This is my first post, wish somebody can give me some ideas on this:
we are establishing our 13485 QMS procedures, is it necessary to indicate the validation of equipments(such as injection machine) in the procedure which is suggested by our consultant, but actually we do not and don't know how to do the validation for the equipment.
many thanks
:)

This is my first post, wish somebody can give me some ideas on this:
we are establishing our 13485 QMS procedures, is it necessary to indicate the validation of equipments(such as injection machine) in the procedure which is suggested by our consultant, but actually we do not and don't know how to do the validation for the equipment.
many thanks
:)


Welcome to the Cove.

I suggest that you include a procedure on Equipment Qualification as Level 2 document and could have a template for doing equipment qualification as Level 3.

I recommend to do equipment qualification for all major equipment like injection molding machine, material dryer, robots, mold temp controller etc.

This is my first post, wish somebody can give me some ideas on this:
we are establishing our 13485 QMS procedures, is it necessary to indicate the validation of equipments(such as injection machine) in the procedure which is suggested by our consultant, but actually we do not and don't know how to do the validation for the equipment.many thanks
:)

A good starting point would be to read (and understand) the FDA guideline on process validation

http://www.fda.gov/Cder/guidance/pv.htm

and also get a good grip of GAMP IV.

Second, you need to qualify personell to perform process validation. This includes management (IMO), otherwise you will never perform a single useful validation.

Third, you need to generate a Master Validation Plan, to prioritize your validation effort. This validation plan should take the guideline on whether or not to validate into consideration. Basically the guideline states that if you can securely verify the result of your process then you do not need to validate, but if you cannot, or have other reasons, e.g. financial, then you should. Since the task of validation is a burdensome one, and the amount of validation that may need to be done if your company is new to the requirements in 13485 is staggering, then I strongly suggest that you consider which of the validations will provide the most value to your company, and which are the easiest ones to perform so you start with some successes.

... validation of equipments(such as injection machine) ... we do not and don't know how to do the validation for the equipment.

http://www.fda.gov/Cder/guidance/pv.htm
Process validation is establishing documented evidence which provides a high degree of assurance that a specific process will consistently produce a product meeting its pre-determined specifications and quality characteristics.

Validation is only required where "routine end-product testing" is not sufficient to assure "product quality".
http://www.fda.gov/Cder/guidance/pv.htm

I would strongly recommend to avoid "straight-forward" approach like "here is the mold machine producing MD's; let's go to validate it".
In my experience (with production of rubber catheters) a mold machine shall rarely be a subject of a validation because the features of a molded product may mostly be inspected with an acceptably-high degree of confidence.
Then, for example, if a Product FMEA says that all deviations from the requirements of the "pre-determined specifications and quality characteristics" are having low RPN's (usually due to, e.g. the features being 100% automatically tested; when such testing is likely to be established for critical/safety features of an MD) then the subject of a validation shall be the test equipment and probably not the mold machine.

Validation is only required where "routine end-product testing" is not sufficient to assure "product quality".


I would strongly recommend to avoid "straight-forward" approach like "here is the mold machine producing MD's; let's go to validate it".
In my experience (with production of rubber catheters) a mold machine shall rarely be a subject of a validation because the features of a molded product may mostly be inspected with an acceptably-high degree of confidence.
Then, for example, if a Product FMEA says that all deviations from the requirements of the "pre-determined specifications and quality characteristics" are having low RPN's (usually due to, e.g. the features being 100% automatically tested; when such testing is likely to be established for critical/safety features of an MD) then the subject of a validation shall be the test equipment and probably not the mold machine.

In my experience, in medical devices, the absolute sinner in a company is the plastics department. The components they produce are usually very cheap, and the attitude behind the production reflects the price. Unfortunately the quality that is the result of this attitude is a varying, to put it mildly, resulting in an output that creates trouble for the rest of the company, since these parts is critical for the products and the processes.

