SIST-TP CEN/TR 17386:2019
(Main)Postal services - Transit time measurement for cross border postal items using real mail feasibility study
Postal services - Transit time measurement for cross border postal items using real mail feasibility study
This document is a feasibility study that was carried out to explore the use of real mail data in measurement of the transit time of end-to-end services for single piece cross-border priority mail. In this document a description is given of the context, the way this study was carried out, the results of the study and the advice given to CEN/TC 331 Postal services and, finally, CEN and the European Commission.
Postalische Dienstleistungen - Messung der Durchlaufzeit von grenzüberschreitenden Postsendungen unter Nutzung von echten Sendungsdaten
Services postaux - Mesure du délai d’acheminement du courrier transfrontière à partir des flux réels - Rapport de faisabilité
Poštne storitve - Merjenje prehodnih časov za čezmejne poštne pošiljke z uporabo študije realnih poštnih zmogljivosti
Ta dokument je študija o zmogljivosti, pri kateri se je raziskovala uporaba realnih poštnih podatkov pri merjenju prehodnih časov od sprejema do vročitve za posamezne čezmejne pošiljke prednostne pošte. V tem dokumentu so podani opis konteksta, način izvedbe te študije, rezultati študije in nasveti za CEN/TC 331 »Poštne storitve« ter CEN in Evropsko komisijo.
General Information
Standards Content (Sample)
SLOVENSKI STANDARD
SIST-TP CEN/TR 17386:2019
01-november-2019
Poštne storitve - Merjenje prehodnih časov za čezmejne poštne pošiljke z uporabo
študije realnih poštnih zmogljivosti
Postal services - Transit time measurement for cross border postal items using real mail
feasibility study
Postalische Dienstleistungen - Messung der Durchlaufzeit von grenzüberschreitenden
Postsendungen unter Nutzung von echten Sendungsdaten
Services postaux - Mesure du délai d’acheminement du courrier transfrontière à partir
des flux réels - Rapport de faisabilité
Ta slovenski standard je istoveten z: CEN/TR 17386:2019
ICS:
03.240 Poštne storitve Postal services
SIST-TP CEN/TR 17386:2019 en,fr,de
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.
---------------------- Page: 1 ----------------------
SIST-TP CEN/TR 17386:2019
---------------------- Page: 2 ----------------------
SIST-TP CEN/TR 17386:2019
CEN/TR 17386
TECHNICAL REPORT
RAPPORT TECHNIQUE
September 2019
TECHNISCHER BERICHT
ICS 03.240
English Version
Postal services - Transit time measurement for cross
border postal items using real mail feasibility study
Services postaux - Mesure du délai d'acheminement du Postalische Dienstleistungen - Messung der
courrier transfrontière à partir des flux réels - Rapport Durchlaufzeit von grenzüberschreitenden
de faisabilité Postsendungen unter Nutzung von echten
Sendungsdaten
This Technical Report was approved by CEN on 21 July 2019. It has been drawn up by the Technical Committee CEN/TC 331.
CEN members are the national standards bodies of Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia,
Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway,
Poland, Portugal, Republic of North Macedonia, Romania, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey and
United Kingdom.
EUROPEAN COMMITTEE FOR STANDARDIZATION
COMITÉ EUROPÉEN DE NORMALISATION
EUROPÄISCHES KOMITEE FÜR NORMUNG
CEN-CENELEC Management Centre: Rue de la Science 23, B-1040 Brussels
© 2019 CEN All rights of exploitation in any form and by any means reserved Ref. No. CEN/TR 17386:2019 E
worldwide for CEN national Members.
---------------------- Page: 3 ----------------------
SIST-TP CEN/TR 17386:2019
CEN/TR 17386:2019 (E)
Contents Page
European foreword . 3
Introduction . 4
1 Scope . 5
2 Normative references . 5
3 Terms and definitions . 5
4 Symbols and abbreviations . 6
5 Existing QoS measurement system . 6
6 Operational measurement . 7
7 Objective of the study . 8
7.1 Objective . 8
7.2 Research questions . 8
7.3 Summary of the purpose of the feasibility study . 9
8 Existing and proven technology . 9
8.1 Image quality . 9
8.2 Image analysis . 10
8.3 Image matching . 10
8.4 Volume of data associated to image analysis and matching function . 11
8.5 Proven technology . 11
9 Key points in the results of the survey . 12
9.1 Overall remarks. 12
9.2 Key points in question 1 . 12
9.3 Key points in question 5.1 . 12
9.4 Key points in question 12 . 12
9.5 Key points in question 14 . 13
9.6 Key points in question 30 . 13
10 Answers to the research questions 1 to 4 . 13
10.1 Answers to research question 1 . 13
10.2 Answers to research question 2 . 14
10.3 Answers to research questions 3 and 4 . 14
11 Decision of CEN/TC 331 Postal services regarding the feasibility study . 15
Annex A (informative) Questionnaire sent to 32 EU postal organizations . 16
Annex B (informative) Results of the survey filled out by 18 Eu postal organizations . 21
2
---------------------- Page: 4 ----------------------
SIST-TP CEN/TR 17386:2019
CEN/TR 17386:2019 (E)
European foreword
This document (CEN/TR 17386:2019) has been prepared by Technical Committee CEN/TC 331 “Postal
services”, the secretariat of which is held by NEN.
Attention is drawn to the possibility that some of the elements of this document may be the subject of
patent rights. CEN shall not be held responsible for identifying any or all such patent rights.
This document has been prepared under a mandate given to CEN by the European Commission and the
European Free Trade Association.
3
---------------------- Page: 5 ----------------------
SIST-TP CEN/TR 17386:2019
CEN/TR 17386:2019 (E)
Introduction
Under the current Standardization Request M/548, CEN and CEN/TC 331 Postal services identified a
need for a feasibility study to use real mail processing in the measurement of the quality performance.
