Track and Trace

Download as pdf or txt
Download as pdf or txt
You are on page 1of 16

T. Borangiu et al. (Eds.): Service Orientation in Holonic and Multi Agent, SCI 472, pp. 119134.

DOI: 10.1007/978-3-642-35852-4_8 Springer-Verlag Berlin Heidelberg 2013


Chapter 8
Assessment of EPCIS Standard for Interoperable
Tracking in the Supply Chain
Kary Frmling
1
, Sagar Parmar
1
, Ville Hinkka
1
, Jaakko Ttil
1
, and Dirk Rodgers
2

1
Aalto University, School of Science, PO Box 15500, FI-02015, Finland
[email protected]
2
RxTrace.com, USA
[email protected]
Abstract. With increasing globalization and loosely-coupled business relations
between different companies, the importance of information exchange standards
is increasing. GS1s Electronic Product Code Information Services (EPCIS) is
currently one of the main standards proposed for inter-organizational data ex-
change for track and trace purposes in the supply chain. However, despite its
strong industrial background and maturity, EPCIS is not yet used as a global
standard for trace and track applications. This paper attempts to analyze why
EPCIS has not been universally adopted as a global data-exchange standard for
track and trace by analyzing three application areas where the use of EPCIS
should provide substantial benefits. The results of this analysis should provide
useful insights into the challenges of introducing new, global standards and
guidelines for similar initiatives in the future.
Keywords: Track and trace, supply chain management, inter-organizational da-
ta exchange, interoperability, EPCIS.
1 Introduction
Inter-organizational data exchange has probably existed in some form for nearly as
long as there has been life on our planet. Data exchange between humans and other
living beings, as well as communities and organizations, is performed by sounds,
gestures, chemical substances, light, smoke or other media. In our current society,
paper-, telephone-, fax-based etc. technologies have been the state-of-the-art during
the 20
th
century. Now these technologies are being replaced or complemented by
technologies using computer networks, where the internet has become the main net-
work for inter-organizational data exchange.
Computer networks make it possible to exchange information much faster than
before, including the possibility to process the data automatically. Also, data trans-
mission delays are no longer dependent of geographical distances. These new possi-
bilities enable organizations to set up inter-organizational operations much faster and
with little or no human intervention, as long as different organizations implement the
same protocols for data exchange. Standards are the cornerstone for enabling such
120 K. Frmling et al.
implementations where the implementation effort does not depend on the number of
organizations that exchange data. Successful examples of such standards in Supply
Chain Management (SCM) are Electronic Data Interchange (EDI) and RosettaNet
standards.
Supply chains are getting more and more geographically spread and loosely
coupled [1], which signifies that they need to be able to set up new supplier relation-
ships at a higher pace than before. Concepts used for this kind of loosely coupled
supply chains are Virtual Enterprises [2] and Extended Enterprises [3]. Loose coupl-
ing is particularly challenging for inter-organizational data exchange, which still to-
day often requires long and expensive setup of EDI communication. Even after EDI
integration, supply chains have great challenges in implementing fundamental opera-
tions such as tracking shipments and deliveries, as well as handling after-sales and
service operations on product individuals. The advance shipping notice (ASN) in EDI
is the message that gets the closest to shipment tracking but it is not suitable or in-
tended even for shipment tracking. This is why most shipment tracking systems are
organization, specific, such as those provided by companies like FedEx, UPS etc.
In order to implement shipment tracking, and after-sales and service operations on
product individuals and so on, it is necessary to identify shipments and products as
individual instances rather than just belonging to some product category. Company-
specific tracking or serial numbers are currently the most used identifiers for product
individuals. However, due to their company-specific nature, they are not suited for
inter-organizational data exchange. One example of a globally accepted and widely
used shipment individual identifier is the serial shipping container code (SSCC) that
is standardized by GS1. The SSCC can easily be used with barcodes. However, the
advances in radio frequency identification (RFID) technology, as well as the decreas-
ing cost of implementing that technology, has opened up completely new possibilities
for the identification of shipment and product individuals. RFID supports product
individual-level identification nearly by definition because all RFID tags are identi-
fied by a unique tag identifier. RFID also provides new application opportunities be-
cause RFID does not need a line of sight, so many identifiers can be read with one
single read operation. Such properties signify that many kinds of inventory operations
can be made automatic, thus providing an increased accuracy of material flows while
speeding up operations and reducing the need for manpower.
The use of RFID tags and product individual-level identification in general creates
challenges for existing SCM information systems. Several initiatives have been
created for implementing inter-organizational data exchange protocols and interfaces,
such as the Auto-ID Centre at MIT, the DIALOG initiative at Helsinki University of
Technology [4] and the peer-to-peer based paradigms of the company Trackway [5].
Elements developed by the MIT Auto-ID Centre were taken up by the non-profit
company EPCglobal, which later became a part of GS1. Especially the electronic
product code (EPC) and the related radio interface protocols defined by EPCglobal
have been undeniable successes as standards for UHF-based RFID tags. UHF tags are
the most used ones in SCM applications due to their long read range, which can be
over several meters in optimal conditions. It is also possible to read hundreds of tags
with one single operation, therefore providing considerable advantage in reception
points for incoming goods, for taking inventories, etc.
EPCglobals information system framework is called the EPC network [6]. It con-
tains specifications for managing RFID readers and collecting information from them
8 Assessment of EPCIS Standard for Interoperable Tracking in the Supply Chain 121
(Application Layer Events, ALE), the EPC Information System (EPCIS) and other
components for looking up information sources about shipments and products. EPCIS
is the standard proposed for being used for inter-organizational data exchange in the
EPC network. When considering the background of EPCIS and the maturity of the
standard, one would expect that it would by now be universally used for tracking
items in inter-organizational settings. However, it seems like it is far from being so in
reality. The main research question set out for this paper is to what extent EPCIS is
being used in reality, as well as identifying the main application domains in which it
is used. The main objective of the paper is to analyze and understand the reasons for
which EPCIS has not yet become universally used, as well as if and when it eventual-
ly will become universally used.
