Up to this point we only record the information such as ser#, controlled gage# during tests on our reports if the customer requires it. Recently during an audit the question was asked about how we investigate past tests if we determine a gage is out of calibration, and if customers are notified or finished product is recalled. We have not had the need to do this yet, no gages have been found out of tolerance enough to change the end quality of the product. Do most operations always record the gage# on all test reports to allow for a retracing if the gage is found out of calibration?

7.6 refers to taking action on any product affected.

I'm going to go out on a limb and say that IMO most places don't record the gage number on the test records. The decision to do so should be a based on the risks associated with failure (and any customer/industry requirements). Like any traceability, much of the time you don't need it, but if you end up having a recall, you really wish you had it. If you have a gage way out of tolerance, you might end up contacting more customers because of lack of data...

On a (sort of) related note, I heard on the news today that they are slaughtering 450 calves in the herd with one mad cow because of lack of traceability on them. It could have been two or three if they had better records.

I have had the opposite experience, most places I see (high tech/defense) require the Cal control ID on the lot traveler or test record, unless only 1 instrument exhists in house. Or a log of items tested by date/instrument/equipment

Like any traceability, much of the time you don't need it, but if you end up having a recall, you really wish you had it. If you have a gage way out of tolerance, you might end up contacting more customers because of lack of data...

On a (sort of) related note, I heard on the news today that they are slaughtering 450 calves in the herd with one mad cow because of lack of traceability on them. It could have been two or three if they had better records.Much of agribusiness, liike many other businesses (fasteners, printer ink, athletic shoes, etc.), depends on "fungibility" [dollar bills and bushels of grain are both fungible, meaning one may be exchanged for another without penalty.] The dollar bill is traceable by virtue of its serial number. This extent of traceability on $1.00 bills may never be of value or use except to folks who play "Liar's Poker," but it certainly comes in handy on larger denominations in criminal investigations.

The current cost of traceability on agricultural products (meat, grains, vegetables) is still high, but when the products are aggregated, such as grain into flour, avocados and oranges or tomatoes into guacamole or juice, cattle into hamburgers, the question of traceability gets lost along with questions such as
"How much rodent feces is acceptable in a carload of grain?"
"How much gristle in a can of stew?"

There's a whole school of thought which decries the American fetish for disinfectant hand wipes, saying the practice fosters the evolution of "super bacteria."

I don't have answers, but I sure have questions. This is definitely a case of "the more you know, the more you worry."

Hoosierken,

We have tools assigned to various departments and they are only used in that department so we know what product was measured with what tool. We are a small organization and this works well for us.

As for taking action on any product affected, we simply refer to our Handling Nonconforming Product procedure in the Calibration procedure. Then if we have a tool that is out of calibration, we inspect the product, determine the extent of the nonconformance (if any) and manage it according to the definition in Handling Nonconforming Products.

mshell

Up to this point we only record the information such as ser#, controlled gage# during tests on our reports if the customer requires it. Recently during an audit the question was asked about how we investigate past tests if we determine a gage is out of calibration, and if customers are notified or finished product is recalled. We have not had the need to do this yet, no gages have been found out of tolerance enough to change the end quality of the product. Do most operations always record the gage# on all test reports to allow for a retracing if the gage is found out of calibration?

7.6 refers to taking action on any product affected.

First, Welcome to the Cove! :bigwave: It's always good to have another Hoosier aboard.

Now to your query. I would have to go along with Howste and say that you need to due a risk analysis, and determine if it is worth the time and effort to minimize the amount of material/product that would have to be recalled (should the need arise - and the fact that you have never had that situation arise thus far should be a factor in that analysis I believe). With the addition of the specific guage information, you could limit your recall to a specific product/customer/date frame, without that information you may have need to recall a larger portion of product from a more diverse customer base and time frame - but your company needs to make the call as to whether the risk is worth it or not.

Thanks very much for the suggestions so far. Allow me to be more specific regarding our business. We are a manufacturer and repair shop of Electric Motors. We have 85 employees and run 3 shifts. Many controlled measuring devices such as handheld digital multimeters are controlled but are used all over the plant. It would be nearly impossible to trace the jobs that a specific meter was used on, but I feel we have a hole in our system with the "what if" situation. Would the simplest more cost effective way to be recording the meter gage # right beside the measurement on the report? I'm really at a loss here and higher management simply places the responsibility on my desk with the caveat of no more cost. :bonk:


:thanx:

Thanks very much for the suggestions so far. Allow me to be more specific regarding our business. We are a manufacturer and repair shop of Electric Motors. We have 85 employees and run 3 shifts. Many controlled measuring devices such as handheld digital multimeters are controlled but are used all over the plant. It would be nearly impossible to trace the jobs that a specific meter was used on, but I feel we have a hole in our system with the "what if" situation. Would the simplest more cost effective way to be recording the meter gage # right beside the measurement on the report? I'm really at a loss here and higher management simply places the responsibility on my desk with the caveat of no more cost. :bonk:


:thanx:

Hoosierken,

At a previous employer's (aerospace/military) we did just what you have suggested - entered the specific gage # beside the measurement (or elsewhere) on the report (in-fact our "reports" had a section for noting the specific type/model number/manufacturer/serial number, or in most instances just the specific gage traceability/inventory number itself would suffice).

Then as long as the reports are maintained and are traceable to the specific product/item tested, you have your gage traceability.

Looks like your on the right track. :cool: Good luck.

I am not disputing the statements above but I am wondering if you don't record the gage number on the test result how can you establish traceability to a National Standard? There may be no requirement to record the number but there is a requirement to establish the traceability. Without the gage number on the test it seems you are saying "trust me, we would never use a non-calibrated gage". We have an area for gage id and tester/operator on every report we generate. The raw data sheet for any test also has the same information.

Dave