Confirmation of software - Exactly what the last paragraph of 7.6 is asking for?

[hokieman]

Can someone please explain in easy to understand language exactly what the last paragraph of 7.6 is asking for? We have several monitoring & measuring devices which are computer driven. Thanks!

 

[BadgerMan]

When software (purchased or internally developed) is used to accept or reject product, its proper functioning needs to be verified both before initial use and periodically thereafter.

 

[antoine.dias]

Calibration of software

I used to work for suppliers to the automotive.

I know Ford had a manual called " Calibration of software ".
A +/- 50 page guide with " exercises " to check software which is used for quality data processing.
The exercises were lists of data you had to use as input. The result given by the software could then be compared with the supposed good result.
It was meant as a " higher standard " for software calibration.

Anyone has a copy to share?

Antoine

 

[antoine.dias]

Calibration/confirmation/verification of software

In another thread this came back again.
Anyone with any information about this topic ?

Antoine

 

[antoine.dias]

Re: Calibration of software

I used to work for suppliers to the automotive.

I know Ford had a manual called " Calibration of software ".
A +/- 50 page guide with " exercises " to check software which is used for quality data processing.
The exercises were lists of data you had to use as input. The result given by the software could then be compared with the supposed good result.
It was meant as a " higher standard " for software calibration.

Anyone has a copy to share?

Antoine

I did not see any reaction on that.
I am reviving this in case someone with a copy missed the post.

I would be very greatful to anyone who can help here ( I does not have to be the FORD manual ).

Best regards,

Antoine

 

[Sebastian]

My experience is related to simple MS Excell sheet, I made to calculate Cp, Cpk for initial samples (PPAP).
Base was QS SPC manual with its calculations for this matter. Everything was fine, until I gave for auditor's question: "How do you determine this indicators?" simple answer: "I made MS Excell sheet according to QS SPC manual."

For next question: "How do you know, results are properly calculated?", I could not find right answer and it finished with major NC.
As proper action in this case, we found use of input values presented in this manual and comparisation to calculation results presented also in this manual for same input values.

I hope it helps.

 

[atitheya]

As proper action in this case, we found use of input values presented in this manual and comparisation to calculation results presented also in this manual for same input values.

The last para in the clause 7.6 means exactly the same, to test the software wether it is giving the results it is supposed to. These tests can be done against standard and known values, where inputs and output value is already available. This is very common with medical instruments, where the computerised machines are tested against a standard input, say a reagant of known value and the output values are compared since they are already available. If the machine is giving results within the range, it is working fine else it needs to be attended to.

Similarly when working with spreadsheet formulae, one can test it with known figures (both input and output) and compare the results to confirm the working of the computer / computerised machine.

Hope this helps to explain