After this introduction, the paper provides an overview of inter-organizational data
exchange in general and EPCIS in particular. Then we provide an overview of differ-
ent domains where EPCIS has been used to some extent and identify three domains
that are of particular interest due to their relatively advanced state regarding inter-
organizational data exchange and experiences from using EPCIS. The level of use of
EPCIS is analyzed in detail for the selected domains, followed by conclusions.
2 Background
Inter-organizational data exchange signifies the transfer of electronic documents or
business data from one computer system to another computer system, i.e. from one
trading partner to another trading partner without human intervention. Data exchange
requires some kind of physical media for transmitting the data from one system to the
other, a protocol that allows determining what system should send when, one or more
interfaces that send and/or receive data and a semantic layer for understanding the
data in a uniform way. When two humans communicate using voice only, frequency
modulation of air vibrations is the media. The protocol is determined by tone, facial
expressions, content etc. The interface for sending is the mouth and ears are used for
receiving. The brains take care of the semantic interpretation of the data received, the
processing of it and sending back new data.
All these same elements (media, protocol, interfaces, semantic interpretation) are
also present in computer-to-computer communication. A major difference between
computer-to-computer communication and human-to-human communication is that
humans can tolerate some degree of noise, errors and misunderstandings. An intellec-
tual dialogue between humans can usually filter out the worst misunderstandings.
However, computers are not yet capable of performing such dialogue, which creates
strict requirements on how unambiguously all data exchange standards must be de-
fined. The further up we go towards the semantic level, the more challenging it is to
create data exchange standards that are unambiguous, while providing a sufficient
power of expression. EPCIS is mainly an interface standard, even though it also con-
tains both protocol and semantic elements. This makes it particularly interesting to
study the level of usage of EPCIS because it may provide an indication of the chal-
lenges faced when attempting to develop standards for even higher levels of semantic
data.
122 K. Frmling et al.
2.1 Overview of Inter-organizational Data Exchange Standards
The classic example of an inter-organizational data exchange standard is traditional
EDI [7, 8]. The best known and most widely used EDI-standards are ANSI X.12
(mainly in the US) and EDIFACT (developed by UN, intended as a global standard)
[9]. EDI standards date back to 1970s and beyond; in the mid-1980 EDI use began to
expand significantly, following the introduction of the first version of ANSI X.12 in
1981 [7] and the EDIFACT starting from 1985 [10].
Lately, it has been suggested that newer standards would overcome the challenges
of traditional EDI [11] by adding to the syntactic and semantic level of EDI standards.
These newer pragmatic level standards have been claimed to help create more flexible
and economical inter-organizational system-to-system integrations. Typically, these
standards utilize the power of the XML (Extensible Markup Language), dating back
to 1996 [12]. XML defines besides the data, through the so called tags, meta-data, i.e.
data on data [10]. However, XML as such defines only the syntax for describing data
and notably not how to exchange XML-based data between the information systems
of separate organizations, nor the semantic meaning of the actual data [13].
Some relatively widely diffused examples of XML-based standards are RosettaNet
[14, 15] and ebXML (Electronic Business using eXtensible Markup Language). Ro-
settaNet is designed to fulll the growing demand for unied methods of B2B integra-
tion among buyers and sellers in the global information industry supply chain [13]. It
has gained significant footing within the semiconductor and electronic components
manufacturing industries [15]. ebXML is suggested to be the most ambitious business
process standard [16] originated from UN/CEFACT (United Nations Center for Trade
Facilitation and Electronic Business) and OASIS (Organization for the Advancement
of Structured Information Standards) in late 1999. ebXML and its related standards
attempt to found a universal electronic market and similar regulations. With ebXML,
companies from various industries can discover applicable trading partners dynami-
cally, negotiate trading process automatically, and conduct business transactions [16].
In this paper we are focusing on inter-organizational exchange of shipment track-
ing data, as well as data related to product individuals in general. Our own work in
this area started in 2000 with the goal to create systems for implementing the Internet
of Things (IoT). In 2001 the first version of such a system was implemented under the
name DIALOG (Distributed information architectures for collaborative logistics)
[17]. DIALOG was used in multi-organizational and international pilots for tracking
shipments and warehouse items in 2002 [18] and 2003 [19]. DIALOG was also used
and demonstrated for storing and retrieving product individual information in after-
sales and maintenance settings, as well as for many kinds of consumer applications. In
later years, DIALOG has mainly been used for asset management, after-sales service
and maintenance and product individual lifecycle management applications [20]. We
are also involved in a standards initiative called Quantum Lifecycle Management with
the Open Group (http://www.opengroup.org/qlm/) for defining data exchange stan-
dards that can be generally applicable to handling any kind of product individual data.
It is in this context that we have been closely following the evolution of EPCglobal
standards as a potential candidate for use in our own applications. EPCIS [6] is the
standard that is technically closest to the kind of data exchange that we have needed
over the years. Unfortunately EPCIS still seems suitable mainly for SCM but we still
8 Assessment of EPCIS Standard for Interoperable Tracking in the Supply Chain 123
think that an analysis of the evolution and the degree of acceptance and use of EPCIS
in real applications can provide valuable insight about the actual need and readiness
for inter-organizational data exchange, as well as about the challenges related to the
launch of standards in this domain.
