Calibration Due date-Critical Connie

Smchandler

Involved In Discussions
I work for a company that is a prime military contractor, and recently my boss suggested removing the day from the calibration label ("May 31, 2021" to "May 2021") to try to improve the calibrations completed on-time metrics, basically trick the people on the floor into thinking its due but not clarifying exactly what day in the month it is due, I am completely against this because I believe there are better ways and the people on the floor need to have a due date, I suggested "EOM" within gagetrak, and debated it by suggesting we review existing gages and try to increase the time between calibration (based on history) to improve the percentage completed on time and reduce overall expenses to calibrations. That wasn't enough for him, what I would need to know is this acceptable per AS9100D or ISO9001 company, would it go against any clauses?
I said it would because our gagetrak system says it is due May 31, 2021 and it doesn't match the printed label says "May 2021", he suggested just writing the procedure to say the label doesn't have to match GageTrak and note it can be done any day of the month. Am I just being over critical or is this acceptable method?
Also any additional recommendations on improving the calibration complete ontime metrics would also be greatly appreciated.
 

Ninja

Looking for Reality
Trusted Information Resource
FWIW...I fall easily on your boss's side.
As much as it sure would be nice to have folks on the floor aware, they can be aware of a month as easily as they can be a day.

What's important here?
1. Gages are accurate and reliable (calibration is part of that)
2. You achieve the goals of the business (reduce overall expenses).
3. other things...

The absolute best way to improve "calibration complete ontime metrics" is to get them calibrated on time...and to define what "on-time" means in the best manner fitting the gage and its use...which may include increasing the calibration period, or may in fact mean shortening it.

Leave Gagetrak out of it...its a tool you use to track stuff...let the needs of the job rule, let tracking it follow.
If GT requires a specific date, simply push all dates to the end of the month.
Whether the valid-through date is "June 30, 2022" or "June 2022"...the next day is too late to use the gage anymore until recal.
If an auditor saw "June 30, 2022" in the software and "June 2022" on the sticker and had an issue...that's what woodsheds are for, to hide what happens next and supply tools for doing it.
(I would write it in the procedure too...takes less time than the woodshed)
 

Smchandler

Involved In Discussions
Great feedback, and sorry to pick on you, but since you are supporting the no day on label, I have to ask what would you think the benefit of having do day on the label would be? Do you think people would not be confused and drag it to at least the last day of the month? So to me why not just make it due the last day. I am lost at what the human benefit factor would be, making the requirement even more vague seems worse to me.
 

Ninja

Looking for Reality
Trusted Information Resource
No worries...questions are how we learn...but don't just learn from me, others here who haven't answered yet may fall on your side and disagree with me (it happens...often).
You ask many questions, let's handle each one.
I have to ask what would you think the benefit of having do day on the label would be?
None whatsoever beyond a visual reminder that calibration is due...just like having the exact day has no value whatsoever beyond a visual reminder that calibration is due. You recognize that it isn't required to have the sticker on there at all, right?
What is required is that the gage is calibrated if you're gonna use it.
You don't think anyone using the gages looks at that sticker, do you?
If there's any value, however miniscule, it's avoiding a stupid finding during an audit for a stray sticker a day past what it states.
Other threads here talk about building in a "grace period" to avoid this same thing.
Do you think people would not be confused and drag it to at least the last day of the month?
I think people will not be confused at all, ... , and drag it to at least the last day of the month...pretty much like they do now, right?
A date on a sticker isn't going to affect human nature.
I am lost at what the human benefit factor would be
I listed the important things about calibration dates...feel free to disagree with them, but I didn't list a human factor.
The gage really doesn't care about humans or what they think.
making the requirement even more vague seems worse to me.
The requirement is that the gage be calibrated if it is used for acceptance criteria...and when it reaches the end of its calibration cycle it doesn't get used until it is recalibrated...that's pretty clear, not vague at all.
I choose to expand that to more like "if this measurement is worth doing, the gage should be calibrated"... If I don't care about it being calibrated, it's because I don't care that the measurement is accurate...so why am I wasting my time?

In my experience, little stickers on everything is the confusing part. We used a database (like GT, but not that) and gave a months warning with a list of gages that would be pulled on XX/XX/XXXX date for recal. On that date, the gages were pulled (or calibrated in place). There was no confusion save for the folks who didn't bother reading what was on the warning list...and they had backups on a staggered cal cycle so it didn't affect their work anyway.

HTH
 
Last edited:

Ninja

Looking for Reality
Trusted Information Resource
True in theory for some gages...but pretty hard to do with many gages. How do you track use of a manual Vernier caliper? I've got a calendar where I can track date pretty easily.
 

Guest

On Holiday
I don't believe you have all the information present to answer the question. Calibration is a function of use. If you have equipment being used - and needing calibration - which isn't under control, such that you know the level of use, then the organization has a bigger issue to fix, first.
 

Ninja

Looking for Reality
Trusted Information Resource
such that you know the level of use
Don't know about you, but I'm in business to make profit.

It costs $35 a year to calibrate a set of handheld mics,
it costs hundreds of dollars a year to track the use of those mics and fight with the guys using them to fill out the use logs and "I have to type in the computer in order to spot check my part?" and the resulting lower morale for forcing them to do something they think is pointless.
I'll pay the $35...I don't see the bigger issue beyond a strive for utopia reaching into my profits.
 

Guest

On Holiday
Also any additional recommendations on improving the calibration complete ontime metrics would also be greatly appreciated.

There are many aspects to running an effective and efficient (that's cost saving) calibration system and it's not much about a specific date of recall of an item - indeed an effective cal system will allow extensions to recalls etc - based on use.

Firstly, experience shows that there's often too many items in the system being calibrated. A significant proportion can be very satisfactorily verified against a know (calibrated) master. This is and has been totally satisfactory - even under AS9100, since the standard specifically permits it.

Are you sure everything in the database actually needs a full calibration (we are talking about recognized calibration with uncertainties here?)

Secondly, and I had some painfully expensive experiences, where gauges were calibrated annually and the useage of the gauges rocketed due to production quantities increasing and the gauges wore out in a few months. So annual calibration was ineffective - and despite the low cost of the calibration, that was nothing to the HUGE costs of recall and remanufacture! I don't know what your situation is but I wouldn't want to go to my boss and say "It only costs $XX to calibrate them annually, but it it costs us $NN,NNN to do a recall because of it"...
 
Top Bottom