Is Failure to achieve a Goal a Major Non-Conformance?

[AndyN]

My point is, Jennifer, that a blanket 30 days isn't appropriate for ALL CAs! Some should be done within 24 hours, if they warrant it. Sure, that's within 30 days, but what about those which don't have a timetable of fewer than 30 days? Design related? Customer related? As with many things in life, doing something to a calendar isn't an appropriate method to show how effective CA is. They could be closing is 100% within 30 days, but they were 20 days late!

 

[Jen Kirley]

My point is, Jennifer, that a blanket 30 days isn't appropriate for ALL CAs! Some should be done within 24 hours, if they warrant it. Sure, that's within 30 days, but what about those which don't have a timetable of fewer than 30 days? Design related? Customer related? As with many things in life, doing something to a calendar isn't an appropriate method to show how effective CA is. They could be closing is 100% within 30 days, but they were 20 days late!

Oh I am sorry I misunderstood you, yes I totally agree.

I have had to defend a year-old open CA with a TS 16949 auditor. I did it by explaining its containment, the fact that it needed some rather extensive resources to fix the scheduling bugs in the app, and a string of notes in my corrective action tracking log what had been verified done on what day, meetings and who committed to what, reaudit results and following actions, etc. He gave up and moved on.

 

[Kevin Mader]

Something to guard against: extensions on extensions on extensions.

In years past, CA closures were prolonged (resource constraints, interest...the usual suspects). It got easy for folks to ask for an extension and then close something out a day or two later. Metrics improved but of course (CAPAs done within our arbitrarily set 30 days, yada-yada) but didn't tell the whole story. I ended up creating a second metric to track the number of extensions given. This curbed padding a number while allowing another to run away. Sad it had to happen that way.

OK - where's my soap box?

Deming warned about setting Goals without a Method. Lloyd Nelson also warned that the most important numbers are unknown and unknowable. Both agreed that abstaining from managing a process because a 'hard' number could not be established was not an answer either. When establishing 'hard' numbers, folks must be intimate with the process, understand control limits and process controls before setting them. Otherwise, numbers are arbitrary and usually end up causing more trouble than you would have had without them. Managing a process, CA (CAPA) or otherwise, requires theory and knowledge of the process. Statistically speaking, is there a difference between 75% and 85%? Examining the process closely might reveal that either exist between the control limits - no difference (each had the equal chance to appear). What happens is that folks kick off the new year with new 'numbers'. Business folks are infamous for doing this. What's needed is appropriate goal setting under the conditions Deming and Nelson suggest. In a nut shell, forget the goal setting and manage the process. Show auditors this kind of data, this level of intimacy, and you won't have to argue much at a closing meeting

Good luck everyone!

 

[Sidney Vianna]

Could an auditor consider this a Major non-conformance?

I will assume you are talking about an external (CB) auditor type.

In addition to the great advice you have already received, especially the one that questions the adequacy of setting a goal such as this, I would say: If I were an external auditor assessing this issue, I would focus on top management and their involvement in the management review, as "status of corrective actions" is a mandatory item to be reviewed. I would interview top management and ascertain if they are aware of the situation and their knowledge of the reasons on why, apparently, closure of corrective actions take longer than you anticipate.

But, without a question, pre-defined set times for CAR closure is not a wise approach.