2.2 Overview of EPC Network
GS1s EPC is a unique code that is assigned to each good or material tagged and is
read when passed through a RFID reader. The RFID technology and associated com-
puter networks use the EPCglobal Network and its EPCIS as an interface to enable
EPC-related data to be captured and queried using a defined set of operations and
associated EPC-related standards, combined with security mechanisms that satisfy the
needs of the parties involved [6]. In other words, EPCIS provides a standard interface
for storage and access of EPC-related data that can be read and written by the autho-
rized parties involved. The hardware and the software components of the framework
can be grouped into important entities, namely
1. EPC tags and readers where tags communicate their EPC information to readers
via radio waves and deliver data to information system of parties involved through
EPC Middleware,
2. EPC Middleware that acts as an interface between read events and companys oth-
er existing data information systems and also for communication with EPCIS,
3. EPCIS, which enables partners to exchange data and events throughout the supply
chain. The EPC Information Services consists of capture, Query Interfaces and a
repository for storage of event data. The Capture interface collects and delivers
EPC event data to the repository and to the accessing applications. An EPCIS Ac-
cessing application is responsible for processes such as warehouse management,
shipping and receiving, and output analysis facilitated by EPC-related data. EPCIS
Query application provides a mechanism for authentication of data exchange be-
tween two parties and facilitates a provision to limit or control the scope of data
that is being accessed by different parties.
4. Object Naming Service (ONS): A look up service that takes an EPC as input and
produces output in the form of a Uniform Resource Locator (URL) of an EPCIS
repository.
5. EPCIS Discovery Services that facilitates the users to find and request access to
specific EPC data.
An RFID tag attached to a trade item contains a unique EPC that globally identifies
that item while it is in the supply chain. The EPC would typically be encoded with a
Serialized Global Trade Item Number (SGTIN) which contains: The EPC Manager
Number (company specific code), Object Class (product number), and Serial Number
for each object tagged. These tags are scanned by readers, which send EPC informa-
tion contained in the tags to EPC middleware. The middleware software filters, col-
lects and stores the information in the EPCIS repository, queries the ONS to find
information related to that product and then gets supply chain event data about that
product from the EPCIS repository of the parties involved. Access to the EPCIS
124 K. Frmling et al.
servers of other trading partners needs authorization and authentication based on pre-
determined business contracts [6]. EPCIS is a standard that is intended for use in track
and trace applications. In a nutshell, it holds the WHAT, WHEN, WHERE and WHY
details of each individual unit of the products moving through the supply chain.
3 Case Studies
To illustrate the extent of EPCIS usage and potential challenges for its deployment,
authors chose to examine different EPCIS use cases. Firstly, use cases were exten-
sively searched from a variety of sources, including article databases, websites and
expert statements. It was found out that EPCIS standard has been used and experi-
mented (term used by several interviewees) in many successful pilots such as in
Swedish Fisheries [21], in electronic pallet tracking [22], in Norwegian Meat manu-
facturing company Nortura, by retailer Wal-Mart and in Hong Kong baggage han-
dling and pharmaceutical manufacturing [23], as an example. Bridge EU project has
also been using the EPCIS standard [24]. Many IT service providers such as Ax-
way, IBM, and Oracle provide EPCIS-based track and trace solutions to customers
[25-27]. The most prominent application domains were found to be SCM, e-pedigree
in the US, and tracking and tracing in general. It became evident eventually that use
cases at least with a fully-fledged EPCIS implementation were not of ample
supply. Thus, three use cases with a relative strong advancement or potential in
EPCIS usage were chosen. These three cases are e-pedigree in the US pharmaceuti-
cal supply chain, Railways and Technical trade. The former two were chosen
due to their apparent advancement and maturity with EPCIS compared to other do-
mains. The third case was chosen because of its potential for a wide-scale use for
EPCIS and also because it was a research project in which some of the authors parti-
cipated, giving first-hand view on the issues of EPCIS implementation.
The three chosen cases are analyzed in the following subsections. The structure of
the analyses proceeds as follows: Firstly, each case is introduced and its state of RFID
deployment is discussed. Thereafter, the use of inter-organizational data exchange and
EPCIS is analyzed. Finally, challenges are discussed and conclusions drawn.
3.1 e-Pedigree
The pharmaceutical industry exhibits one of the most complex supply chains involv-
ing many trading partners including manufacturers, re-packagers, distributors, third
party logistics providers, hospitals, clinics and retail pharmacies [28]. Approximately
a billion sellable units moved through this intricate chain in the United States (US) in
2008 [29]. The health care industry is under constant pressure to supply prescription
drugs to patients safely and securely and, at the same time, reduce the overall costs
and improving service levels [25, 30].
Drug-counterfeiting poses a serious threat to patient safety and also harms the im-
age of manufacturers, their profitability and revenue [28]. According to WHO reports,
the illegal sale of counterfeit drugs globally results in loss of revenues to the tune of
over US $35 billion and might reach US $75 billion [30]. To combat this threat, gov-
ernments are concluding that secured and transparent distribution of drugs between
8 Assessment of EPCIS Standard for Interoperable Tracking in the Supply Chain 125
trading partners must be insured. Markets such as Turkey, Brazil, France and the US
have begun to mandate that manufacturers and distributors who supply drugs to citi-
zens conform to some sort of e-Pedigree standards, such as GS1 EPCIS or GS1 Drug
Pedigree Messaging Standard (DPMS), for tracking and tracing purposes and 2D
matrix barcodes or RFID for carrying product information on each saleable package
[27]. For effective drug identification and traceability, drugs are mass serialized
where each drug package is given a unique identification such as an EPC. Matrix
codes and RFID tags are used to encode the unique identification and make it ma-
chine readable. In some countries the government specifies the technology, but in
others it is left up to the industry to agree on standards.
The US pharmaceutical supply chain is being driven towards the adoption of ePe-
digree because of State and Federal laws. The GS1 DPMS standard was specifically
created enabling the pharmaceutical industry in the US to generate interoperable elec-
tronic document pedigrees for drugs that were serialized (California) or not serialized
(Federal, Florida and elsewhere). Because the California law is the first to require the
use of serialization, companies that are facing the future effective date of that law are
now experimenting with systems that are based on GS1s EPCIS standard.