This feasibility study has been carried out, researching the costs versus the benefits of a method for
transit time measurement for cross border postal items using real mail.
Based on this study, is became clear that the time and resources needed to develop such a method are
too extensive to justify it. Therefore, the members of CEN/TC 331 Postal services have decided that this
method is not feasible and that the method will not be developed within the current Standardization
Request M/548.
4
---------------------- Page: 6 ----------------------
SIST-TP CEN/TR 17386:2019
CEN/TR 17386:2019 (E)
1 Scope
The scope of this document is a feasibility study that was carried out to explore the use of real mail data
in measurement of the transit time of end-to-end services for single piece cross-border priority mail. In
this document a description is given of the context, the way this study was carried out, the results of the
study and the advice given to CEN/TC 331 Postal services and, finally, CEN and the European
Commission.
2 Normative references
The following documents are referred to in the text in such a way that some or all of their content
constitutes requirements of this document. For dated references, only the edition cited applies. For
undated references, the latest edition of the referenced document (including any amendments) applies.
EN 13850:2012, Postal Services - Quality of Services - Measurement of the transit time of end-to-end
services for single piece priority mail and first class mail
3 Terms and definitions
For the purposes of this document, the following terms and definitions apply.
ISO and IEC maintain terminological databases for use in standardization at the following addresses:
• IEC Electropedia: available at http://www.electropedia.org/
• ISO Online browsing platform: available at https://www.iso.org/obp
3.1
International Postal Corporation
IPC
cooperative association of postal operators in Asia Pacific, Europe and North America
3.2
Office of Exchange
OE
postal facility which handles the foreign mail departing to and/or arriving from another country
3.3
PostEurop
trade association that represents European public postal operators
3.4
Research question
RQ
investigations carried out as part of the feasibility study
3.5
Sequencing sort
sorting of the mail pieces in the order of the postman delivery
3.6
Test Letters
TL
tracked and traced letter with which the cross-border time is measured between OE’s
5
---------------------- Page: 7 ----------------------
SIST-TP CEN/TR 17386:2019
CEN/TR 17386:2019 (E)
3.7
UNEX™
postal QoS system of IPC which contains several specific modules and is used to measure international
letter service performance between 37 postal operators using test mail methodology, one of which
modules measures the end-to-end postal quality of service for single pieces priority in Europe
4 Symbols and abbreviations
CFC Culler Facer Canceller
DPI dots per inch
FC Facer Cenceller
KPI Key Performance Indicator
OCR Optical Character Recognition
QoS Quality of Service
RFID Radio Frequency Identification Technology
RMM Real Mail Measurement
RoI Return of Investment
SPPM Single Piece Priority Mail
USP Universal Service Provider
5 Existing QoS measurement system
The IPC UNEX™ CEN module measures the transit time of end-to-end services for single piece priority
and first class mail in Europe, according to the standard EN 13850:2012 Postal Services – Quality of
Services – Measurement of the transit time of end-to-end services for single piece priority mail and first
class mail. The measurement is end-to-end from posting in the original country (C) to delivery to the
final addressee in the destination country (I). This includes the time of collection, sorting and
transportation as indicated in Figure 1.
Figure 1 — Postal process from origin country to destination country
This performance measurement monitored since 1994, indicated the percentage of mail flows reaching
the final addressee within J+3 and J+5 after posting. The percentage objectives have been set
respectively to 85 % (J+3) and 97 % (J+5) (see the 1997 European Union on Postal Services Directive
(97/67/EC Directive)).
In order to accurately evaluate these two performances, CEN Test Letters representative of the real mail
streams are produced and panellists located in urban and in rural areas are in charge to record the
6
---------------------- Page: 8 ----------------------
SIST-TP CEN/TR 17386:2019
CEN/TR 17386:2019 (E)
posting and receiving dates of these test letters. In addition to this manual operation, there are around
40 % of these test letters with a specific RFID device which can be read automatically when the letter is
under a RFID antenna located at the entrance and at the exit of each sorting centre. That means for
these 40 % of test letters it is possible to record automatically the date when the letter is detected at the
various locations indicated in Figure 1: D E F G H.
In 2016 and 2017, 797 mail flows have been measured between 32 EU countries. There were around
1,200 panellists handling around 66 000 test letters per year.
The following Figure 2 provides the J+3 performance (also identified as speed indicator) across years
since 1997.
Figure 2 — J+3 performance across years
6 Operational measurement
In order to process the cross-border mail flows between EU countries, bilateral agreements between
USP in each country are in place in order to perform the transport from sender to receiver.
The transport between the sender (C) and the receiver (I) requires services to be delivered by each USP
in country A and in country B in addition to the international transport - see Figure 3.
Three different legs have been identified:
— Leg 1 corresponds to the transport from the sender in country A up to the hand-over to the first
carrier for the international transport – ref to segment C to E in Figure 1;
— Leg 2 corresponds to the international transport up from end of Leg 1 up to the hand-over to the
USP of the receiving country B. Multiple carriers may be involved – ref to segment E to F.in Figure 1;
— Leg 3 corresponds to the transport from end of Leg 2 up to the delivery to the addressee in country
B – ref to segment F to I in Figure 1.
7
---------------------- Page: 9 ----------------------
SIST-TP CEN/TR 17386:2019
CEN/TR 17386:2019 (E)
Figure 3 — Operational measurement
For the traffic from country A (sender) to country B (receiver), the Operational Responsibility is defined
as the following:
— The Inbound Post B is only responsible for Leg 3;
— Leg 2 which may include multiple carriers is the sole responsibility of the Outbound Post A in
addition to the Leg 1.
