Chris, you need to get:
- ANSI/AIIM MS44 " Recommended Practice for Quality Control of Image Scanners",
- AIIM X441 "AIIM Scanner Test Target - for use with AIIM MS44", and
- ANSI/AIIM TR34 "Sampling Procedures for Inspection by Attributes of Images in Electronic Image Management (EIM) & Micrographics Systems"

Unless the value of the records you are scanning is quite low, you want to be sure to use BOTH quality control methods described in MS44.

TR34 has a menu of sampling rates you can use for MS44, and you can use different rates for each of the QC methods. If you talk to experts in statistical quality control, you will find they tell you the error-detection rate is LOWER when you try to QC every image than if you use a correctly-designed statistical sampling method (or two, in this case). I'm NO expert in statistical quality control, so don't take my word for it:  find and ask your own.

ISO Technical Committee TC171 is working on various angles of the scanning QC process, but isn't ready to publish anything yet, so these are the currently-recommended tools.

I agree with Todd's comments.

Best of luck,

Frederic J. Grevin
Email: [log in to unmask]
Phone: 212-788-8615

-----Original Message-----
From: Records Management Program [mailto:[log in to unmask]] On Behalf Of Todd Johnson
Sent: Wednesday, January 12, 2011 14:40
To: [log in to unmask]
Subject: Re: [RM] Sampling

If the volume is large I would recommend sampling. Get all parties to agree on what percentage should be sampled and go from there. Verifying every image is very time consuming and might not be worth the effort depending on what's being imaged and what the cost is of having a "bad" image.

Todd Johnson, CRM, ermm
Analyst - Records & Information Management
Southern Star Central Gas Pipeline
4700 Hwy 56
Owensboro, KY 42301
Direct: (270) 852-4512

On Wed, Jan 12, 2011 at 1:24 PM, Chris Flynn <[log in to unmask]> wrote:

> Hello,
> I have some folks that are looking at an imaging project, using Image Now.
> Any experts out there? i would love to tap your brain. They are also looking
> at sampling rather than verifying every image (my recommendation is to
> verify every image). Any thoughts on sampling?
> Chris Flynn

List archives at
Contact [log in to unmask] for assistance
To unsubscribe from this list, click the below link. If not already present, place UNSUBSCRIBE RECMGMT-L or UNSUB RECMGMT-L in the body of the message.
mailto:[log in to unmask]