RFID Deployment. While RFID has been around for many years, the use of the tech-
nology has spread slowly, even if most of the technical challenges even at unit-level
have been solved. There are also available possibilities for secure authentication of
RFID tags and the technology for preventing the tags from being copied even if their
use is scarce [23]. The biggest challenge in RFID deployment in the healthcare supply
chain seems to be the high cost of tags compared to widely used 2D barcodes [31]. The
extent of use of RFID tags varies according to the size of the company. Bigger distribu-
tors such as McKesson prefer to use RFID tags while average size companies like As-
traZeneca and Genzyme favor GS1 2D Datamatrix barcodes. Some companies who
make biological drugs have chosen to avoid placing RFID tags on their packages out of
concern that radio signals involved with RFID tags and readers could possibly modify
the effectiveness of their drugs, although recent research studies indicate that this is not
the case [31, 32]. Other drug manufacturers intend to employ 2D barcodes on the labels
they attach to bottles of prescription drugs, and then an RFID tag in addition to barcodes
at the case-level [31, 33]. The cost of deployment of RFID tags is higher compared to
2D barcodes at this point and for companies tagging millions of products every year the
incurred cost can be difficult to justify.
Many pharmaceutical companies have executed pilots with RFID tags at the unit-
level but only Purdue Pharma and Pfizer in the US have applied the technology for an
extended period of time [34-35]. Today it appears that most companies facing the
serialization requirement of the California law are preparing to apply 2D barcodes on
their products at the unit-level. Some are planning to make use of RFID in addition to
barcodes at the case- and pallet-levels.
Given the history of RFID experimentation in the US pharmaceutical supply chain
particularly the recent elimination of RFID at the item-level by Pfizer, a long-time
experimenter with the technology it appears that the use of RFID for compliance
with the California serialization mandate will be limited to cases and pallets for some
manufacturers [36].
126 K. Frmling et al.
EPCIS vs. DPMS. By early 2000, most trading partners in the US pharmaceutical
supply chain were sharing basic data of shipments and payments received electroni-
cally at least between manufacturers, the larger distributors and larger chain phar-
macies. Companies in the US pharmaceutical supply chain showed little interest in
serialization and track and trace until 2004 when the State of California, and later
Florida, enacted a prescription drug pedigree law that included a requirement that all
drugs distributed within their borders contain a unique identifier. The effective date of
these requirements continued to drift outward until 2008 when the state legislature set
it as 2015 and 2016 for drug manufacturers, mid-2016 for distributors and mid-2017
for pharmacies. EDI was used for sharing data but it lacked the flexibility and control
to identify and track the entry of counterfeit drugs in the supply chain.
In 2006 GS1 ratified a document-based pedigree standard, known as the GS1
DPMS, which is used by US drug distributors in the State of Florida to comply with
their document-based, non-serialized drug pedigree regulation. The GS1 DPMS stan-
dard defines a complex XML schema that allows an entire supply chain history of
events for one or more drug packages to be stored in a single XML file. DPMS ePedi-
grees are exchanged between trading partners in the same way that EDI documents
are. The type of events stored within the single file bear a resemblance to the EPCIS
Commissioning, Shipping and Receiving events.
Larger pharmaceutical manufacturers and distributors such as AstraZeneca,
Genzyme, and AmerisourceBergen have slowly begun to embrace the idea of building
applications around GS1s EPCIS standard to help them solve the new challenges
with capturing and storing serial number events [27, 33, 37]. AmerisourceBergen has
experimented with IBMs EPCIS solution in the past which also supports DPMS.
EPCIS is their preferred mechanism of data exchange, but they also plan to have the
ability to receive pedigree data in the DPMS format [26].
A larger group of IT giants including IBM, SAP, Axway, and Samsung are sup-
porting and providing track and trace solutions based on the EPCIS standards [38].
Axway has been particularly active in promoting the development of EPCIS-based
standards for monitoring the pharmaceutical supply chain. A growing number of suc-
cessful pilots and projects have been carried out using EPCIS. Axways solution also
meets most of the global compliance regulations of many countries such as Brazil,
Turkey, Israel and the US [27].
While EPCIS helps in the products track and trace by the trading partners, by
sharing information on the products visibility, DPMS ensures the products security
and authenticity. The problem is that the type of inter-organizational data exchange
that the EPCIS standard query interface implements is not document-based. In gener-
al, track and trace systems based on GS1s EPCIS standard can be used for sharing
supply chain event data about serialized products. Though, it still does not have some
of its parts well-defined and it lacks the specificity needed to implement all of the
pedigree laws in different countries today. Secondly, DPMS is not interoperable
with other systems and it has several non-compliance issues that must be dealt with
first. A mechanism combing the advantages of both EPCIS and DPMS needs to be
developed.

8 Assessment of EPCIS Standard for Interoperable Tracking in the Supply Chain 127
Challenges. By design, using the standard EPCIS query interface as the data ex-
change mechanism between trading partners would result in distributing snippets of
pedigree information for each package of drugs across all of the repositories of the
previous owners of those drugs. The result is some amount of confusion about what
path companies should take.
Confusion has been inadvertently introduced into the US pharmaceutical supply
chain by the recent activities of the GS1 Traceability in Healthcare, Network Centric
ePedigree (NCeP) work group which is made up of a few larger companies in US
pharmaceutical supply chain. Their goal was to define one or more ways to produce
an interoperable electronic pedigree system that makes use of the GS1 EPCIS stan-
dard, not only for event data capture, but also for inter-organizational event data
exchange [40]. The group has recently published descriptions of the operational cha-
racteristics that would result from seven different NCePs models for consideration by
regulatory bodies who might want to enact new pedigree legislation [39]. In particu-
lar, it is hoped that the US Congress may enforce a new law that can be met by one of
the seven NCeP models and that DPMS will not be needed.