This operational responsibility is not related as such to the UNEX™ measurement but is an agreed split
of responsibilities between all the shareholders for cross-border mail traffic.
As there are sufficient information with the 40 % of test letters automatically detected due to presence
of the RFID component transported in these test letters and other messages transmitted between the
postal organisations, carriers, etc., it is possible for IPC to precisely determine the operational
measurement Leg 1, Leg 2 and Leg 3 for the test items equipped with RFID.
7 Objective of the study
7.1 Objective
The feasibility study shall answer the question of whether new techniques also using methods of
tracking and tracing of cross border letter mail within the EU could lead to a reliable and diagnostically
useful end-to-end survey to measure the transit time of end-to-end services for single piece cross-
border priority mail.
The method to perform this task is listed in the following four Research Questions:
7.2 Research questions
7.2.1 Question 1
Can images captured by sorting machines in sorting centres and offices of exchange per item
throughout Europe (EC member states) be stored and used in a survey in such a way that
— Individual mail items can be identified and;
— The date of posting (cancelling information with dates on stamps or the date in the franking image)
and tracking data per item throughout a big part of Europe is registered?
8
---------------------- Page: 10 ----------------------
SIST-TP CEN/TR 17386:2019
CEN/TR 17386:2019 (E)
7.2.2 Question 2
Can postal operators make these data available, also from a legal perspective?
7.2.3 Question 3
Can real mail data such as described in the questions 1 and 2 be completed with data in order to create
an end-to-end result?
7.2.4 Question 4
Can data from domestic surveys be used in combination with real mail track and trace data, as meant in
question 3, to produce an end-to-end survey to assess the norms for cross border mail within the
European Union in the Postal Directive 2008/6/EC?
7.3 Summary of the purpose of the feasibility study
— to analyse the possibility of acquiring automatically images of letters processed in sorting machines
installed in Office of Exchange (or in a sorting centre close to the OE) in both countries (sender and
receiver – from country A to country B as indicated in Figure 1).
— to be able to recognize individual mail items from image identification.
— to be able to track and trace individual items between OEs and to measure the time spent during
the international transport.
— to analyse the possibility to complement this international transport measurement with the
domestic QoS measurement in country A and country B in order to create an end-to-end result as
required by the CEN Standard EN 13850:2012 Postal Services - Quality of Services - Measurement of
the transit time of end-to-end services for single piece priority mail and first class mail.
8 Existing and proven technology
8.1 Image quality
This feasibility study has been launched based on the regular improvements made either in image
quality of the cameras and in image matching technology during the last several decades:
— The development of automatic mail sorting equipment started in the 70s,
— A decade later in the 80s, automatic OCR (optical character recognition) was gradually introduced
dealing with typewritten and then handwritten address recognition.
— During the next 10 years, several recognition engines were combined in the same logic to carry out
the full interpretation of addresses written on the letters enabling the domestic sorting up to
sequencing which is the last sortation phase before the postman delivery.
In order to perform the full address identification only once (at the start of the domestic processing
chain), a unique barcode (also called Tag Id code) is printed on each mail piece in order to easily
identify this letter all along the journey from the sender to the final addressee. All information required
for processing are associated to this unique Tag Id code. The sorting information is retrieved when the
Tag Id of the letter is detected and read in a sorting equipment.
9
---------------------- Page: 11 ----------------------
SIST-TP CEN/TR 17386:2019
CEN/TR 17386:2019 (E)
This barcode printing solution has a number of advantages to facilitate the tracing and tracking of each
mail piece, but has also some constraints which have to be bared by postal operators, mainly due to the
costs associated to
— The barcode printer,
— The ink,
— The maintenance and spare parts,
— The need to apply an additional label when the mail piece is plastic wrapped (i.e. magazine) as
printing on the plastic is not possible at the speed of the transport.
— The cost of the label applicator and the cost of labels applied on each item.
In 2000s, due to all these operational constraints, new technological solution of mail piece identification
based only on image analysis and image matching has been introduced removing the need of applying a
barcode on mail pieces and therefore all the associated above-mentioned costs.
8.2 Image analysis
The objective of the image analysis is to extract a Set of Parameters (SoP) which characterizes the image
of each letter. This SoP, also called the mail piece signature, remains nearly invariant when another
image of the same letter is acquired by another machine in another sorting centre.
In order to extract an accurate signature, the image quality should be at least a multi-grey level image
with a minimum definition of 8 pixels per mm (or 200 dpi – dot per inch).
During the export sorting:
— The destination country is identified by reading the address,
— The mail pieces are sorted to the machine outputs according to the OE destinations,
— Sample of letters (also named Test Letters - TLs) are selected in the different mail flows and the
signatures of these TLs are recorded in files (a file per OE destination),
— Associated to each Test Letter, the sequence of mail pieces processed before and after each TL for
the same OE destination are analysed and their signatures are recorded in the same data file for
that OE destination.
— Then, during the transport of the physical mail pieces to the inbound OEs, the data files containing
all these signatures of Test Letters and other associated mail pieces are electronically sent their
relevant OEs destinations.
8.3 Image matching
For cross border mail flows, the Image matching is the process performed in each inbound OE where
the image signatures of imported mail flows are compared with the signatures of TL (Test Letters)
acquired in all outbound OEs (see processes a to c as described above).
For each image processed in the inbound OE, the matching may provide several TL candidates of images
acquired in the outbound OEs, each candidate having its own confidence level value. The final decision
is made after analysing the letters sorted before and after each TL candidate – see “step d” as described
in Image Analysis paragraph above.
10
---------------------- Page: 12 ----------------------
SIST-TP CEN/TR 17386:2019
CEN/TR 17386:2019 (E)
As a result, the image matching function is not only a comparison between two images acquired in 2
sorting centres, but a more complex comparison between of a sequence of letters processed in Offices of
Exchange, firstly in the outbound OE and then the inbound OE.
