Merlin Barcode Quality Industry Best Practices Work Group
Minutes of 11/5/2002 Meeting
The Merlin Barcode Quality Industry Best Practices Work Group held its final meeting
on November 5, 2002, from 10:00 a.m. to 12:00 p.m. at the USPS facility in Rosslyn,
VA. Attendees included:
Industry
Wanda Senne, World Marketing Inc., Industry Co-Chair
Mark Breunig, Domino Amjet
Hank Cleffi, MFSA
Stephen Colella, Calmark Inc.
Jay Gillotte, Presort Services
Dennis MacHarg, Advance Presort Service
Dale Miller, R. R. Donnelley
Bob O'Brien, Time Customer Service Inc.
Anita Pursley, Quebecor World
Bob Reeves, Perry Judds
Kathleen Siviter, PCSi (PostCom)
Chuck Somerville, Scitex
Joel Thomas, NAPM
Diane Ward, MSP
Terry Wehr, MSP
M. Winn, R. R. Donnelley
USPS
Scott Hamel, USPS Co-Chair
Hernan Borja, USPS Engineering
Paulette Kelly, USPS Business Mail Acceptance
Karen Magazino, USPS Mail Preparation & Standards
Bell and Howell
Dan Parenti, Bell and Howell
Introduction
The purpose of the final workgroup meeting was to review the Postal Service's
responses (copy attached) to a list of remaining MERLIN concerns identified by
industry. The workgroup prioritized the list and reviewed the responses where it
was felt that clarification was needed or further discussion necessary. A recap of
these issues appears below.
USPS' Responses on MERLIN concerns
The following questions from the "MTAC Mailer Barcode Quality Workgroup Q & A's"
list were discussed with disposition as noted.
5. Q: What is the status of the checklist for requesting MERLIN reports?
A: Will be completed 11/1/02.
Q: Which reports are optional, which are automatic?
A: If a mailing passes, we will furnish the Summary Report. If a mailing
fails, reports associated with the failure will be furnished.
The workgroup asked for further information on this item. The USPS said that
the checklist has been drafted and currently is being circulated internally
for final approval. At that time, a copy will be distributed to the USPS
field offices as well as to workgroup members. The USPS also said that
mailers can request additional reports on mail which passes MERLIN testing,
although fulfillment of that request may not occur until the next day as the
reports will be run during the equipment down time.
Industry asked if a "standing order" for a list of reports could be made. The
USPS responded that this concept produces unnecessary paperwork and prefers
that mailers submit a reports request form for each mailing of concern.
Mailers asked if a summary from multiple mailings would be available. The
USPS responded that it is working on this feature within PostalOne, where the
USPS could pull up reports on multiple mailings by pulling up one company
name. MERLIN retains mailing data for only 30 days, but PostalOne has
indefinite data retention. The reports from PostalOne are not the same format
as the reports from MERLIN, but could contain similar information. The USPS
has not established a definite time frame for accomplishing this feature.
6. Q: A statement as to what a mailer should do if local USPS personnel are not
following correct procedures is needed.
A: See the September issue of Mailers Companion.
The workgroup asked for further clarification on this item. The USPS
responded that if local USPS personnel are not following correct procedures,
the USPS Area MERLIN coordinator should be contacted. Mr. Hamel noted that if
problems persist locally and the Area coordinators are not helping, the mailer
should contact postal headquarters.
Industry noted that the list of USPS MERLIN coordinators still has not been
added to the USPS' web site (which was verified at the time of the writing of
this report). The USPS will check on the status and add the information to
its web site.
The USPS also clarified that if the cause of a barcode failure cannnot be
determined by a Mailpiece Design Analyst (MDA), the test results are
overturned by the supervisor.
7. Q: Where do things stand regarding the promise to have appeal mail that is
sent to Chicago entered into the mail stream? Industry does not want the mail
delayed by being returned to the post office where it was initially tested.
Q: Industry also wants the mail in the sample tested to be deemed mailed the
day the mail was submitted to the USPS.