Confusion over which approach to the inter-organization ePedigree data exchange
will be necessary to remain compliant with state and federal regulations in the future
is causing an understandable hesitancy to invest in solutions. While this is causing
slow adoption of EPCIS in the US pharmaceutical supply chain it is likely that the
confusion will be resolved by the end of year 2012 when the US federal government
either takes a specific action or takes none at all. In either case, there could be a surge
of adoption of EPCIS as part of the ePedigree deployments by supply chain members
around that time. While there are known examples of successful implementation of
systems based on the EPCIS standards on one hand, there are certain obstacles that
hold this standard from actually taking off in the pharmaceutical supply chain for
applications other than ePedigree compliance. Amongst them, inter-organizational
data exchange security issues are primary. Companies might not be willing to share
certain data beyond their boundaries and especially with their competitors. Discovery
Services, a standard planned by GS1 to address this issue must still be developed and
ratified, and then must be accepted and adopted by all pharmaceutical companies
worldwide. Worldwide adoption of the EPCIS standard is a very slow process.
The pharmaceutical supply chain is very complex involving the movement of huge
numbers of drugs and information across it. With companies needing to implement
item-level serialization, the key issue is to comprehend the supply chain based on
large amounts of EPCIS data. The challenge here is to extract useful information by
combining human reasoning and analysis frameworks, reduce complexity of the data
and make informed decisions based on it. Secondly, according to Fabian and Gnther
[40], EPCIS is prone to attacks from the internet and other possible hacking exploits.
Attackers or hackers might gain control of servers/databases handling EPCIS infor-
mation and forge them. This could be a serious security breach for companies and can
increase their operational business risk. Systems to tackle these problems need to be
improved and developed for the worldwide use.
128 K. Frmling et al.
3.2 Railways
Railways have been an early domain of RFID usage. Already in the 90s, RFID was
applied in railways in several countries in Europe, Asia and Americas [41], and nearly
all railcars were RFID tagged in the US by 1994, for example [42]. In Europe, there
has also been considerable interest in RFID in railway operations lately. It has either
been discussed, piloted or even deployed in several European countries and also in
Russia. The main usage has been tagging of wagons or locomotives, which enables
automated identification and tracking of rolling stock units. With the use of RFID or
other advanced tracking systems arises the question of how to transfer and share the
generated tracking information intra- and especially inter-organizationally. Informa-
tion sharing of tracking information in railways is discussed in this section mostly
from the case perspective of Finnish and Swedish railway operations. Both the Fin-
nish and Swedish railway operators and administrators have been early-adopters of
RFID tracking and have collaborated in their research and design of the systems. The
data sources for the case are public documents and four informants in total from the
Finnish Traffic Agency (FTA), VR (a state-owned Finnish railway operator) and
Swedish Transport Agency (STA).

RFID Deployment. Enabling and pushing inter-organizational data exchange for-
ward, RFID has been deployed increasingly in the European railway operations. On a
higher level, European Commission has published a Technical Specification for Inte-
roperability (TSI) 2006/861/EC regarding rolling stock freight wagon, which specifies
RFID air standard ISO18000-6 type A to be used if RFID is applied [43]. It has been
argued, however, by the European Rail Infrastructure Managers (EIM) in their posi-
tion paper on RFID standards published in 2009 that the air standard specified by the
2006/861/EC is outdated and standard ISO18000-6 C, better known as EPC Gen 2
Class 1, should be used instead [44]. This EPC compliant standard has been adopted
at least in the Finnish and Swedish railways.
In Finland, VR has recently installed Gen 2 Class 1 EPC compliant RFID tags to
all of its cargo wagons [45]. The company intends to use the tags in three different
ways: firstly, it has streamlined its own railyard operations by using hand-held RFID
readers to identify wagons and transfer data intra-organizationally; secondly, it will
start to track its wagons using the upcoming network of fixed RFID readers to be
installed by the Finnish Transport Agency; and thirdly, it has planned to install its
own fixed readers to offer its customers tracking data of transportations. Regarding
the second point, FTA has planned to start deploying the fixed RFID readers along the
Finnish railway network at the end of 2011. The purpose is to install around 120 read-
ers to places where it has rolling stock monitoring sites. FTA will use them to collect
data for preventive maintenance use, and identification and measurement data will be
distributed to rolling stock owners and maintainers as well.
In Sweden, the RFID deployment is not necessarily as wide as in Finland but has
been planned and designed extensively. Between 2005 and 2008, SJ (Swedish state-
owned railway operator) tested RFID systems based on active tags but a satisfactory
standard was not found. From then on, they have been piloting RFID with passive
8 Assessment of EPCIS Standard for Interoperable Tracking in the Supply Chain 129
EPC tags and GS1s Global Returnable Asset Identifiers (GRAI). The results have
been satisfying, and the plans include using the tags with the upcoming fixed RFID
reader network installed by the Swedish Traffic Administration. There are already 10
readers in pilot use between Falkping and Gothenburg, and the long-term goal is to
install between 500 and 700 readers along the Swedish railway network.
As can be concluded from the Finnish and Swedish RFID implementations, there
seems to be a fertile ground for inter-organizational data interchange from a technolo-
gical perspective. EPC compliant RFID tags are used and they have gained a strong
position as means of identification. What needs to be considered next is whether there
is a demand and willingness to share data between organizations and if so, whether
EPCIS is the means to accomplish this goal.