8.4 Volume of data associated to image analysis and matching function
The most important factor to be taken into consideration for integrating these two functions - Image
Analysis and Image Matching - is the volume of data to be stored and managed.
The size of a multi-grey level mage with a definition of 8 pixels per mm (in a compressed format) varies
between 350 Kbytes (for small letters) and 700 Kbytes (for large letters). An average value can be
estimated to 450 Kbytes considering the ration of flats (20 %) and letters (80 %) in the live mail.
Therefore, the data storage for 100 000 images is estimated to 45 GB
The data generated for an image signature depends on the quantity of information contained in the
image, but an average value is estimated to 30 KB. For the same quantity of images (100 000 images),
the storage of all associated signatures is estimated to 3 GB
For an OE where the volume of mail to be sent abroad to EU countries is around 500 000 letters per
day; the required the total storage capacity required will be around 240 GB per day.
This evaluation can be updated to consider the low volume processed in some of the EU countries. For
the other countries, the integration of the real mail measurement will require a storage capacity close to
Terabytes as there is a need to consider several days for the tracking of the cross-border mail flows.
As a consequence, “Big Data” technology is clearly needed to be able to efficiently manage such data
volumes. The other optimization is the quantity and type of data to be stored at the different data
centres: the outbound and inbound OEs and at the central data processing and management. The trade-
off is between real time processing (i.e. matching images when import mail arrives in inbound OE) and
non-real time processing off-line (i.e. data transmission between OEs when the physical mail is
transported between the same OEs).
8.5 Proven technology
This technology of tracing and tracking mail pieces using only images with no barcode printed on mail
items has been integrated by a few postal solution suppliers. As some of them have communicated on
this subject via their website, it is a public information which can be listed hereafter:
Siemens Postal and Parcel Automation indicates that they integrated the Mail item identification with
fingerprint technology in the flat sorting systems delivered to Deutsche Post in Germany.
Solystic indicates that they integrated the tracing of letters using the image with no additional barcode
printed for the postal operator in the following countries: France, Finland and UK. This technology
based on image tracking is called “V-Id” which means “Virtual Identification”.
In these EU countries; postal operators integrated such solution mainly for the domestic sorting but
also for foreign mail in order to be able to video-code these items in an off-line mode (the mail pieces
have to be identified when reprocessed the second time and printing barcode on foreign mail is not
allowed).
All these methods of tracking and tracing mail pieces for all flows (small letters and flats) are now in full
operation for more than10 years. This technology has been really proven.
11
---------------------- Page: 13 ----------------------
SIST-TP CEN/TR 17386:2019
CEN/TR 17386:2019 (E)
9 Key points in the results of the survey
9.1 Overall remarks
The questionnaire includes 30 questions. An overview of the answers of the posts which provided their
feedback is attached in Annex B.
The following questions brought forward key points that may have a direct impact on this Real Mail
Measurement feasibility study: Question 1, 5.1, 12, 14 and 30. What follows is a summary of those key
points.
9.2 Key points in question 1
Automatic sorting equipment are in place in 83 % of the 18 countries. That means in countries where
there is currently no automation, it is not possible to acquire automatically images of mail pieces when
they are sorted.
In these countries, the cross-border mail volume (export and import) is low. In some cases, such as
manual sorting at a frame, it could be possible to envisage the integration of a camera device at each
operator position. An image of these manually sorted letters will be automatically acquired when the
mail piece is handled by the operator.
Where there is currently no automation in place, a modification of the existing operational environment
will be needed for the acquisition of images.
9.3 Key points in question 5.1
39 % of postal operators indicate that mail preparation per destination may be carried out by big
customers. The questionnaire was referring only to SPPM (Single Piece Priority Mail) processing, but
for the question 5.1, some posts may have included the processing of bulk mail in their answer. If it is
the case, the percentage of 39 % may be much too high.
If in some cases the preparation per destination is performed by the sender, these mail pieces will not
be reprocessed in a sorting equipment at the outbound OE before being dispatched to the destination
country. The images of such pre-sorted letters will not be taken into account in the Real Mail
Measurement of the cross-border traffic.
9.4 Key points in question 12
3 statements on meter mail were included in the questionnaire as there are some requirements in the
control of information when printed in the franking mark.
72 % of postal operators indicate that the date of posting is written in a readable (question 12.1). For
the check of the accurate date of posting, only 39 % of postal operators (question 12.2) indicate that
they are controlling this date and only 22 % reject the mail pieces and request the sender to reprint this
meter mail as it is non-compliant to the mandatory rules.
The date printed in the franking mark of the meter mail cannot be considered as a valid date of posting
for all postal operators (for those that have intensive controls on posting marks on the envelopes, this
could however be considered as valid information). In general, this printed date can only be considered
as an information among others to facilitate the correct matching between images.
12
---------------------- Page: 14 ----------------------
SIST-TP CEN/TR 17386:2019
CEN/TR 17386:2019 (E)
9.5 Key points in question 14
89 % of postal operators indicate that mail is sent in bags to countries where the cross-border volume
is low. That means, the sequence of mail as sorted in trays in the automatic equipment will be only kept
for a handful of mail pieces. The transfer of mail pieces – from trays to bags in the outbound OE and
then from bags to domestic trays in the inbound OE – may reduce the confidence level required for
matching the images.
It will be necessary to confirm that the transfer of mail pieces in different types of containers in the
delivery chain has not a detrimental effect in the Image Matching performance. For the domestic
processing and automatic sorting, the mail pieces are always transported in the same type of container
(tray) from the first sorting site up to the delivery office with no transfer in temporary containers such
as bags.
Some posts indicated that for the transport to EU destinations, there is not only mail transfer in bags;
but also mail transfer in the standard IPC tr
...
