IATF 16949 Clause 10.2.4 - Error Proofing

jfriess

Involved In Discussions
What if your only error proofing is a device(with PLC) that accepts/ rejects parts based on temperature? The gages are calibrated and verified daily to ensure that they are accepting rejecting parts based on the limits that are set. It is not possible to have "rabbits".
 

Bran

Involved In Discussions
Per IATF 16949:2016, 3.1 Terms and definitions for the automotive industry

Page 12: Error proofing

Product and manufacturing process design and development to prevent manufacture of nonconforming products.

I'm finding "Process design that prevents manufacture of nonconforming product" a difficult concept. I can think of many process design elements that will reduce occurrence of nonconforming product, but none that will truly prevent it. Does anyone have any ideas?

A past BIQS auditor suggested I interpret "error-proofing" as:
  • Auto Shutdown
  • Auto measurements or Automatic data collection points
  • Auto QMR
  • Not auto-advancing
  • Intentional System Delays
Problem with this list is that none of these actually meet the IATF definition of "preventing manufacture". The system I implemented mirrors our EHS "error proofing" (snip below), but reading the IATF definition I think this is stretching the requirement to be more than is intended?

Should automated systems that reduce the occurrence of nonconforming product be considered "error proofing"? How would EP devices like this be tested for a simulated failure? They rely on programming which is validated prior to implementation

1578083214828.png
 
Last edited:

Miner

Forum Moderator
Leader
Admin
Say you have a component that can be installed backwards. You could change the design to be 100% symmetrical so that it no longer matters, or you could change the design such that it physically cannot be assembled backwards.
 

Bran

Involved In Discussions
Specifically, I am in an organization that produces bulk material and does no product design
 

Mikey324

Quite Involved in Discussions
Re: IATF 16949 Clause 10.2.4 - Error-proofing question

Error-proofing device failures shall have a reaction plan.

Do they mean Mistake proofing rather than error proofing?

This just means how will you react if your error proofing device fails. If it can't detect a bad part, what will you do? 100% sorting? Increased visual inspection? Things like that.
 

brandieb1230

Involved In Discussions
We recently received a non-conformance on clause 10.2.4 as we didn't have proof that our rabbit was certified. We build parts that have a decent amount of content - rabbits are built to test cameras on line for content and color. We told them it is certified because it faults at various stations and doesn't fault at others. That was not acceptable.

How do you calibrate your rabbits or challenge parts?
 

Mikey324

Quite Involved in Discussions
We recently received a non-conformance on clause 10.2.4 as we didn't have proof that our rabbit was certified. We build parts that have a decent amount of content - rabbits are built to test cameras on line for content and color. We told them it is certified because it faults at various stations and doesn't fault at others. That was not acceptable.

How do you calibrate your rabbits or challenge parts?



What was the nonconformance? 10.2.4 doesn't say anything about the need to be "certified". It says:

"The process shall include the testing of error-proofing devices for failure or simulated failure (your rabbits). Records shall be maintained (do you have these?). Challenge parts, when used, shall be identified, controlled, verified, and calibrated where feasible. Error-proofing device failures shall have a reaction plan. (do you have this?)''


Calibrated where feasible. It sounds to me like that wouldn't be feasible in your case. My questions would be do you have records of testing them? You do have to test your method, whatever it may be. You want to make sure a failure is actually flagged as such. If you do have records of testing them, I'm not sure how the NC was justified by the auditor.

Maybe you can share the finding as it was written? the statement of NC and the requirement it failed to meet?
 

Sebastian

Trusted Information Resource
Certified, controlled, inspected, managed, verified, released, calibrated - whatever.

Challenge part - who made it, how it was made, when it was made, how it was verified, when it was verified , who test it, what was result, are there release records? These are sample questions to verify, how it is managed internally.

I was working for company when challenge parts have "best before date" defined.
Reason was, sample part modified initially to be NOK and used to verify test machine performing 100% iproduct nspection, after some time transform into OK part. Period was defined according to analysis of consecutive measurements of challenge part made before this limitation was implemented.
 

brandieb1230

Involved In Discussions
What was the nonconformance? 10.2.4 doesn't say anything about the need to be "certified". It says:

"The process shall include the testing of error-proofing devices for failure or simulated failure (your rabbits). Records shall be maintained (do you have these?). Challenge parts, when used, shall be identified, controlled, verified, and calibrated where feasible. Error-proofing device failures shall have a reaction plan. (do you have this?)''


Calibrated where feasible. It sounds to me like that wouldn't be feasible in your case. My questions would be do you have records of testing them? You do have to test your method, whatever it may be. You want to make sure a failure is actually flagged as such. If you do have records of testing them, I'm not sure how the NC was justified by the auditor.

Maybe you can share the finding as it was written? the statement of NC and the requirement it failed to meet?


Thanks for your assistance:

Statement of NC: The system for calibrating/certifying "rabbits" used to test error-proofing process is not effective.
Requirement: The organization shall have a documented process to determine the use of appropriate error- proofing methodologies. Details of the method used shall be documented in the process risk analysist (such as FMEA) and test frequencies shall be documented in the control plan> The process shall include the testing or error-proofing devices for failure or simulated failure. Records shall be maintained. Challenge parts, when used, shall be identified, controlled, verified, and calibrated where feasible. Error-proofing device failures shall have a reaction plan.
Objective Evidence: There is no evidence that "rabbits" used to verify automated, in process inspections are verified/ certified against standard.

Our reaction plan is set up to follow help chain- contact controls and supervisor.
The assembly lines are set up where the rabbit part is required to run within the first hour of production- if it doesn't run the line shuts down.
Emails are sent out to various plant leadership that shows the rabbit ran and would notate any failure. - we do not have any proof of testing them besides this email.
 
Top Bottom