A: The USPS agrees.
Industry asked for further response to this item. The USPS clarified that it
was waiting for a final answer re logistics from Great Lakes Area on this issue.
The USPS agrees that the mail will be deemed mail the day it is submitted. Industry
would like an acknowledgment that the mail has been entered in Chicago, which the
USPS said it is incorporating into its draft procedures.
The USPS has to ensure that the postage has been paid before the mail can be
entered in Chicago. Currently, the USPS deducts the postage for the appeal
pieces. The USPS will look at a way to change the procedures so that the
postage payment documentation is attached. The USPS plans to have this matter
resolved by early December.
When mailers are submitting pieces for pre-testing, they must provide
the USPS with the return postage for return of the sample pieces, the USPS
clarified. Appeal pieces are sent by the USPS via Express Mail to the appeals
center in Chicago. The pieces are then run, usually immediately, and the
results are faxed to the Chicago RCSC, who writes a letter and faxes it back
to the original mailing site.
8. Q: We need a report on how erroneous MERLIN results that are overridden, as a
result of a second level review, or on appeal, are corrected in the record so
that mailer quality history that reside in PostalOne! are accurate.
A: USPS does not plan to produce such a report.
Industry asked for further discussion on this item. Mailers are concerned
about the accuracy of the mailer history information that resides in
PostalOne! and questioned the USPS as to how the mailer can verify that the
information has been corrected in the mailer history when a MERLIN test result
has been overridden by the supervisor. The USPS responded that the override
is reflected in the report summary. The USPS said it will manage this process
internally.
Ms. Kelly said that all documentation on second level overrides comes to her
for review and to date there are not many. Industry said that some USPS
locations do overrides regularly, but some never do them. The USPS said that
the programming for PostalOne! will be done to make sure the data gets
uploaded so the history is accurate. If a mailer suspects data problems, they
should contact the USPS. The USPS noted that this data now resides in the
PERMIT system and if a barcode readability report comes off MERLIN which a
second level review then overrides, the mailing goes into the PERMIT system as
accepted 100%.
9. Q: What is the status of the barcode template for flats?
A: The design is in process. Both Brown and Quebecor are assisting.
Industry asked for further information on this item. Karen Magazino gave an
update on the status of the barcode template for flats, which currently is
being developed by the USPS and mailers. A draft of the template was
distributed to the workgroup members for feedback by 11/15. The template is
being developed with input from mailers and USPS MDAs. Ms. Magazino said that
the USPS wants to keep it as simple and clean as possible. Industry agreed
that the template should be made of a size to be able to measure maximum flats
dimensions and Ms. Magazino said that the USPS is exploring a template that
could be folded down, so that it will be more portable and still be able to
measure maximum flats dimensions.
The template will include specifications for both AFSM 100-compatible flats
and FSM 1000-compatible flats. The USPS also plans to resurrect the small
barcode template, and noted that the old DPBC gauge is obsolete.
The USPS plans to circulate a plastic draft version of the barcode template
for flats in early 2003 and finalize it for distribution as soon as approved.
12. Q: MERLIN can only read 11 or 12 degrees from the baseline of a barcode and
yet there is no positional skew requirement for a flat. Is this causing error
reports where there should be none?
A: MERLIN has a system limitation when calculating the positional skew for
flats. MERLIN reports the findings with the "unrecognized barcode" and
"Recognized, not decoded" barcode error messages. The operator is to review
the reported errors to ensure that the errors are not a product of the
positional skew system limitation. A template is developed to assist the
operator with the task.
Industry asked for further clarification of this item. USPS Engineering said
that pattern skew problems result in barcode reading equipment not being able
to recognize the barcode, but that the MERLIN system doesn't clearly identify
pattern skew as the problem so it returns error codes such as "unrecognized
barcode." The USPS relies on operator intervention to catch the pattern skew
problem, and has developed a template for operators to identify pattern skew
and override.