SLOVENSKI STANDARD
kSIST-TP FprCEN/TR 17386:2019
01-junij-2019
Poštne storitve - Merjenje prehodnih časov za čezmejne poštne pošiljke z uporabo
študije realnih poštnih zmogljivosti
Postal services - Transit time measurement for cross border postal items using real mail
feasibility study
Postalische Dienstleistungen - Messung der Durchlaufzeit von grenzüberschreitenden
Postsendungen unter Nutzung von echten Sendungsdaten
Services postaux - Mesure du délai d’acheminement du courrier transfrontière à partir
des flux réels - Rapport de faisabilité
Ta slovenski standard je istoveten z: FprCEN/TR 17386
ICS:
03.240 Poštne storitve Postal services
kSIST-TP FprCEN/TR 17386:2019 en,fr,de
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.
---------------------- Page: 1 ----------------------
kSIST-TP FprCEN/TR 17386:2019
---------------------- Page: 2 ----------------------
kSIST-TP FprCEN/TR 17386:2019
FINAL DRAFT
TECHNICAL REPORT
FprCEN/TR 17386
RAPPORT TECHNIQUE
TECHNISCHER BERICHT
March 2019
ICS 03.240
English Version
Postal services - Transit time measurement for cross
border postal items using real mail feasibility study
Services postaux - Mesure du délai d'acheminement du Postalische Dienstleistungen - Messung der
courrier transfrontière à partir des flux réels - Rapport Durchlaufzeit von grenzüberschreitenden
de faisabilité Postsendungen unter Nutzung von echten
Sendungsdaten
This draft Technical Report is submitted to CEN members for Vote. It has been drawn up by the Technical Committee CEN/TC
331.
CEN members are the national standards bodies of Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia,
Finland, Former Yugoslav Republic of Macedonia, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania,
Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Romania, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland,
Turkey and United Kingdom.
Recipients of this draft are invited to submit, with their comments, notification of any relevant patent rights of which they are
aware and to provide supporting documentation.
Warning : This document is not a Technical Report. It is distributed for review and comments. It is subject to change without
notice and shall not be referred to as a Technical Report.
EUROPEAN COMMITTEE FOR STANDARDIZATION
COMITÉ EUROPÉEN DE NORMALISATION
EUROPÄISCHES KOMITEE FÜR NORMUNG
CEN-CENELEC Management Centre: Rue de la Science 23, B-1040 Brussels
© 2019 CEN All rights of exploitation in any form and by any means reserved Ref. No. FprCEN/TR 17386:2019 E
worldwide for CEN national Members.
---------------------- Page: 3 ----------------------
kSIST-TP FprCEN/TR 17386:2019
FprCEN/TR 17386:2019 (E)
Contents Page
European foreword . 3
Introduction . 4
1 Scope . 5
2 Normative references . 5
3 Terms and definitions . 5
4 Symbols and abbreviations . 6
5 Existing QoS measurement system . 6
6 Operational measurement . 7
7 Objective of the study . 8
8 Existing and proven technology . 9
9 Key points in the results of the survey . 12
10 Answers to the research questions 1 to 4 . 13
11 Decision of CEN/TC 331 Postal services regarding the feasibility study . 15
Annex A (informative) Questionnaire sent to 32 EU postal organizations . 16
Annex B (informative) Results of the survey filled out by 18 Eu postal organizations . 21
B.1 General remarks . 21
B.2 Results . 21
2
---------------------- Page: 4 ----------------------
kSIST-TP FprCEN/TR 17386:2019
FprCEN/TR 17386:2019 (E)
European foreword
This document (FprCEN/TR 17386:2019) has been prepared by Technical Committee CEN/TC 331
“Postal services”, the secretariat of which is held by NEN.
This document is currently submitted to the Vote on TR.
This document has been prepared under a mandate given to CEN by the European Commission and
supports essential requirements of EU Directive(s).
3
---------------------- Page: 5 ----------------------
kSIST-TP FprCEN/TR 17386:2019
FprCEN/TR 17386:2019 (E)
Introduction
Under the current Standardization Request M/548, CEN and CEN/TC 331 Postal services identified a
need for a feasibility study to use real mail processing in the measurement of the quality performance.
This feasibility study has been carried out, researching the costs versus the benefits of a method for
transit time measurement for cross border postal items using real mail.
Based on this study, is became clear that the time and resources needed to develop such a method are
too extensive to justify it. Therefore, the members of CEN/TC 331 Postal services have decided that this
method is not feasible and that the method will not be developed within the current Standardization
Request M/548.
4
---------------------- Page: 6 ----------------------
kSIST-TP FprCEN/TR 17386:2019
FprCEN/TR 17386:2019 (E)
1 Scope
The scope of this document is a feasibility study that was carried out to explore the use of real mail data
in measurement of the transit time of end-to-end services for single piece cross-border priority mail. In
this document a description is given of the context, the way this study was carried out, the results of the
study and the advice given to CEN/TC 331 Postal services and, finally, CEN and the European
Commission.
2 Normative references
The following documents are referred to in the text in such a way that some or all of their content
constitutes requirements of this document. For dated references, only the edition cited applies. For
undated references, the latest edition of the referenced document (including any amendments) applies.
EN 13850:2012, Postal Services - Quality of Services - Measurement of the transit time of end-to-end
services for single piece priority mail and first class mail
3 Terms and definitions
For the purposes of this document, the following terms and definitions apply.
