A Supply Chain Management System to Prevent Counterfeiting and
Trace Different Transactions Instead of using PUF Device
Yusuke Abe
1
, Kosei Arisaka
1
, Kitahiro Kaneda
2
and Keiichi Iwamura
1
1
Tokyo University of Science, 6-3-1 Nijuku, Katsushika-ku, Tokyo 125-8585, Japan
2
NAGASE & CO., LTD., 5-1 Nihonbashi-Kobunacho, Chuo-ku, Tokyo 103-8355, Japan
Keywords: Blockchain, Supply Chain, Raw Material, Physical Unclonable Function (PUF), Counterfeiting.
Abstract: The distribution of counterfeit products in supply chains has been increasing in recent years. Physical
unclonable function (PUF), which takes advantage of the difficulty of duplication inherent in devices, is
attracting attention as a way to overcome this problem. However, PUF can only be applied to a few objects,
notably semiconductor chips, and is, therefore, unable to cover the wide variety of products in a supply chain.
Moreover, it is necessary to use noise reduction technology, such as a fuzzy extractor, to remove noise from
the output through PUF. There is a concern that costs may increase to implement such technology. Therefore,
this paper proposes a system that can perform the same function as PUF on objects for which PUF has not yet
been established, without using noise reduction technology. An arbitrary feature of an object is measured, and
if the feature satisfies a certain criterion, the object can be safely delivered. In addition, the proposed method
is able to distinguish between individual transactions between one company and another. This prevents
unauthorized resale and diversion by controlling even the location of the products once they are dispatched
from the supplier.
1 INTRODUCTION
Logistics has evolved over the years with the
dramatic advances in information technology (IT),
and has become an inseparable part of modern life.
Here, the distribution process from the procurement
of raw material to the delivery of products to
consumers is called the supply chain. The supply
chain consists of suppliers, logistics providers,
wholesalers, retailers, and end users. To improve the
added value of products and services for customers, a
management system that optimizes the integrated
management of objects, money, and information has
been attracting attention in recent years. This is
known as supply chain management (SCM).
Companies involved in the supply chain are aware
of SCM and focus on how to provide products
efficiently. Therefore, damage caused by counterfeits
around the world still cannot be stopped. A report
(OECD and EUIPO, 2016, 2019) by the Organisation
for Economic Cooperation and Development
(OECD) and the European Union’s (EU) Intellectual
Property Office explains that the global trade value of
counterfeit and pirated goods reached $461 billion in
2013 and $509 billion in 2016. For example, have you
ever wondered whether the product you purchased
through e-commerce is authentic and has been
delivered to you through the right channels? To
eliminate such concerns, products are currently
managed through the physical attachment of radio-
frequency identification (RFID) tags or barcodes (QR
codes). For those with malicious intent, however, it is
easy to physically remove such tags or codes from the
products. This makes it possible to attach the original
RFID or barcode to a fake product and sell the
counterfeit. When such an attack occurs, it is usually
very difficult for end users to determine the
authenticity of the product, necessitating counter-
measures. In addition, each company manages its own
transaction history. Even if the end user reads the
information from the tag, the user will only be able to
view the information that has been made accessible to
the public at the discretion of the individual company.
Given this background, technologies such as
blockchain and PUF are expected to be used. As for
the information the user browses, it is expected that
blockchain, which is a distributed ledger that is
difficult to tamper with, will be introduced into SCM.
This technology can be used to track products with
reliability. For intentional tag replacement, an
individual identification system that uses a
technology called physical unclonable function
Abe, Y., Arisaka, K., Kaneda, K. and Iwamura, K.
A Supply Chain Management System to Prevent Counterfeiting and Trace Different Transactions Instead of using PUF Device.
DOI: 10.5220/0010571301010108
In Proceedings of the 18th International Conference on e-Business (ICE-B 2021), pages 101-108
ISBN: 978-989-758-527-2
Copyright
c
2021 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
101
(PUF) has been proposed, instead of information that
is externally attached to products, such as RFID or
barcodes. PUF is a function that outputs different
eigenvalues for each object, utilizing unique physical
properties of the product that are difficult to replicate.
For example, when a semiconductor chip with the
same circuit receives the same input, the output is the
same for all chips, but the response time is slightly
different. This technology identifies individual chips
with the same circuit using the difference in response
time as a unique property of the device, which makes
it difficult to duplicate. A buyer of the device can
determine the authenticity by utilizing this feature.
However, PUF has a drawback in that it cannot be
applied to all products. Due to PUF’s features, it is
limited to a very small number of products, such as
semiconductor chips. The technology is not versatile
and cannot cover a wide variety of commerce. In this
paper, we propose a system that can perform the same
functions as PUF for a range of materials (powders,
liquids, individuals, precious metals, etc.) as
examples of a commodity supply chain for which
PUF cannot be used. Mere substances are often
identified by their composition and size, and few
individual identification technologies, like PUF, have
been studied that clearly confirm the match. It is,
however, possible to judge if a product is legitimate
based on whether it meets a criterion. This judgment
is called normal judgment. In normal judgment, by
measuring the physical properties of a product, a
substance that has the specified components and size
is judged to be genuine, and a substance that does not
meet the specified components and size is judged to
be fake. Accordingly, when an end user purchases
and receives a substance, it is preferable to have a
technology that not only determines whether the
product is legitimate by satisfying the specified
features, but also utilizes unpredictable values like
PUF. Hence, the main goal is to realize an SCM
system that can be used for various products, with or
without the application of PUF.
The remainder of this paper is organized as
follows. Chapter 2 explains the related work against
counterfeits, Chapter 3 describes the proposed
methods, Chapter 4 is devoted to the evaluation, and
Chapter 5 provides a summary.
