ISO 9001:2000 Clause 7.5.2 Validation of Processes - Level II Response

[howste]

IMO destructive testing is product testing, which can be used to validate the process.

I'm just having a hard time thinking of an example where there is no way (destructive or otherwise) to verify the output of a process.

How do you verify that an airbag or a hand grenade is going to function as intended?

 

[Rachel]

How do you verify that an airbag or a hand grenade is going to function as intended?

Point taken...but can that logic not be applied to many situations?

For example...we have gasketing operations in our facility. We make the gasketing compound; in some cases we sell the compound to companies for their own use, and in some cases the customers supply us with parts and we apply the gasket for them.

Now, we have no way of verifying that any one gasketed part is going to function as intended, but we do have material specs that give us a good indication of the hardness/curing time/etc. of the material. That, and the products are dedsigned and developed to perform in certain environments (extreme cold/heat, moisture, etc.). We may test certain specs on the completed gaskets, like hardness and height, but we don't necessarily immerse a pair of mated parts in water to see if the set leaks. So, we're testing the product, not the process. Right?

(Does anyone know what I'm asking here?...because I'm not sure *I* am at this point!)

I guess I'm just having a hard time distinguishing between testing the product and the process - because in many cases, you test the process by achieving the required product specs.

Am I out to lunch here?

 

[Randy Stewart]

Don't Over Think It!

Rachel,
We do fastener plating here. We check our paint samples each day. If my paint viscosity is correct and the sample parts show proper plating thickness, then the parts being run on that line are considered "good".
Our processes are designed to meet a certain spec (GM, Ford, etc.) and salt spray test requirements. Once the parameters are set (spec requirements), my inputs are verified (viscosity, spin time, bake time, etc.) and my output of the process is checked (samples). My process is set. I don't need to test every piece off the line, nor do I need to do salt spray testing on each part/product.
You are doing the same. You know that given a certain weight, height and hardness, that the gasket is suppose to meet a certain performance criteria.

 

[Todd_w]

7.5.2 for chemical industry

I work in the chemical industry. I CAN do destructive testing on a sample of each shipment. In my mind, if the testing meets the customer’s specifications, both the process and product have been validated. The clause reads “…where the resulting output cannot be verified”. I don’t see how this applies to my company since I can validate a representative aliquot of a homogenous mixture . The only thing I can’t guarantee is the condition of the product upon receipt by the customer. Generally, product quality complaints resulting from the production process are very low. Packaging, delivery and poor condition of the product due to transportation are our major customer complaints.

Please correct me if I’m not looking at this correctly.

 

[Douglas E. Purdy]

It has been a week - You must be correct!

I work in the chemical industry. I CAN do destructive testing on a sample of each shipment. In my mind, if the testing meets the customer’s specifications, both the process and product have been validated. The clause reads “…where the resulting output cannot be verified”. I don’t see how this applies to my company since I can validate a representative aliquot of a homogenous mixture . The only thing I can’t guarantee is the condition of the product upon receipt by the customer. Generally, product quality complaints resulting from the production process are very low. Packaging, delivery and poor condition of the product due to transportation are our major customer complaints.

Please correct me if I’m not looking at this correctly.

Todd,

I am not the one to correct you! I am searching the threads to get a better understanding of revalidation in 7.5.2, and I have hardly any chemistry in my background, but with your verification - how do you know the remainder of the product would pass the destructive test? Isn't that why it is a validation activity as oppose to verification?

Just Pondering the Difference!

Doug

 

[Paul Simpson]

Why validate?

I thought we had killed off this discussion in an earlier thread.

I'm not sure how to read this requirement, at all. I struggle with the following sentence under:
7.5.2 Validation of processes for production and service provision

"The organization shall validate any processes for production and service provision where the resulting output cannot be verified by subsequent monitoring or measurement."

A bit of background: “Validation of processes” has its origins in the special processes clause of BS 5750 when it first came out. The element required an organisation to consider manufacturing processes that could not be fully assessed by inspection and testing. There are some good examples above including welding, heat treatment, painting, bonding and plating. The principle for each of these processes is that the only we you can know the process has been successful is by destroying the part concerned.
• So a weld can only be fully tested by breaking it
• You can only demonstrate that heat treatment has given the correct hardness to the correct depth by taking a section and making hardness tests

And so on for all of these other processes.

Now we do not want to do this for each part that comes off the process (yield rate = 0), so hence the requirement to validate the process – if we have qualified welders and we destructively test the first few parts of a new design we provide confidence that the process will give us the standard of weld we require and we can ship good parts instead of broken pieces! But we have to control the process to ensure we keep our confidence in the process – hence the need to follow elements a to e

a) the only way to validate a process is by using known methods / knowledgeable people (e.g. development and sign off of a weld process with defined parameters following destructive test of components)
b) if we want to be confident the process is in control we may want to ensure that only known / calibrated equipment (e.g. weld sets) is used or that only qualified people (e.g. coded welders) operate the process
c) we may have work instructions for the task (e.g. weld procedures)
d) we may ask for people to make specific records to demonstrate the process has been operated in accordance with defined process methods (e.g. we ask the welder to record the weld wire feed rate, voltage, current and shield gas settings for each job)
e) periodically we go back and check that the process is still operating effectively (e.g. by destructively testing a sample part or series of parts).

But the important thing is - if you can go back and measure / inspect a part (or service) you are entitled to claim exemption from clause 7.5.2