Inter-Organizational Data Exchange. Inter-organizational data exchange does exist
in Finnish and Swedish railway operations already today. In Finland, VR uses EDI to
exchange inter-organizational data for instance orders, invoices, transportation doc-
uments, tracking data with its customers [46]. It has also a web application for
smaller customers without permanent contracts. There are also prospects for further,
richer data interchange if VR begins to install its own fixed readers and offer cargo
tracking to its customers, as has been planned. Furthermore, the tracking data from
FTAs fixed readers, including technical measurement information, is planned to be
used once the system is up and running. It is not clear yet if the data provided to cus-
tomers should be more detailed than just the wagon identification and direction. It
could include cargo contents, for example, but then the data content would be more
refined than what is directly available in FTAs system. The data standard to be used
in the exchange of the tracking information of the fixed readers will be ultimately
decided by the FTA but it will take into account the needs of other stakeholders, in-
cluding VR. EPCIS has been brought up in the discussions every now and then but it
has not been favored in any way and no decision for the data exchange standard has
been made yet. FTA and VR have been following the EPCIS experimentation of their
Swedish counterparts, however.
In Sweden, the development has moved a bit regarding use of EPCIS. The EPC pi-
lot projects have involved use of EPCIS since the beginning and no other standard has
been tested or considered for the data interchange. The 10 fixed readers that are oper-
ating currently use EPCIS to share tracking data between the Falkping freight ter-
minal and Gothenburg harbor. EPCIS has been regarded to perform well without any
major problems.
In summary, stakeholders in both Finnish and Swedish railways exchange data in-
ter-organizationally even without EPCIS [47]. The need for standardizing has been
acknowledged as well, as is illustrated by the aforementioned position paper on RFID
by EIM suggesting that a standardized RFID message should be defined by a TSI.
However, it still remains unclear which standard will offer the best solution and be-
come dominant. The standard to be used did not seem to play a big role in the Finnish
railways as long as it proved to function properly, according to the informants.
The Swedish railways have embraced EPCIS more than the Finnish railways but the
benefits over other ways to share data, like EDI, are harder to identify. It has been
130 K. Frmling et al.
suggested by the STA that after the installation of the fixed RFID readers along the
railway network is finished, it is then the role of markets to use the data provided and
develop new services and businesses based on it [49]. But we argue that this may not
happen so easily at least with EPCIS if companies do not see a clear benefit in
EPCIS over their existing systems. Even VR, who plans to provide RFID based track-
ing systems to its customer in the future, has not decided what data standard to use. If
they already have a well working EDI infrastructure with their customers, it may be
hard to justify new implementations based on EPCIS.
Possibilities for Standardization of Inter-organizational Data Exchange. Clearly,
there is a need for a widely used standard in inter-organizational data exchange in
railway operations. In Swedish railways, for instance, 60-70% of the wagons come
from other European countries [48]. If standardized messages were transferred be-
tween organizations from other countries, there would be a big potential for exchange
of tracking information. Then again, in the Finnish railways, there are fewer stake-
holders and they are less international than in Sweden. Another possible obstacle to a
European wide standard for inter-organizational data exchange is that some countries
are already building their own tracking systems in railways according to the infor-
mants. This might lead to problems of interoperability. One solution or remedy to this
could be that an administrative body would impose the standard to be used. In rail-
ways, the European Commission could define the standard in its TSI but as for now,
there is not that kind of progress to be seen in the near future, the informants say.
3.3 Technical Trade
To spread the use of RFID/EPC technology, GS1 Finland together with two universi-
ties launched an exploratory case study project together with 16 companies 12 man-
ufacturers, 3 wholesalers and one Logistics Service Provider (LSP) acting mainly in
technical trade industry and especially in heating, plumbing and air-conditioning
(HPAC) field. The main purpose of the project was to find suitable application areas
for RFID/EPC technology in the technical trade supply chain and create a roadmap
for adopting supply chain-wide RFID/EPC technology implementation.

RFID Deployment. The project was divided into three phases: (1) Research the ap-
plicability of RFID technology for the technical trade industry, (2) Test the technolo-
gy in practice, if RFID turns out to be suitable for the industry, and (3) create a road-
map for promoting the use of RFID to the whole industry.
In the first phase of the project, the participating manufacturer and the wholesaler
companies found enough benefits in their operations to consider larger RFID tracking
implementation. During the second phase, testing, they also found ways to tackle major
industry specific problems of adopting RFID tracking, such as ensuring the reading on
metal components typical of the industry. After the test, the biggest wholesaling compa-
ny participating in the project announced that they will start to implement RFID at first
to improve their receiving in the second half of 2012. As this wholesaler has a 43 %
market share in HPAC wholesaling in Finland, the participating manufacturers listened
8 Assessment of EPCIS Standard for Interoperable Tracking in the Supply Chain 131
rather carefully to the intentions of that wholesaler. That announcement also gave guide-
lines for creating a roadmap and helped to set concrete targets and milestones for the
project participant companies for adopting RFID. This announcement also encouraged
the participant companies to discuss about practical issues of adoption such as what kind
of RFID tagging the wholesaler prefers and what would be the format of ASN messages
that the wholesaler wants.
Inter-organizational Data Exchange. As GS1 Finland coordinated and initiated the
project, their purpose was to propose technological solutions that were in accordance
with global GS1 standards. One of the managers of GS1 Finland formulated this in
the following form at a steering group meeting in the early phase of the project:
GS1 is not involved in any project, which does not support global GS1 standards!
The companies involved in the project did not express any resistance against
the use of EPCIS architecture, if it works. In the beginning of the project most of the
participant companies did not even use EDI; therefore they could apply any new
standard when developing their logistics operations to use electronic data messages,
because their existing systems needed considerable modifications in any case.
Possibilities for Standardization of Inter-organizational Data Exchange. When
researching the roadmap for industry-wide RFID/EPC implementation, the research
team in the project had a problem: Especially GS1 Finland would have liked to rec-
ommend implementing EPCIS standards, but they realized that there is no software or
integrator company that could provide a ready EPCIS product. The system integra-
tor company, who is market leader in Finland and in Nordic countries in SCM RFID
implementations, does not at least yet have the ability to offer solutions based on
EPCIS. One major Finnish software company has developed an EPCIS product in
an EU project some time ago, but according to GS1 representatives, it still requires
further development to be suitable for implementation in the technical trade industry.