If I were to put a financial view on my validation efforts, then I would always prioritize validation of plastics before any thing else (excluding sterilization of course :) )

Check out this attachment
http://elsmar.com/Forums/attachment.php?attachmentid=6583&d=1174179173
posted in this thread by Qualityalways
http://elsmar.com/Forums/showthread.php?t=20841&highlight=ghtf
and other threads on the subject, for example the results of a search for GHTF (Global harmonization Task force):
http://www.ghtf.org/
Good luck
György

Aristo
Attached you will find a IQ/OQ (Installation Qualification/Operational Qualification) validation protocal. This plan is based on plastic injection molding machines, You will need to tweak for use on your machines. Use the operations manual received with your machine to tweak validation procedure. As previously stated as long as you can measure the results of the device manufactured you will not have to do process qualification.

it's very helpful!!
:thanx:

Aristo
Attached you will find a IQ/OQ (Installation Qualification/Operational Qualification) validation protocal. This plan is based on plastic injection molding machines, You will need to tweak for use on your machines. Use the operations manual received with your machine to tweak validation procedure. As previously stated as long as you can measure the results of the device manufactured you will not have to do process qualification.

Seems like a good starting point, but I fail to see things like qualification of materials, qualification of surroundings, air, humidity, training of personell, cooling water, worst case scenario and establishing of process control... :2cents:

I strongly oppose to not performing PQ on an injection moulding machine, just as I strongly oppose to not defining the worst case scenario or identifying the properties of the material needed for the process. It will greatly diminish the effect of the validation to the point of rendering it useless. :(

we are establishing our 13485 QMS procedures, is it necessary to indicate the validation of equipments(such as injection machine) in the procedure which is suggested by our consultant, but actually we do not and don't know how to do the validation for the equipment.
:)

My questions/confusions start off similar to Aristo's but veer off in another direction. I am updating from ISO9001:2000 to ISO 13485. As long as my company has held an ISO9001 cert, we have always claimed an exemption to 7.5.2 Validation. The reason behind the exemption is that we feel we can verify our product through inspection and test.

A few years ago, one of our customers was in here performing an audit for production of medical devices. They jumped all over me for not having installation qualifications in place for our surface mount line which includes a solder paste stencil machine, pick&place machine, and solder oven. I told them we verified product through inspection and test but they would have nothing of it. I asked several times to show me their requirement but they could produce nothing. Management wanted to pacify the customer so installation validations were done along with validation of software. (The software validated was written internally to control purchasing parts and inventory control. IMHO this didn't need to be done either.)

We recently bought replacement surface mount equipment and validations were performed again. I asked the operator why he was performing the validation. He said he didn't know why. Only that it was done before.

Continuing on...
ISO 13485 calls out for the validation of computer software used for the production/service process. We use software with a continuity test box (used to check wiring on medical devices) so I am assuming that requires validation. From my previous example, I am assuming an inventory control program does not require validation.

So where am I going with this?
1) If I can truly verify product through inspection and test, do I need to perform installation qualification / validation?
2) I can no longer claim exemption to 7.5.2 under ISO 13485 because I have to validate the test box software?
3) Were the requirements placed upon my company by the customer right or wrong?

As I am suppose to be the expert on all this quality stuff, talking to others here has led me to even greater confusion. (One person told me, you're the expert, go look in the mirror and ask the question.)

So where am I going with this?
1) If I can truly verify product through inspection and test, do I need to perform installation qualification / validation?
2) I can no longer claim exemption to 7.5.2 under ISO 13485 because I have to validate the test box software?
3) Were the requirements placed upon my company by the customer right or wrong?

As I am suppose to be the expert on all this quality stuff, talking to others here has led me to even greater confusion. (One person told me, you're the expert, go look in the mirror and ask the question.)

Do you perform 100% inspections of the process output? ISO/TR 14969 indicates that process validation is required unless the output is fully verifiable. Does your customer have issues with the quality of the product you are providing?

From the details you have provided, I don't believe your customer's requirements are unreasonable. I also believe you would have a hard time justifying exemption to 7.5.2. :2cents:

Do you perform 100% inspections of the process output?
Yes.

ISO/TR 14969 indicates that process validation is required unless the output is fully verifiable. Does your customer have issues with the quality of the product you are providing?
At the time of the audit, there were no issues regarding the quality of the product. The customer peforming the audit was sold to another company. The new company never has inquired about the validation situation or has made an issue of it.

From the details you have provided, I don't believe your customer's requirements are unreasonable. Unreasonable? Maybe not, but I was waiting for a "you must do this BECAUSE....". I gave them the reason why we weren't doing it. I expected a reason from them to do it.

also believe you would have a hard time justifying exemption to 7.5.2. :2cents:
I can see not being able to exempt the clause due to the software requirement of 7.5.2 in ISO 13485. As far as performing IQs, OQs, or PQs though, as of right now, I'm holding my ground and saying "not applicable" as we perform 100% inspection and test.

