Re: Identification for 7.6
I agree in principle that there's no way around the "specified intervals" requirement, and that auditors have to call 'em as they see 'em. The problem with the whole thing is that the calibration interval, regardless of accounting for risk, will always be arbitrary at some level. No one knows when or if anything will need to be calibrated or adjusted and when something does, the need almost never has anything to do with chronology. For this reason, too much emphasis is placed on dates and not enough on actual measurement issues, and too many meaningless CAs are done for things being beyond their dates when being beyond the date hasn't been shown to affect anything. This is not the fault of auditors, though--if the organization has broken its own rule, it's a nonconformity that should be documented.
As I suggested earlier, devices should always be verified when measurements are close to spec limits or when spec limits are very tight. If anyone thinks that stickers are helpful, stickers should be used.
I'm going to have to disagree with you on this one. If the organization specified the calibration/verification intervals based on risk then there's a good reason to write a nonconformity if the gages aren't recalled by their due date. It's exceeded the organization's own risk criteria. The organization determined that to go beyond that date was a risk they were unwilling to accept.
Now if the organization specified the intervals based on nonsensical reasons, then it's not the auditor's fault that there's no value in the finding. You suggest that the auditor's job is to identify the risks to the organization. I say that it's the organization's responsibility to identify their risks. If there's no value, the responsibility rests squarely on the organization.
Now if the organization specified the intervals based on nonsensical reasons, then it's not the auditor's fault that there's no value in the finding. You suggest that the auditor's job is to identify the risks to the organization. I say that it's the organization's responsibility to identify their risks. If there's no value, the responsibility rests squarely on the organization.
As I suggested earlier, devices should always be verified when measurements are close to spec limits or when spec limits are very tight. If anyone thinks that stickers are helpful, stickers should be used.
