Could someone help me better understand the differance between "Validation" and Verification"?

Thanks!

:bonk:

How about this,

System > Validate> Verify

We start some process, it could be a new project or a corrective action or whatever.

We validate through the tools at our disposal that it is feasable and works.

After we validate (short term) a process what to we do? Leave it alone or verify that our actions were indeed valid?

Part two next-------------------------------->

I assume you are looking at the 7.3 design section. Verification is basically makeing sure what you designed meets what you were trying to design. Validation means it actually works in the intended application.

For example, say you are designing a bumper for a car. Specifications say it must:

a. Mate with the rest of the car.
b. Resist corrosion per some specification XXX.
c. Withstand a 5 mph impact with no damage.

To verify your design you could:

a. Perform a dimensional analysis of your blueprint/cad data with the rest of the vehicle and make sure everything will fit.
b. Evaluate the expected coating/plating performance to be used against the requirements in specification XXX.
c. Evaluate the material and design to make sure it is strong enough to withstand.

To validate you could make several prototype bumpers and:

a. Mount it to a car.
b. Run a salt spray / environmental test
c. Crash a car at 5 mph and check for damage.

Hope this helps.

Tom

Tom,

Another view using the automotive example. A customer requests an automobile that can accelerate from 0 - 60 in 15 seconds. Design engineer produces a design, based on car weight, for a 200 hp engine. We build a prototype engine and test for 200 hp therefore verifying the design. We then put it in the car and validate the car goes from 1 - 60 in 15 seconds.

Validation assures we meet the customers requirements. Verification assures the design output meets the design input.

I have a rather simplistic view on the differences of verification and validation. If I get an e-mail stating that someone on the Cove made a statement and I ask, say Energy, and he says "yes, that is so" I could say that the statement is validated based on the fact that energy is a frequent visitor here, and he has never told me a falsehood (except maybe a fish story or two;) )

But, If I ask Mike S., he'll not only tell me it is so, he'll find the quote and send it to me. He just verified the statement by not only his experience and competency but by providing back up evidence.

There are some things made that you really can't verify, through hardening of gear blanks for one. But the process can be validated by regulating and monitoring the furnace, process controls of product going in, and any types of nondestructive testing and your past track record in producing these parts.

Just in case this distinction was missed:

As noted in the body of the posts above, keep in mind that there are two types of validation being discussed here: product validation (from a customer's perspective) and process validation.

Regards,

Kevin

look at it this way validation is the process of verification.
If a nightclub requires all patrons to be over 21 years of age, they must validate this is true. They do that by placing a person at the entrance to verify age by checking ID's.

In otherwords, the only way to validate something is to verify it. The act of checking is verification. The process of checking is validation.

In aviation maintenance verification is checking to see if systems and components work on the ground (MOC's) and validation is the test flight activity (MTF) itself.

Thanks for all of the input!

One more question:

we are a Heat Treating company. All we do is take the customer part, heat treat it to harden it, and then send it back to the customer. We test for hardness to make sure that it is correct.

How does verification and validation fit into this?

Thanks!:bonk:

Verification will be to do a hardness check to make sure you hardened it to the correct specification. Validation will most likely be up to your customer - unless you have the ability to run the part in a real world type test.

In your situation I normally state something to the effect that verification is the act of inspection. Validation is the responsibility of the customer. As long as the customer does not complain, and pays the invoice, the part is considered validated and the design process can be considered closed.

Tom

Dbzman,

Say that you can’t verify hardness of the part because the hardness test is a destructive test method (it will leave a dimple where you don’t want one). How will you ensure that the heat treat process produces the right results?

You must validate the process by knowing what the process is capable of producing through various validation and verification techniques. Perhaps the only way to ensure shipping the right hardness is by knowing the process intimately, through a prior validation process, where you can offer reasonable assurance that the outcome will meet specifications (process control monitoring, sample bar of material with similar composition, etc).

If I run a process in statistical control, my outcomes are predictable. By performing SPC, I monitor the process and know that the outcomes are what they should be, without having to perform any verification activities (beyond initial process validation, of course).

Regards,

Kevin

dbzman said:

Thanks for all of the input!

One more question:

we are a Heat Treating company. All we do is take the customer part, heat treat it to harden it, and then send it back to the customer. We test for hardness to make sure that it is correct. Verification

How does verification and validation fit into this?

Thanks!:bonk:

When you setup the line to process the next batch of parts for this customer and take a first piece sample to assure that hardness is within the requirements you are validating. (You do this by testing a part -verification)

Can anybody tell me now the diference about efficiency and effectiveness?:confused:

tattva said:

Can anybody tell me now the diference about efficiency and effectiveness?:confused: Why don't we start with the ISO 9000:2000 definitions:
effectiveness
extent to which planned activities are realized and planned results achieved

efficiency
relationship between the result achieved and the resources usedEffectiveness is about how well something works. Efficiency seems to be about how much it costs to do it (in terms of $$, people, or other resources).

Originally posted by dbzman
Could someone help me better understand the differance between "Validation" and Verification"?

Thanks!

:bonk:

I think in TS 7.5.2 Validation of processes for production and service provision has clear definition, it said:

The organization shall VALIDATE any processes for production and service provision where the resulting output cannot be VERIFIED by subsequent mornitoring or measurement..............

SO, VALIDATE IS A PROCESS
VERIFY IS AN ACTION

According to the ISO definitions, verification is confirmation that a requirement has been met. Validation is confirmation that a requirement has been met related to the intended use of the product (or process).

Originally posted by dbzman
Could someone help me better understand the differance between "Validation" and Verification"?

Thanks!

:bonk:

Verification: Paper
Validation: Product

:)

Originally posted by tomvehoski
Verification will be to do a hardness check to make sure you hardened it to the correct specification. Validation will most likely be up to your customer - unless you have the ability to run the part in a real world type test.

In your situation I normally state something to the effect that verification is the act of inspection. Validation is the responsibility of the customer. As long as the customer does not complain, and pays the invoice, the part is considered validated and the design process can be considered closed.

Tom

I´m in the foam seat industry and we do perform hardness test to seats, and recently one of our customers in Plymouth found out one headrests was not in SPC, they said it had 14.5 N for I.L.D., the SPC is 30.6 to 39.5, they sent it back to us, made us to present a contianment program, hand & visual verification, and many other things, president of this company even said: how in God´s Name we were QS certified and worse trying for TS.

The issue here is that the piece was returned to us and try another shot for the piece´s hardness " IT WAS 31.6 N". So is it a good thing to leave validation to the customer?

Any ideas about another way to perform validation? Thanks im advance :truce:

It sounds like you need to do a correlation of measurement methods with your customer, and educate them on the proper method. More than likely your method is right, since (I assume) you're the experts...

From TS2;
Verification - "...ensure that the design and development outputs have met the design and development input requirements".
Outputs are design records, BOM, test requirements, process studies,etc.,etc. that are required to meet the input requirements.

Validation - "... ensure that the resulting product is capable of meeting the requirements ...".
This is perfromance testing of the product.

tattva,

I would not consider your situation as verification or validation under the design requirements. That should have been long ago in development when they tested the parts you make to verify that the specification worked in safety, comfort, durability, etc. for the passenger.

I agree with howste - you need to get with your customer and figure out what went wrong in the test. I'm not familiar with your type of testing, but the measurement you noted being so far off makes immediatly suspicious of their measurement.

We used to have issues with customers on hardness of steel. We would be given a spec of 58-60 HRC. Customers would reject parts because they measured 57.9 on their digital tester. Hard to convince them that even a certified test block is +/- 0.5, so 50% of my tolerance is used up in my calibration standard.