ISO and IEC maintain terminological databases for use in standardization at the following addresses:
• IEC Electropedia: available at http://www.electropedia.org/
• ISO Online browsing platform: available at http://www.iso.org/obp
3.1
International Postal Corporation
IPC
cooperative association of postal operators in Asia Pacific, Europe and North America
3.2
Office of Exchange
OE
postal facility which handles the foreign mail departing to and/or arriving from another country
3.3
PostEurop
trade association that represents European public postal operators
3.4
Research question
RQ
investigations carried out as part of the feasibility study
3.5
Sequencing sort
sorting of the mail pieces in the order of the postman delivery
3.6
Test Letters
TL
tracked and traced letter with which the cross-border time is measured between OE’s
5
---------------------- Page: 7 ----------------------
kSIST-TP FprCEN/TR 17386:2019
FprCEN/TR 17386:2019 (E)
3.7
UNEX™
postal QoS system of IPC which contains several specific modules and is used to measure international
letter service performance between 37 postal operators using test mail methodology, one of which
modules measures the end-to-end postal quality of service for single pieces priority in Europe
4 Symbols and abbreviations
CFC Culler Facer Canceller
DPI dots per inch
FC Facer Cenceller
KPI Key Performance Indicator
OCR Optical Character Recognition
QoS Quality of Service
RFID Radio Frequency Identification Technology
RMM Real Mail Measurement
RoI Return of Investment
SPPM Single Piece Priority Mail
USP Universal Service Provider
5 Existing QoS measurement system
The IPC UNEX™ CEN module measures the transit time of end-to-end services for single piece priority
and first class mail in Europe, according to the standard EN 13850:2012 Postal Services – Quality of
Services – Measurement of the transit time of end-to-end services for single piece priority mail and first
class mail. The measurement is end-to-end from posting in the original country (C) to delivery to the
final addressee in the destination country (I). This includes the time of collection, sorting and
transportation as indicated in Figure 1.
Figure 1 — Postal process from origin country to destination country
This performance measurement monitored since 1994, indicated the percentage of mail flows reaching
the final addressee within J+3 and J+5 after posting. The percentage objectives have been set
respectively to 85 % (J+3) and 97 % (J+5) (see the 1997 European Union on Postal Services Directive
(97/67/EC Directive)).
In order to accurately evaluate these two performances, CEN Test Letters representative of the real mail
streams are produced and panellists located in urban and in rural areas are in charge to record the
posting and receiving dates of these test letters. In addition to this manual operation, there are around
6
---------------------- Page: 8 ----------------------
kSIST-TP FprCEN/TR 17386:2019
FprCEN/TR 17386:2019 (E)
40 % of these test letters with a specific RFID device which can be read automatically when the letter is
under a RFID antenna located at the entrance and at the exit of each sorting centre. That means for
these 40 % of test letters it is possible to record automatically the date when the letter is detected at the
various locations indicated in Figure 1: D E F G H.
In 2016 and 2017, 797 mail flows have been measured between 32 EU countries. There were around
1,200 panelists handling around 66 000 test letters per year.
The following Figure 2 provides the J+3 performance (also identified as speed indicator) across years
since 1997.
Figure 2 — J+3 performance across years
6 Operational measurement
In order to process the cross-border mail flows between EU countries, bilateral agreements between
USP in each country are in place in order to perform the transport from sender to receiver.
The transport between the sender (C) and the receiver (I) requires services to be delivered by each USP
in country A and in country B in addition to the international transport - see Figure 3.
Three different legs have been identified:
— Leg 1 corresponds to the transport from the sender in country A up to the hand-over to the first
carrier for the international transport – ref to segment C to E in Figure 1;
— Leg 2 corresponds to the international transport up from end of Leg 1 up to the hand-over to the
USP of the receiving country B. Multiple carriers may be involved – ref to segment E to F.in Figure 1;
— Leg 3 corresponds to the transport from end of Leg 2 up to the delivery to the addressee in country
B – ref to segment F to I in Figure 1.
7
---------------------- Page: 9 ----------------------
kSIST-TP FprCEN/TR 17386:2019
FprCEN/TR 17386:2019 (E)
Figure 3 — Operational measurement
For the traffic from country A (sender) to country B (receiver), the Operational Responsibility is defined
as the following:
— The Inbound Post B is only responsible for Leg 3;
— Leg 2 which may include multiple carriers is the sole responsibility of the Outbound Post A in
addition to the Leg 1.
This operational responsibility is not related as such to the UNEX™ measurement but is an agreed split
of responsibilities between all the shareholders for cross-border mail traffic.
As there are sufficient information with the 40 % of test letters automatically detected due to presence
of the RFID component transported in these test letters and other messages transmitted between the
postal organisations, carriers, etc., it is possible for IPC to precisely determine the operational
measurement Leg 1, Leg 2 and Leg 3 for the test items equipped with RFID.
7 Objective of the study
7.1 Objective
The feasibility study shall answer the question of whether new techniques also using methods of
tracking and tracing of cross border letter mail within the EU could lead to a reliable and diagnostically
useful end-to-end survey to measure the transit time of end-to-end services for single piece cross-
border priority mail.
The method to perform this task is listed in the following four Research Questions:
7.2 Research questions
7.2.1 Question 1
Can images captured by sorting machines in sorting centres and offices of exchange per item
throughout Europe (EC member states) be stored and used in a survey in such a way that
— Individual mail items can be identified and;
— The date of posting (cancelling information with dates on stamps or the date in the franking image)
and tracking data per item throughout a big part of Europe is registered?
8
---------------------- Page: 10 ----------------------
kSIST-TP FprCEN/TR 17386:2019
FprCEN/TR 17386:2019 (E)
7.2.2 Question 2
Can postal operators make these data available, also from a legal perspective?
7.2.3 Question 3
Can real mail data such as described in the questions 1 and 2 be completed with data in order to create
an end-to-end result?
7.2.4 Question 4
Can data from domestic surveys be used in combination with real mail track and trace data, as meant in
question 3, to produce an end-to-end survey to assess the norms for cross border mail within the
European Union in the Postal Directive 2008/6/EC?
