People Helping People !
    Dismiss Notice
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.

"production -related software verification..." (7.1.5.2.1-i)

Discussion in 'IATF 16949:2016 - Automotive Quality Systems' started by John C. Abnet, Nov 20, 2017.

Tags:
  1. John C. Abnet

    John C. Abnet Well-Known Member

    Joined:
    May 23, 2017
    Messages:
    106
    Likes Received:
    49
    Trophy Points:
    27
    Good day all-
    We would like some input from this group in regards to interpretation of this specific clause line item.

    Here our current TWO interpretations...

    1- This applies when, within a manufacturing organization ("production-related")
    that organization verifies software ("...software verification...") as part product and/or process
    control.

    In other-words, IF the organization is verifying software.



    2- This applies when, within a manufacturing organization ("production-related") software is used to verify product and/or process control (e.g. on a press, the software that provides "thermostat" to control and record melt temperatures; the software that interprets a laser being used for length monitoring, etc... )

    In other-words, this applies to the software on most automated pieces of direct manufacturing equipment in a modern facility and the organization must arrange "calibration/verification" of all equipment controllers, which are governed by software.


    Thanks in advance for your feedback regarding this topic.
     
  2. Serious Man

    Serious Man Active Member

    Joined:
    Oct 24, 2017
    Messages:
    50
    Likes Received:
    15
    Trophy Points:
    7
    Temperature control system
    1. Hardware - thermocouple
    We know temperature and confirm thermocouple output voltage.
    2. Software - steering program of PLC
    We know input voltage and confirm PLC output signal steering heater.
    Program has criteria telling PLC software how to handle certain input values.
    Of course PLC is a piece of hardware too, but without program telling its software what to do it's a total zero.
     
  3. John C. Abnet

    John C. Abnet Well-Known Member

    Joined:
    May 23, 2017
    Messages:
    106
    Likes Received:
    49
    Trophy Points:
    27
    Thanks for the feedback "serious man". Can I assume the "steering program" and "thermocouple" are items you produce for sale? (or are these items that verify inputs and/or outputs?)

    I'd be curious if others have thoughts/examples on this as well.

    Thank you.
     
  4. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    1,965
    Likes Received:
    1,023
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    In my experience, software is treated in 2 ways:

    Checks (verification) to ensure that the software does what it's supposed to do, that is passes good product (process) and fails out of spec product.

    Controls to ensure the correct version is being used (treat is as work instructions for a machine, not people)

    Otherwise it's a design output (product) and falls under the design controls (good luck with THAT)!
     
  5. John C. Abnet

    John C. Abnet Well-Known Member

    Joined:
    May 23, 2017
    Messages:
    106
    Likes Received:
    49
    Trophy Points:
    27
    Greatly appreciate your feedback Andy-glad you jumped in.

    Based on your past experience, let me provide a specific example...
    e.g. A plastic injection molding press has a built in controller, which verifies...
    - Pressure
    - Dwell (hold pressure time)
    - Barrel temperature
    - Cavity Temperature
    - Injection velocity
    - Clamp force
    - Screw speed
    ....etc..etc...

    Is this clause implying that the software used by this controller needs verified? This seems unreasonable/not feasible.

    I'd appreciate your thoughts on this example.

    Thank you.
     
    Last edited: Nov 22, 2017
  6. Serious Man

    Serious Man Active Member

    Joined:
    Oct 24, 2017
    Messages:
    50
    Likes Received:
    15
    Trophy Points:
    7
    I meant, it is example of complete control system included in control plan for manufacturing process where temperature is a process parameter, e.g. melting.
    System consisting of hardware and software elements, which both have to be verified independently.
     
  7. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    1,965
    Likes Received:
    1,023
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    I'd treat them simply as "work instructions" for the controller. At some stage (and under a known version of controller firmware) parts were verified as being to specification (there was a production run which was checked for dimensions, damage/burning, shot shots, flash and so on) which could be used as evidence of verification. After that any changes to the controller instructions should also be considered for re-verification.
     
  8. John C. Abnet

    John C. Abnet Well-Known Member

    Joined:
    May 23, 2017
    Messages:
    106
    Likes Received:
    49
    Trophy Points:
    27
    Seems to be a sound and viable approach. Sincere thanks for your input on this topic.

    Is it safe to assume then, that our interpretation "1" (as follows) is the intent of the clause?...
    1- This applies when, within a manufacturing organization ("production-related"),
    that organization verifies software ("...software verification...") as part product and/or process control.

    In other-words, IF the organization is verifying software.
     
  9. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    1,965
    Likes Received:
    1,023
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    I'm not exactly certain what you're asking, John, however, since parts are going to be PPAPed, then this process will, in effect, be a validation of ALL the process control requirements (isn't it?) which includes instructions for whatever controls the manufacturing process...
     
  10. hogheavenfarm

    hogheavenfarm Well-Known Member

    Joined:
    Jul 30, 2015
    Messages:
    198
    Likes Received:
    136
    Trophy Points:
    42
    Maybe a different example -

    A cnc program to control a waterjet machine generated from a dxf drawing. Is the part cut by the waterjet the correct size? How do you know? Many factors come into play here, cutting nozzle offset, height, taper, etc. We conduct an FPI on the first part to determine that yes, the dxf is correct and the programming produced the part we want. I then call the "program", which is the entire cutting process, "validated" for that particular setup and part.
     
  11. Serious Man

    Serious Man Active Member

    Joined:
    Oct 24, 2017
    Messages:
    50
    Likes Received:
    15
    Trophy Points:
    7
    Requirement is very simple - keep in mind that not only hardware can mislead you, but software too.
    So if hardware needs verification/calibration using master, software needs it too.

    So what is a software master?
    It is set of input data combined with output result.
    Software calibration is procedure of running software on "master input data" and comparing achieved result to "master output result".

    Do not mix here later activities like product/process validation, PPAP, anything related to process output.
    You have to have reliable control tools in our hand to start thinking about product/process validation.
     
  12. Andy Nichols

    Andy Nichols Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    1,965
    Likes Received:
    1,023
    Trophy Points:
    112
    Location:
    In the "Rust Belt"
    I'd caution using such terms to describe software - particularly in this context. For a start, the whole area of control of measuring equipment is fraught with mis-understanding. Auditors frequently don't have much idea and organizations mis-label the term. Let's keep software controls to verification and validation only. All they are is instructions. To treat them as something akin to calibration is, in my experience, very dangerous...
     
    Last edited: Nov 24, 2017
    John C. Abnet likes this.

Share This Page