Therefore, in the end of the research project GS1 ended up to recommend to use other
interface standards widely used in the industry at the moment, but keep EPCIS in
mind in a way that the system could be developed in the future to apply EPCIS.
The settings for the project seem to be ideal for applying EPCIS, because there was
a strong involvement of GS1 in the project, and the project participant companies did
not use any other dominant standard, which they would prefer to rely. However,
project participant companies had to accept that EPCIS is not mature enough for im-
plementation in this domain. Therefore GS1 did not want to risk the enthusiasm to-
wards RFID tracking by recommending a standard that is still under development.
4 Conclusions
We started developing systems and interfaces for inter-organizational exchange of
tracking data in the supply chain because of the clear need for such solutions by
companies doing international investment projects in 2001. We have therefore been
closely following the evolution and the take-up of such solutions in practice. Despite
technical challenges of EPCIS compared to alternative approaches [5], we would have
132 K. Frmling et al.
expected EPCIS and related standards to become universally used in SCM and similar
applications. However, the conclusion of the case studies performed is that EPCIS is
still mainly being used on an experimental scale in domain-specific applications.
From the case studies, we have attempted to identify the main reasons that would
explain why EPCIS is not used more universally. We have identified at least the fol-
lowing potential explanations for the phenomenon:
Companies may not be ready or willing to share information with other organiza-
tions. Most RFID success stories are mainly intra-organizational applications,
where the use of EPCIS does not provide enough value for the cost that it causes.
EPCIS is too tightly coupled with RFID and SCM-focused, which makes it hard to
extend for use in other parts of business, thereby limiting the potential profits that
could be achieved.
EPCIS is too complex or costly to implement compared to the benefits. Even
though validated EPCIS implementations are provided by many software provid-
ers, the required implementation and systems integration may still be perceived as
being too high.
Further research would be required in order to understand the impact of these causes
for the non-use of EPCIS or other information exchange standards. Such an increased
understanding might help us to create standards that would be more widely accepted.
Since the popularity of RFID started increasing over ten years ago, it has been said
that we only need to find the right silver bullet that will make RFID and inter-
organizational information exchange explode at some moment. However, we are
still waiting for that silver bullet to be found.
References
1. Holmstrm, J., Frmling, K.: Design Patterns for loosely coupled trace, track, configura-
tion, and check operations in multi-company environments. In: Proceedings of EUROMA
Conference 2005, pp. 329337 (2005)
2. Kim, T.Y., Lee, S., Kim, K., Kim, C.H.: A modelling framework for agile and interopera-
ble virtual enterprises. Comput. Ind. 57, 204217 (2006)
3. Estrem, W.A.: An evaluation framework for deploying Web Services in the next genera-
tion manufacturing enterprise. Robot. Comput. -Integr. Manuf. 19, 509519 (2003)
4. Krkkinen, M., Ala-Risku, T., Frmling, K.: The product centric approach: a solution to
supply network information management problems? Comput. Ind. 52, 147159 (2003)
5. Frmling, K., Harrison, M., Brusey, J., Petrow, J.: Requirements on Unique Identifiers for
Managing Product Lifecycle Information - Comparison of Alternative Approaches. Int. J.
of Computer Integrated Manufacturing 20, 715726 (2007)
6. EPC Information Services (EPCIS) Specification, EPCGlobal Ratified Standard (2007),
http://www.gs1.org/gsmp/kc/epcglobal/epcis/
epcis_1_0_1-standard-20070921.pdf (accessed August 23, 2012)
7. Emmelhainz, M.A.: Electronic Data Interchange: Total Management Guide. Van Nostrand
Reinhold, New York (1990)
8. Damsgaard, J., Truex, D.: Binary trading relations and the limits of EDI standards: The
Procrustean bed of standards. Eur. J. Inform. Syst. 9, 173188 (2000)
8 Assessment of EPCIS Standard for Interoperable Tracking in the Supply Chain 133
9. Nurmilaakso, J.M.: Adoption of e-business functions and migration from EDI-based to
XML-based e-business frameworks in supply chain integration. Int. J. Prod. Econ. 113,
721733 (2008)
10. Goldfarb, C.F., Prescod, P.: Charles, F. Goldfarbs XML Handbook, 5th edn. Prentice Hall,
New Jersey (2004)
11. Chong, A.Y.L., Keng-Boon, O.: Adoption of inter organizational system standards in
supply chains: An empirical analysis of RosettaNet standards. Ind. Manage. Data
Syst. 108, 529547 (2008)
12. W3C: Extensible Markup Language (XML) 1.0, 5th edn. (2008),
http://www.w3.org/TR/xml/ (accessed August 23, 2012)
13. Kauremaa, J., Nurmilaakso, J.M., Tanskanen, K.: E-business enabled operational linkages:
The role of RosettaNet in integrating the telecommunications supply chain. Int. J. Prod.