The reason for process validation is only present when you cannot verify the outcome. That is the reason why sterilization has to be validated. However, if you have economic reasons that justify validation that is a whole different subject.

Software validation is applicable only when the software controls a process, releases or approves products or when the software maintains the records required under the scope of the certificate etc.

:bigwave:

I agree with Gert's points. Given that you are able to fully verify the process output, validation is not required by 13485. Is your customers insistance on validation based on their interpretation of 13485 requirements?

You might seek some guidance from your registar.

So where am I going with this?
1) If I can truly verify product through inspection and test, do I need to perform installation qualification / validation?
2) I can no longer claim exemption to 7.5.2 under ISO 13485 because I have to validate the test box software?
3) Were the requirements placed upon my company by the customer right or wrong?

As I am suppose to be the expert on all this quality stuff, talking to others here has led me to even greater confusion. (One person told me, you're the expert, go look in the mirror and ask the question.)[/QUOTE]


My thoughts are the reason you need to validate is that in your processes you can not fully verify the outcome. You can do some functional testing, but you can not test, or most likely do not test, that all processes were completed as required to meet your, or your customer's needs

All the operations you listed in relation to surface mount, solder, bake, are all special processes. While you can verify they were done, you can not vouch for their integrity without destructive testing. You can show some solder got on a certain spot, and was cured in a bake, but you can't verify how long that solder would last, if it has adequate coverage to stay put in use conditions, etc. These processes are pretty commonly defined as special processes.

I have seen solder fail when implanted in the body. it worked functionally great at the supplier, but after implantation it became aparent that the solder was not adequate to perform its' purpose. I would look into validating all of the processes you just mentioned not to show if they are done, but how well they were done. In 13485, this is just the nature of the game.

The fact is true that you should do validations to prove the process, and the verification after the validations have been completed. My answer to your auditor (and has been ourselves) that at the time (pre-ISO 9000 and longer for 13485) your validations may not have been done or documented. If you are making a medical product and you are in the US, you are probably registered with the FDA, regardless of ISO, you still have the obligation to perform (or document why you didn't perform) validations.

Now, going forward; since you've obviously not had quality issues or made changes to the processes - I'd perform a retrospective validation on those same processes. That way when you are audited again, you can show a validation, which would also include a verificaiton of the documentation and the operators performing the job to the documentation. Some auditors just need to check off a box, they aren't always going to go in to depth. Keep in mind that IF you make changes to any of the processes that have been validated and are verified on an on going basis; you WILL need to perform and document a regular validation.

Sorry, I understand your pain. Tata :frust:

Just a note here, retrospective validation requires valid data. Most of us don't have valid data for retrospective validation purposes. So going down that road requires a wee bit of investigation into old records to make sure that it is actually possible. :bigwave:

The fact is true that you should do validations to prove the process, and the verification after the validations have been completed. My answer to your auditor (and has been ourselves) that at the time (pre-ISO 9000 and longer for 13485) your validations may not have been done or documented. If you are making a medical product and you are in the US, you are probably registered with the FDA, regardless of ISO, you still have the obligation to perform (or document why you didn't perform) validations.

Now, going forward; since you've obviously not had quality issues or made changes to the processes - I'd perform a retrospective validation on those same processes. That way when you are audited again, you can show a validation, which would also include a verificaiton of the documentation and the operators performing the job to the documentation. Some auditors just need to check off a box, they aren't always going to go in to depth. Keep in mind that IF you make changes to any of the processes that have been validated and are verified on an on going basis; you WILL need to perform and document a regular validation.

Sorry, I understand your pain. Tata :frust:
I have been following this very interesting trail. Some examples of the level of validation work required/carried out for surface mount line would be helpful(Is anyone willing to share their work).
I am not a medical device/automotive manufacturer but would think on following lines:

For paste printing following should suffice:
a. Solder paste selection/evaluation i.e. idle time response, first print, paste release consistency, print definition, SIR & tack time testing(if not happy with the supplier information), hot & cold slump testing, halide content, X-Ray & cross sectional analysis to check level of wetting and voiding, pre-placement and post reflow inspection results. Corrosive Gas Testing, Thermal Cycling and vibration testing may also be carried out.

b. Stencil design, footprint design, tooling support, process settings, etc.
c. Use of paste manufacturer recommended stencil/USC cleaning solvent, paper.
c. Operator training, SOPs
d. Maintenance schedules

If the results of the above under controlled conditions are shown to give positive results conforming to IPC A 610, then all that should be required as a control is 2D/AOI inspection before component placement.