7.3 Summary of the purpose of the feasibility study
— to analyse the possibility of acquiring automatically images of letters processed in sorting machines
installed in Office of Exchange (or in a sorting centre close to the OE) in both countries (sender and
receiver – from country A to country B as indicated in Figure 1).
— to be able to recognize individual mail items from image identification.
— to be able to track and trace individual items between OEs and to measure the time spent during
the international transport.
— to analyse the possibility to complement this international transport measurement with the
domestic QoS measurement in country A and country B in order to create an end-to-end result as
required by the CEN Standard EN 13850:2012 Postal Services - Quality of Services - Measurement of
the transit time of end-to-end services for single piece priority mail and first class mail.
8 Existing and proven technology
8.1 Image quality
This feasibility study has been launched based on the regular improvements made either in image
quality of the cameras and in image matching technology during the last several decades:
— The development of automatic mail sorting equipment started in the 70s,
— A decade later in the 80s, automatic OCR (optical character recognition) was gradually introduced
dealing with typewritten and then handwritten address recognition.
— During the next 10 years, several recognition engines were combined in the same logic to carry out
the full interpretation of addresses written on the letters enabling the domestic sorting up to
sequencing which is the last sortation phase before the postman delivery.
In order to perform the full address identification only once (at the start of the domestic processing
chain), a unique barcode (also called Tag Id code) is printed on each mail piece in order to easily
identify this letter all along the journey from the sender to the final addressee. All information required
for processing are associated to this unique Tag Id code. The sorting information is retrieved when the
Tag Id of the letter is detected and read in a sorting equipment.
9
---------------------- Page: 11 ----------------------
kSIST-TP FprCEN/TR 17386:2019
FprCEN/TR 17386:2019 (E)
This barcode printing solution has a number of advantages to facilitate the tracing and tracking of each
mail piece, but has also some constraints which have to be bared by postal operators, mainly due to the
costs associated to
— The barcode printer,
— The ink,
— The maintenance and spare parts,
— The need to apply an additional label when the mail piece is plastic wrapped (i.e. magazine) as
printing on the plastic is not possible at the speed of the transport.
— The cost of the label applicator and the cost of labels applied on each item.
In 2000s, due to all these operational constraints, new technological solution of mail piece identification
based only on image analysis and image matching has been introduced removing the need of applying a
barcode on mail pieces and therefore all the associated above-mentioned costs.
8.2 Image analysis
The objective of the image analysis is to extract a Set of Parameters (SoP) which characterizes the image
of each letter. This SoP, also called the mail piece signature, remains nearly invariant when another
image of the same letter is acquired by another machine in another sorting centre.
In order to extract an accurate signature, the image quality should be at least a multi-grey level image
with a minimum definition of 8 pixels per mm (or 200 dpi – dot per inch).
During the export sorting:
— The destination country is identified by reading the address,
— The mail pieces are sorted to the machine outputs according to the OE destinations,
— Sample of letters (also named Test Letters - TLs) are selected in the different mail flows and the
signatures of these TLs are recorded in files (a file per OE destination),
— Associated to each Test Letter, the sequence of mail pieces processed before and after each TL for
the same OE destination are analysed and their signatures are recorded in the same data file for
that OE destination.
— Then, during the transport of the physical mail pieces to the inbound OEs, the data files containing
all these signatures of Test Letters and other associated mail pieces are electronically sent their
relevant OEs destinations.
8.3 Image matching
For cross border mail flows, the Image matching is the process performed in each inbound OE where
the image signatures of imported mail flows are compared with the signatures of TL (Test Letters)
acquired in all outbound OEs (see processes a to c as described above).
For each image processed in the inbound OE, the matching may provide several TL candidates of images
acquired in the outbound OEs, each candidate having its own confidence level value. The final decision
is made after analysing the letters sorted before and after each TL candidate – see “step d” as described
in Image Analysis paragraph above.
10
---------------------- Page: 12 ----------------------
kSIST-TP FprCEN/TR 17386:2019
FprCEN/TR 17386:2019 (E)
As a result, the image matching function is not only a comparison between two images acquired in 2
sorting centres, but a more complex comparison between of a sequence of letters processed in Offices of
Exchange, firstly in the outbound OE and then the inbound OE.
8.4 Volume of data associated to image analysis and matching function
The most important factor to be taken into consideration for integrating these two functions - Image
Analysis and Image Matching - is the volume of data to be stored and managed.
The size of a multi-grey level mage with a definition of 8 pixels per mm (in a compressed format) varies
between 350 Kbytes (for small letters) and 700 Kbytes (for large letters). An average value can be
estimated to 450 Kbytes considering the ration of flats (20 %) and letters (80 %) in the live mail.
Therefore, the data storage for 100 000 images is estimated to 45 GB
The data generated for an image signature depends on the quantity of information contained in the
image, but an average value is estimated to 30 KB. For the same quantity of images (100 000 images),
the storage of all associated signatures is estimated to 3 GB
For an OE where the volume of mail to be sent abroad to EU countries is around 500 000 letters per
day; the required the total storage capacity required will be around 240 GB per day.
This evaluation can be updated to consider the low volume processed in some of the EU countries. For
the other countries, the integration of the real mail measurement will require a storage capacity close to
Terabytes as there is a need to consider several days for the tracking of the cross-border mail flows.
As a consequence, “Big Data” technology is clearly needed to be able to efficiently manage such data
volumes. The other optimization is the quantity and type of data to be stored at the different data
centres: the outbound and inbound OEs and at the central data processing and management. The trade-
off is between real time processing (i.e. matching images when import mail arrives in inbound OE) and
non-real time processing off-line (i.e. data transmission between OEs when the physical mail is
transported between the same OEs).