Econ. 127, 343357 (2010)
14. Bussler, C.: B2B Integration: Concepts and Architecture. Springer, Berlin (2003)
15. Boh, W.F., Soh, C., Yeo, S.: Standards development and diffusion: A Case Study of Roset-
taNet. Commun. ACM 50, 5762 (2007)
16. Tsai, T., Huang, W., Chang, C., Wu, F., Chou, S.T.: eXFlow: a Web Services-compliant
system for supporting B2B process integration. Inform. Syst. and E-Business Manage-
ment 5, 4764 (2007)
17. Huvio, E., Grnvall, J., Frmling, K.: Tracking and tracing parcels using a distributed
computing approach. In: Proceedings of the 14th Annual Conference for Nordic Research-
ers in Logistics (NOFOMA 2002), pp. 2943 (2002)
18. Krkkinen, M., Ala-Risku, T., Frmling, K.: Efficient Tracking for Short-Term Multi-
Company Networks. Int. J. Phys. Distrib. Logist. Manag. 34, 545564 (2004)
19. Krkkinen, M., Ala-Risku, T., Frmling, K., Collin, J., Holmstrm, J.: Implementing in-
ventory transparency to temporary storage locations: a solution design experiment in
project business. Int. J. of Managing Projects in Business 3, 292306 (2010)
20. Holmstrm, J., Frmling, K., Ala-Risku, T.: The uses of tracking in operations manage-
ment: Synthesis of a research program. Int. J. Prod. Econ. 126, 267275 (2010)
21. Hild, N.: Success of EPCIS Pilot in Swedish Fishery: Swedish pilot applies EPCIS stan-
dard to food traceability. eTrace (June 22, 2010)
22. Kwok, S.K., Ting, S.L., Tsang, A.H.C., Cheung, C.F.: A counterfeit network analyzer
based on RFID and EPC. Ind. Manage. Data Syst. 110, 10181037 (2010)
23. Swedberg, C.: Norsk Lastbaerer Pool Inserts RFID Into the Norwegian Food Chain. RFID
Journal (January 19, 2011)
24. European Union, BRIDGE Building Radio Frequency Identification Solutions for
the Global Environment (2009), http://www.bridge-project.eu/data/File/
BRIDGE_Final_report.pdf (accessed October 1, 2011)
25. Fowler, P., Loughead, K.: Life Sciences Strategies: A Practical Approach to Healthcare
Supply Chain Collaboration, Axway White Paper (2011)
26. Bacheldor, B.: IBM Adds E-Pedigree Features to WebSphere RFIDIC. RFID Journal (Au-
gust 15, 2007)
27. Teichert, E.: Axway to Bolster Patient Safety and Compliance for AstraZeneca, Fierce
Pharm (2010), http://www.fiercepharma.com/press_releases/axway-
bolster-patient-safety-and-compliance-astrazeneca
(accessed October 11, 2011)
28. GS1 Healthcare: GS1 Healthcare Reference Book 2011/2012, GS1 Global Office, Brussels
(2011)
134 K. Frmling et al.
29. HDMA Factbook: Healthcare Distribution Management Association, Centre for Health-
care Supply Chain Research (2010)
30. Sachdeva, S., Pati, D.P.: Leveraging RFID and 2D Barcodes: ePedigree and Beyond. TCS
White Paper (2009)
31. Vaczek, D.: Anteing Up for RFID. Pharmaceutical & Medical Packaging News (July 28,
2008)
32. Swedberg, C.: Researchers Find Biologics Not Harmed by RFID RF Signals. RFID Jour-
nal (March 25, 2011)
33. Healthcare Packaging: GSMS gains an e-Pedigree edge. Healthcare Packaging (April
2010)
34. OConnor, M.C.: Cephalon Announces Item-Level Pilot. RFID Journal (August 9, 2005)
35. OConnor, M.C.: McKesson Starts RFID Pilot for Viagra. RFID Journal (February 17,
2005)
36. Rodgers, D.: Why GS1 EPCIS Alone Wont Work For California Pedigree, Part 1,
RxTrace (2011), http://www.rxtrace.com/2011/10/why-gs1-epcis-
alone-wont-work-for-california-pedigree-part-1.html
(accessed August 23, 2012).
37. Medical News Today: IBM and Golden State Medical Supply Create Drug Safety Model
to Help Combat Counterfeiting. Medical News Today (2009)
38. Basta, N.: Information Technology Serialization Efforts Energize Track-and-Trace Tech-
nology, Pharmaceutical Commerce (2010),
http://www.pharmaceuticalcommerce.com (accessed August 23, 2012)
39. GS1 GSMP, NCeP - Technical Analysis Sub-Group (2011),
http://www.gs1.org/docs/healthcare/Healthcare_Traceability_
Pedigree_Background.pdf (accessed August 23, 2012)
40. Fabian, B., Gnther, O.: Security challenges of the EPCglobal network. Communications
ACM 52, 121125 (2009)
41. Landt, J.: The history of RFID. IEEE Potentials 24, 811 (2005)
42. Weis, S.A.: RFID (Radio Frequency Identification): Principles and Applications (2007),
http://www.eecs.harvard.edu/cs199r/readings/rfid-article.pdf
(accessed August 23, 2012)
43. European Union: Commission decision concerning the technical specification of interope-
rability relating to the subsystem rolling stock - freight wagons of the trans-European
conventional rail system (2006), http://eur-lex.europa.eu/LexUriServ/
LexUriServ.do?uri=OJ:L:2006:344:0001:0467:En:PDF
(accessed September 23, 2012)
44. EIM, EIM position paper on RFID standards (2009),
http://www.eimrail.org/pdf/techpapers/EIM%20Position%
20Paper%20on%20the%20use%20of%20RFID.pdf (accessed August 23, 2012)
45. Sipil, A.: VR Transpoint Builds New, Efficient and Multi-Functional Logistics System
(2011), http://www.vilant.com/company/blog/2011/05/23/vr-
transpoint-builds-new-efficient-and-multi-functional-
logistics-system/ (accessed August 23, 2012)
46. Linked: VR Transpoint kehitt uusia palveluja. Linked (The magazine of VR Transpoint),
p. 11 (January 2011)
47. MacLeod, M.: Roll out the superhighway. Supply Management 4, 2627 (1999)
48. Ivansson, G., Andersson, L.: RFID in Rail, Swedish Transport Administration (2011),
http://www.gs1.se/global/EPC/RFID_in_Rail/Swedish_Transport_
Administration.pdf (accessed August 23, 2012)

You might also like