RECMGMT-L Archives

Records Management

RECMGMT-L@LISTSERV.IGGURU.US

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Frederic Grevin <[log in to unmask]>
Reply To:
Records Management Program <[log in to unmask]>
Date:
Wed, 14 Aug 2013 23:25:00 +0000
Content-Type:
text/plain
Parts/Attachments:
text/plain (33 lines)
Deborah Martin asked "What are other companies doing for quality control (QCing) after documents have been scanned to verify the pages are scanned correctly and legible. E.g., are companies QCing every page, 10 pages per box, etc. What processes are used for QCing images?"

Deborah, we require our conversion contractors to use BOTH QC methods (there are two) described in ANSI/AIIM MS441988 (R1993) "Recommended Practice for Quality Control of Image Scanners" (http://www.aiim.org/Research-and-Publications/Standards/Catalog/220). 

The contractor's sampling rate varies, but we generally recommend they choose from ANSI/AIIM TR34-1996 "Sampling Procedures for Inspection by Attributes of Images in Electronic Image Management (EIM) and Micrographics Systems" (http://webstore.ansi.org/RecordDetail.aspx?sku=ANSI%2FAIIM+TR34-1996). Screen inspection and match proof print would have different sampling rates, with match proof print at a lower rate than screen inspection.

Finally, we perform screen inspections of the contractor's work -- and, if necessary match proof print inspections -- usually at a sampling rate of 30:1 at the start of a project. We'll reduce the sampling rate if the quality is acceptable AND consistent. We increase the sampling rate if it appears the quality of the work deteriorates. 

In all cases (contractor QC and ours), we require the QC results to be logged (and the contractor's log must be available to us for review upon request).

I have been working with document conversion since 1973 (it was microfilm then), originally in the world of library preservation and inter-library loans. In those days, we visually inspected every image. Every statistician with whom I've spoken since then has vehemently insisted that inspection of each and every image is MUCH LESS EFFECTIVE than a well-designed (statistically-speaking) random sampling program. 

I am NO statistician, but I can tell you from years of personal experience that the inspector's attention span falls off rapidly after a fairly short time of 100% inspection. So I'm inclined to think those statisticians are correct. 

I suspect that at least some of these 100% inspection programs are for "due diligence" reasons, rather than for real value in a QC sense ("yes, judge, we DO check EVERY IMAGE!!!").

You didn't ask about the quality control of the image INDEX, but I'll hope you do have a system for that ...

Best regards,

Fred
---------------------------------
Frederic J. Grevin
[log in to unmask]
212-312-3903
Vice-President, Records Management
New York City Economic Development Corporation

List archives at http://lists.ufl.edu/archives/recmgmt-l.html
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]

ATOM RSS1 RSS2