The USPS also is looking at a fix for pieces that "pop up" during feeding,
inhibiting the machine's ability to read the barcode. USPS Engineering noted
that this occurs most commonly with letters and cards and has to do more with
the speed at which the MERLIN machine is run (e.g., at 4000/hr these run
smoother). The USPS said that its policy is that if the bottom edge of the
piece is not down when it is fed, these pieces are ignored in the sample.
This "fix" will be part of a MERLIN software upgrade in January 2003.
15. Q: We need to address the problem of barcode failure due to the camera
reading through the label on page one of piece, to copy on page 2 or 3, or
through ink jet address on the cover of a piece, to the copy on the next
page(s).
A: This is an industry issue to resolve.
The workgroup asked for further discussion on this item, which also ties in
with Question # 7 on the "Policy Issues" list. Ms. Pursley further explained
the issue, which is that pieces pass the USPS reflectance meter test even
though they failed MERLIN. In some cases, pieces approved by MDAs are then
failing MERLIN.
The USPS reflectance meters are not available for mailers to purchase and are
not being properly maintained. Mailers want to know how to deal with
reflectance issues. Attempts are being made to have the densitometers
manufacturer talk with the Postal Service to see if these instruments could be
designed to gauge reflectance so that mailers have a tool to use in mailpiece
design.
The USPS asked that mailers submit samples where these problems are occurring.
Samples of 25-pieces should be submitted to Paulette Kelly at USPS so that
further analysis can be done.
16. Q: Industry would like an explanation of how USPS can use a 200 by 200 pixel
image in which each pixel is 0.005" x 0.005" to determine whether a standard
stated in 10/000th of an inch is or is not being met. (See list for further
explanation of this issue.)
A: USPS Engineering provided a detailed response to this inquiry in the Q&A
handout.
Industry wants further clarification, and feels that the USPS'
response does not adequately address the question. After some discussion, it
was agreed that the USPS and Bell and Howell will further discuss this issue,
with more input from industry as deemed necessary, and forward a response to
those concerned.
18. Q: We need to resolve the issue of whether the lamp used by MERLIN can cast a
shadow on the POSTNET barcode causing MERLIN to falsely report non-existing
errors.
A: MERLIN does not use a three dimensional method for image processing.
Industry asked for further explanation of the USPS' response on this item.
USPS Engineering explained that MERLIN shines a light directly on the target
and reflectance is analyzed. Since there is no object between the light and
the target, no shadow can be produced. After some discussion it was
determined that if industry wants to further pursue this issue, it should
provide additional information or clarification of the issue to USPS
Engineering.
19. Q: Industry would like to know what mail is being rejected off production
barcode sorters, and why, and what errors MERLIN is finding.
A: The Postal Service does not track mail in this fashion.
Industry wants further discussion on this issue. Industry wants the USPS to
perform a MERLIN test of mail rejected off production barcode sorters to
identify the top five problems so that industry can work to make quality
improvements in those areas. After some discussion, the USPS said it will
take this matter under consideration.
The following questions from the "Policy Issues from Mailer Barcode Quality
Industry Best Practices Meetings" list were discussed with disposition as noted.
1. Q: Industry would like to see a policy requiring MERLIN operators to complete
a certain percentage of MERLIN tests in the first part of the 60-day grace
period, to ensure that the majority of the 30 tests required to be completed
over the 60 days are not completed during the latter part of the grace period,
which would give mailers little time to identify and resolve barcode quality
problems.
A: 30 tests will be run at all DMUs during the grace period. The pace will
be set locally. Mailers may approach the district and area on a case by case
basis if an extension is warranted.
Industry asked for further discussion on this issue, but the USPS said that
it's position remains firm in that it will not set a daily pace for conducting the
30 tests within the 60-day grace period. Mailers can request an extension
of the grace period from the Area MERLIN coordinator if they believe there is a
problem with the number of tests conducted over the range of the period.
2. Q: Industry is requesting a clarification and a policy statement to