8.5 Proven technology
This technology of tracing and tracking mail pieces using only images with no barcode printed on mail
items has been integrated by a few postal solution suppliers. As some of them have communicated on
this subject via their website, it is a public information which can be listed hereafter:
Siemens Postal and Parcel Automation indicates that they integrated the Mail item identification with
fingerprint technology in the flat sorting systems delivered to Deutsche Post in Germany.
Solystic indicates that they integrated the tracing of letters using the image with no additional barcode
printed for the postal operator in the following countries: France, Finland and UK. This technology
based on image tracking is called “V-Id” which means “Virtual Identification”.
In these EU countries; postal operators integrated such solution mainly for the domestic sorting but
also for foreign mail in order to be able to video-code these items in an off-line mode (the mail pieces
have to be identified when reprocessed the second time and printing barcode on foreign mail is not
allowed).
All these methods of tracking and tracing mail pieces for all flows (small letters and flats) are now in full
operation for more than10 years. This technology has been really proven.
11
---------------------- Page: 13 ----------------------
kSIST-TP FprCEN/TR 17386:2019
FprCEN/TR 17386:2019 (E)
9 Key points in the results of the survey
9.1 Overall remarks
The questionnaire includes 30 questions. An overview of the answers of the posts which provided their
feedback is attached in Annex B.
The following questions brought forward key points that may have a direct impact on this Real Mail
Measurement feasibility study: Question 1, 5.1, 12, 14 and 30. What follows is a summary of those key
points.
9.2 Key points in question 1
Automatic sorting equipment are in place in 83 % of the 18 countries. That means in countries where
there is currently no automation, it is not possible to acquire automatically images of mail pieces when
they are sorted.
In these countries, the cross-border mail volume (export and import) is low. In some cases, such as
manual sorting at a frame, it could be possible to envisage the integration of a camera device at each
operator position. An image of these manually sorted letters will be automatically acquired when the
mail piece is handled by the operator.
Where there is currently no automation in place, a modification of the existing operational environment
will be needed for the acquisition of images.
9.3 Key points in question 5.1
39 % of postal operators indicate that mail preparation per destination may be carried out by big
customers. The questionnaire was referring only to SPPM (Single Piece Priority Mail) processing, but
for the question 5.1, some posts may have included the processing of bulk mail in their answer. If it is
the case, the percentage of 39 % may be much too high.
If in some cases the preparation per destination is performed by the sender, these mail pieces will not
be reprocessed in a sorting equipment at the outbound OE before being dispatched to the destination
country. The images of such pre-sorted letters will not be taken into account in the Real Mail
Measurement of the cross-border traffic.
9.4 Key points in question 12
3 statements on meter mail were included in the questionnaire as there are some requirements in the
control of information when printed in the franking mark.
72 % of postal operators indicate that the date of posting is written in a readable (question 12.1). For
the check of the accurate date of posting, only 39 % of postal operators (question 12.2) indicate that
they are controlling this date and only 22 % reject the mail pieces and request the sender to reprint this
meter mail as it is non-compliant to the mandatory rules.
The date printed in the franking mark of the meter mail cannot be considered as a valid date of posting
for all postal operators (for those that have intensive controls on posting marks on the envelopes, this
could however be considered as valid information). In general, this printed date can only be considered
as an information among others to facilitate the correct matching between images.
12
---------------------- Page: 14 ----------------------
kSIST-TP FprCEN/TR 17386:2019
FprCEN/TR 17386:2019 (E)
9.5 Key points in question 14
89 % of postal operators indicate that mail is sent in bags to countries where the cross-border volume
is low. That means, the sequence of mail as sorted in trays in the automatic equipment will be only kept
for a handful of mail pieces. The transfer of mail pieces – from trays to bags in the outbound OE and
then from bags to domestic trays in the inbound OE – may reduce the confidence level required for
matching the images.
It will be necessary to confirm that the transfer of mail pieces in different types of containers in the
delivery chain has not a detrimental effect in the Image Matching performance. For the domestic
processing and automatic sorting, the mail pieces are always transported in the same type of container
(tray) from the first sorting site up to the delivery office with no transfer in temporary containers such
as bags.
Some posts indicated that for the transport to EU destinations, there is not only mail transfer in bags;
but also mail transfer in the standard IPC trays.
9.6 Key points in question 30
All postal operators (100 %) indicate that inward processing steps (from G to I – Figure 1) are identical
for all domestic and imported mail flows.
For these inward steps (G to I), there are two different set of Test Letters: Domestic Test letters (for the
National QoS measurement) and cross-border Test letters (for the CEN module QoS measurement).
Even if the EN 13850:2012 standard is defined for the end-to-end performance measurement, there
should be no variation in the transport time in the section (G to I) between domestic and cross-border
Test Letters. The RFID tags of Test letters (domestic and cross-border) are detected by the same
antennas and are transferred to the same data centre. Analysis of both information could validate if
variations are (or are not) measurable between the domestic flows or cross-border flows in the mail
inward processing section G to I.
10 Answers to the research questions 1 to 4
10.1 Answers to research question 1
The question is: Can images captured by sorting machines? Can individual mail items be identified?
In the EU countries where the mail volume is high, the manual sorting is not possible to perform the
complete processing up to the delivery in a short time period (D+1) for the domestic priority mail.
Therefore, automatic sorting equipment have been installed in sorting centres and in OEs. In this
equipment there is always an image acquisition module (camera) which takes an image of each item
entering in the domestic postal network.
When an image is acquired, the signature of this image is generated. For the identification of that mail
piece when reprocessed in another sorting site in the same country or in a foreign country (inbound
OE), it is necessary to perform a valid image matching with several items (and not for only one mail
piece). In other words, the signature of an image is not e
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.