What specifically must be calibrated per ISO 9001:2000?

[Marc]

The important thing is (and this has been the same for years) you have to be able and ready to justify when (frequency) and why you calibrate and/or "verify" a measurement device, and it has to include the type of use, the frequency of use, the effects of it not being in calibration (such as dropping the device between calibrations and/or verifications), and things like that.

While the word risk wasn't much used in the past, these days it is especially with respect to (for example) what if someone drops the measurement device between calibrations and a lot of product passes that is not in spec. Even though the word risk wasn't much used, it was taken into consideration "where appropriate". I have seen quite a few scenarios where a measurement device would be verified at the start of each run and re-verified at the end of the run.

This was very common in semiconductor and electronic assemblies. You'd run your "Golden Masters" at the start of each run and you'd run them again at the end of each run.

 

[Sidney Vianna]

While the word risk wasn't much used in the past, these days it is especially with respect to (for example) what if someone drops the measurement device between calibrations and a lot of product passes that is not in spec. Even though the word risk wasn't much used, it was taken into consideration "where appropriate". I have seen quite a few scenarios where a measurement device would be verified at the start of each run and re-verified at the end of the run.

That's exactly right. The people who erroneously claims that the previous versions of ISO 9001 did not have "informal risk evaluation" somehow embedded in the standard never truly understood the intent of the standard in the first place.

 

[Marc]

I don't remember too much "formal", documented risk assessment in low risk processes. It was all done during the quality planning phase (not to mention DFMEAs during the design phase), usually in team meetings. Usually you'd find the meeting minutes had content that discussion was held.

But then again, in quality planning for airbags (as an example) we did do FMEAs which would be in the product file. Any time it was a high risk process any risk analysis performed was documented (often multiple FMEAs for different process segments). Semiconductor and electronic assemblies an FMEA or other documented risk analysis was simply done as part of the "standard" process.

This is why I have said so many times - Most companies are already doing appropriate risk analysis. It isn't new any more than "interaction of processes" was.

 

[Ragnarok]

Hello to all

please i have 2 questions

1) If we have attendance machines do we need to calibrate them too?

2) Are software that show average response time, order amounts etc ... in need of calibration and considered measurement resources?

Thanks for the help

 

[itsbiodiversity]

Hello to all

please i have 2 questions

1) If we have attendance machines do we need to calibrate them too?

2) Are software that show average response time, order amounts etc ... in need of calibration and considered measurement resources?

Thanks for the help

I believe I can chime in here.

Attendance machines (for employees I am assuming) would only require calibration under a couple of scenarios that I can think of. One would be if a project specifies, for example, 10 employees perform a certain task or some sort of specification involving either number, type, etc. of employee.

Software that shows average response time of a customer service team would not be calibrated; however, average response time of an electrical meter would be calibrated (depending on the use). If the software is used to maintain compliance (such as a control chart, etc.) you would want to run a validation (unless purchased, off the shelf, no mods).

Please let us know we can help more!!! And for the other pro - fessionals out there please feel free to correct me. Peace.