Obviously the controls for PCB revision change etc. should be in place. A process checklist should suffice.

For Pick & Place:
1. Program Change control
2. SOPs for operation, maintenance, reel replenishments, component salvage, etc.
3. A separate evaluation showing equipment placement accuracy, and placement consistency.

If above has been carried out then an AOI/X-ray inspection as a control measure should suffice.

In my opinion Reflow soldering is a special process:

1. The reflow profile conforming to the paste manufacturer and the component manufacturer's recommendations should be achieved and then soldering results verified using methodologies like visual, X-Ray, Optical endoscopy, AOI, inspection as well as cross-sectional analysis(HPLM,SEM/EDX), vibration testing, etc. SMT engineer's responsibility should be limited to specifying what minimum profile specifications the components should meet. The design engineer should select components bases on those recommendations or vice versa. Component and PCB metallurgy should also be taken into account.

2. SOPs, maintenance procedures and operator training should be put in place.

IMHO, subsequent inspection/verification methodolgies like AOI, X-ray, endoscopy, etc. should be good enough control at the end of reflow.

Obviously the test conditions selected for some of the above testing will be dependent upon the worst conditions the end product/ process could be subjected to.

In my opinion all of the above pertains to process qualification. It will be interesting to see with real examples of what and how validation/qualification has been carried out by the medical companies for equipment and processes.

WOW Divensi! What an incredible first post! I'm glad you found us, and look forward to seeing more posts from you.

Welcome divensi :bigwave:

I echo what Brad has stated - expect to hear more from you :)

Yes, I agree with Gert as well; reading in between the lines it sounds like there may be sufficent records to evaluate to perform a retrospective Val on at least part of the processes. However, when the new machine was bought there's no wiggle room, Validations should be preformed the on all the equipment anyway.

Have you checked the FDA website for Validation guidance documents? (there's also one referencing off the shelf software for review) You didn't mentions what type of devices were being produced or if you were registered (the FDA will not be as kind in many areas)-[U]talk to your consultant, I assume he's got FDA experience, not just an ISO background. here's a generic guidance procedure from the FDA. ....fda.gov/cdrh/ode/425.pdf

:frust:

Tata

Just a little opinion on this...

Validation is one of those frustrating components with doing work with the FDA. However, a good validation program is quite valuable.

1. Establish a good, logical validation program. There is no reason to have an "everything including the kitchen sink" validation plan for every piece of equipment. For example, you might have abbreviated plans for identical equipment added that has already been validated. Maybe develop a process-based approach that allows different approaches depending on the nature of the equipment. Just be consistent.

2. If you have to do it, instill value in your validation. Saying, if you do it, make it worth something by identifying valuable information that may otherwise have not been identified (or more accurate, not documented). Too, initial validation gives you a good baseline for the operation of the equipment. If/when something changes in your process, you have information to inspect and find what changed. If you don't have that baseline, it becomes a little more challenging to determine if (and when) anything changed with the equipment.

Locating a competent, experienced consultant in this area may be well worth the money spent. They tend to see a lot of programs and are generally pretty knowledgable about current FDA trends and the like. Granted that is not absolute, but I do know of some really good consultants that fit this bill.

Good luck, and let us know how this progresses.

Nobody shared real examples to elaborate. Not surprised at all in this secretive world. :sarcasm:

Nobody shared real examples to elaborate. Not surprised at all in this secretive world. :sarcasm:

:modcop: No need for sarcasm here. :modcop:

Your post -and your request - was very specific, and it is far from certain that any of our participants have validated a similar setup. Even if they had, it is not certain that they were allowed to share the documentation.

No Worries! :D:frust:I was just curious. Perhaps curiousity kills the joy of life:nopity:

Nobody shared real examples to elaborate. Not surprised at all in this secretive world.

Well, to an extent I see your point. It is frustrating dealing with validation. Primarily, IMHO, it's one of those things the FDA has said "Do", and has not left better specifics of how to do it.

As Gert correctly suggested, it is difficult (and most times proprietary) to share validation plans. For example, does your equipment have electronic records? Well, how has your organization defined electronic records? Do you have a standard template to follow for validation plans? Are you a medical device? Are there other components working with the system? on and on and on.

Getting the specifics is not the important thing. Just like a QMS, you should get with your upper management and develop an approach which address the regulatory requirements, but more important, establish that the equipment meets your business need.