Validation of Foreign Body Detection Device
What procedure did you utilize to validate your Foreign Body Detection device?
Dear Robert Rogers,
What kind of device ?
Product / Process ?
Any particular standard ? It can matter.
Rgds / Charles.C
during the recent GFSI event spoke to several auditors. It was brought to my attention that a frequent non-conformance issued is regarding the lack of validation for their foreign body detection device.
Therefore, I wanted to see what others on the site have done to satisfy this requirement
Have they used any SOP such as:
Ensure equipment is installed properly
System capable of operating to maximum sensitivity with no affect from the surrounding environment or equipment
Establish production sensitivity so that there are no false rejects
Determine achievable detection capability by passing control samples through the weakest area for detection
You have mentioned 2 functions, sop, validation which tend to be (haccp) partitioned. In reality they can obviously overlap.
I assume the auditors/NCs you mention were referring to a haccp implementation.
Can personally only refer to metal detectors (MD). IMEX, haccp-wise, the validation aspect tends to be (narrowly) associated with validation of critical limits typically for a MD/CCP via, for example, use of (“wand”) standards. This function should clearly involve things like sensitivity maintenance aspects, but practically not always the case IMEX, ie (+) or (-) test(s). Or perhaps sensitivity etc checks are outsourced as a (annual?) certification “sop”.
There are also a range of viewpoints here regarding the exact status of the detector, eg as a CCP itself, a monitor, a verification tool . May depend on the process.
A popular operational reference for Metal detectors is the Codex validation procedure. Just search (or google) "codex validation". One recurring difficulty if you compare standards and forum threads is as to the variety of interpretations of validation.
Rgds / Charles.C
Codex mentions three areas Validation, Verification and Monitoring.
It seems
Validation would be the process to determine what level of control is possible. i.e. determining sensitivity capability or to what size contamination is able to be consistently and reliably detected and rejected
Verification would be ensuring this level of control is met. i.e. hourly checks with metal samples to confirm detection and rejection
Monitoring would be evaluating the information the program or process provides. i.e. evaluate customer complaints for metal, review verification logs, evaluate rejected products or failed hourly checks
Most organizations have procedures for verification and monitoring and lack information as to how and why they determined the particular size or type contamination that they are using during their verification checks.
Other processes are validated based on scientific evidence, mainly regarding the chemical and biological concerns.
With physical contamination, there are not any studies, that I am aware of, that covers the broad spectrum of products, system types and sizes, facility and other application variables.
If this is done on an individual basis I am curious what sort of steps the industry takes to satisfy validation requirements
I suggest you try this thread as a starter. Illustrates the background to most of the facets mentioned in yr post. As you will see, it is sort of agreed that there is no single (correct?) solution.
http://www.ifsqn.com...dpost__p__52215
There are some equally detailed items here specifically on validation per se, often for specific standards of interest. If you have one, I suggest a search here in that forum.
Rgds / Charles.C
What procedure did you utilize to validate your Foreign Body Detection device?
Unless you are building a new line, it's my opinion that you can't "validate" a foreign body detection device within the strict definition of Codex, as the validation has to be done before the equipment is purchased and installed.
What I have done is taken the FDA's "7mm guidance", equipment manufacturers specifications and common industry practice for size of test wands and mashed that into a document explaining why we have that certain metal detector. Is it proper "validation"? Probably not, but it's about the best that can be done until it's time to replace the equipment.
Marshall
mgourley
if a facility was not working toward any certifications in the past and had MD systems in place but now has contract with a customer whom requires certification or they want to comply with changes in the Food Safety Laws would they be obligated to purchase new equipment just to validate the process?
Since the environment and specific location could have an effect on detection capability, it would seem the validation should be performed after installation yet prior to production.
Thanks to all for their input.
mgourley
if a facility was not working toward any certifications in the past and had MD systems in place but now has contract with a customer whom requires certification or they want to comply with changes in the Food Safety Laws would they be obligated to purchase new equipment just to validate the process?
Since the environment and specific location could have an effect on detection capability, it would seem the validation should be performed after installation yet prior to production.
No, they would not be obligated to purchase new equipment. This is what I was referring to in my post. In my situation, the metal detectors were here long before I was and when they were installed, nobody was concerned with "validating" whether or not they were suitable for the purpose.
So now, I can't actually validate the selection of equipment, but I can show that what we have is industry standard and it operates correctly and to the industry standard for the type of metal detection equipment that it is.
Codex is pretty specific in the definition of "validation" and "verification". SQF, not so much. SQF has sort of melded the two definitions as far as I can tell.
My contention was that the only way to "validate" equipment, per Codex, is if it is new equipment, not currently in use.
Marshall
No, they would not be obligated to purchase new equipment. This is what I was referring to in my post. In my situation, the metal detectors were here long before I was and when they were installed, nobody was concerned with "validating" whether or not they were suitable for the purpose.
So now, I can't actually validate the selection of equipment, but I can show that what we have is industry standard and it operates correctly and to the industry standard for the type of metal detection equipment that it is.
Codex is pretty specific in the definition of "validation" and "verification". SQF, not so much. SQF has sort of melded the two definitions as far as I can tell.
My contention was that the only way to "validate" equipment, per Codex, is if it is new equipment, not currently in use.
Marshall
Understood
Here is a quote from the Codex validation document referred earlier –
(underline is mine)IV. CONCEPT AND NATURE OF VALIDATION
Validation focuses on the collection and evaluation of scientific, technical and observational information to determine whether control measures are capable of achieving their specified purpose in terms of hazard control. Validation involves measuring performance against a desired food safety outcome or target, in respect of a required level of hazard control.
Validation is performed at the time a control measure or a food safety control system is designed, or when changes indicate the need for re-validation (see section VII). Validation of control measures is, whenever possible, performed before their full implementation.
I guess it is your interpretation when the instrument is formally implemented as a control measure, although logically the validation should surely be done around the time of first usage from a FS point of view. This mini-zigzag is a common pragmatic alternative IMEX.
Here is a previous forum attachment of a “validation SOP” for a labelling control measure vs allergens which quite well illustrates the use of a Validation “trial period” (Canadian origin from memory).
validation study example.pdf 152.46KB 133 downloads
Slightly OT but if you are interested, this post and surrounding thread discusses some validation interpretative issues, particularly with respect to SQF standard –
http://www.ifsqn.com...dpost__p__43020
Rgds / Charles.C