2 RELATED WORK
2.1 Blockchain in Supply Chain
Current supply chains have difficulty in tracking
product history (traceability). Even if consumers view
the product history, they cannot determine if the data
such as “who”, “when”, “where”, “what”, and “how”
are correct. Therefore, a platform for sharing accurate
information is needed, and there is a lot of research
being done on the use of blockchain, a secure and
highly available distributed ledger. Dietrich et al.
(2021) and Pournader et al. (2020) survey and review
many blockchain projects in the supply chain.
Hackius et al. (2017) sought input from logistics
experts and found that most experts are positive. Tijan
et al. (2019) argue that blockchain can minimize
major issues in logistics such as order delays, errors,
and multiple data entry.
2.2 Detection of Counterfeiting
This section lists the issues regarding RFID-based
external tag technology, PUF-based technology, and
identification by substance features. The study of the
issues is a stepping stone in proposing effective
methods.
Sun et al. (2019) claim that RFID and the
information associated with it cannot be tampered
with, and that different users are provided with
different query permissions to maintain their
authenticity. Toyoda et al. (2017) argue that end users
can reject counterfeits by having each entity transfer
products and their ownership while determining the
authenticity of the RFID tags. No matter how much
the authenticity of the tag is guaranteed, as shown by
Sun et al. (2019) and Toyoda et al. (2017), an end user
has no way of checking the authenticity of the content.
The physical space where products exist and the
cyberspace where authenticity is guaranteed are not
well connected. The use of RFID tag anti-
counterfeiting technology with PUF, as described by
Devadas et al. (2008), is therefore not a sufficient
solution.
Previous studies (Hori et al., 2015; Aniello et al.,
2019; Negka et al., 2019) focus mainly on individual
identification systems that utilize PUF.
Three points are worth noting. First, as mentioned
previously, PUF can only be applied to specific
products. Target products need to be expanded to
meet a wide range of modern needs. Second, noise
reduction technology, called a fuzzy extractor (Dodis
et al., 2004), is required. When using PUF, ideally, a
certain semiconductor chip should always produce a
same output, but in reality, it is difficult because of its
vulnerability to noise. In addition, PUF uses a minute
variation in each semiconductor chip. There are
therefore problems such as not being able to get an
appropriate output or getting a similar output from
different devices. Fuzzy extractor is a means to solve
ICE-B 2021 - 18th International Conference on e-Business
102
this problem. However, installing a fuzzy extractor
makes the process more complicated and increases
the processing time and circuit size. This is also a
factor that increases implementation costs. Third,
systems that use PUF only determine whether the
device shipped and the device delivered to the end
user is an exact match. Strictly speaking, this is not an
authenticity judgment. This just verifies that the end
user has received the product declared by the supplier.
However, it is generally not important for the end user
that the shipment and purchased product are identical.
In many cases, it is important that the product meets
a required criterion and is legitimate. Consider, for
example, the situation in which a diamond is
purchased from a catalog. The purchaser does not
necessarily want a diamond that is exactly the same
as the picture in the catalog. It is important that the
diamond’s carats, hardness, and size meet the criteria.
The system proposed by Koike (2010) verifies
legitimacy based on the features extracted from the
target object. The normal judgment is more
predictable than PUF because the judgment value of
a legitimate device is fixed. We take the example of a
diamond once again. An attacker realizes that if a
product has a specified range of carat, hardness, and
size, it is considered real. In this case, even if the
identification target is out of range, the attacker can
pass off a counterfeit as genuine by creating a device
that outputs a judgment value satisfying the criteria.
By the way, it is difficult to predict an output of PUF,
even if the value of one device is stored, it cannot be
diverted to other devices. The normal judgment does
not have this feature of PUF. Therefore, there are few
proposals for judgment systems based on features of
objects.
In summary, technology is required to output
different values for each individual (or each
transaction between companies), similar to PUF,
without fixing the output value of the normal
judgment. This prevents fraud by logistics providers.
3 PROPOSED METHOD
3.1 Overview
In this study, we propose a technology to determine
authenticity by normal judgment. The goal of the
proposed system is to achieve the same functionality
as PUF for the SCM of substances that have been
difficult to identify in the past. The system consists of
a registration device, verification device, and
identification device, as shown in Figure 1. The
registration device in SCM is assumed to be used by
suppliers who generate and ship products. A
verification device is used by logistics providers who
transport products. The identification device is used
by end users who receive the products. The focus is
on commercial transactions; therefore, suppliers and
end users have no incentive to commit fraud, and
fraud or errors by outsiders or logistics providers can
be controlled. Furthermore, each function is realized
by blockchain, which improves the common
information management and traceability among
other companies. Blockchain improves the efficiency
of the entire SCM and clarifies where responsibility
lies. The following sections describe in detail the
algorithms of the registration device in Section 3.2,
the verification device in Section 3.3, and the
identification device in Section 3.4.
3.2 Registration Device
The registration device is handled by a supplier. It
serves to generate a product-specific key and registers
the key in the blockchain by inputting the feature
values of the product and information, which are
different for each product. This device consists of
four elements: measurement, judgment, generation,
and registration.
The measurement section extracts the feature
value from the target as input signal P1 and outputs
the measurement value.
The judgment section determines whether the
product is legitimate based on the value obtained
from the measurement section. For this purpose, the
feature value to be acquired for each product and its
legitimate range are set in advance. A judgment value
is output after determining that the product is
legitimate if within the range, and that it is
illegitimate if it is outside the range. The value is then
expressed as a relatively large bit string of 128 bits,
for example, as binary values of legitimate or not
legitimate. In this way, noise reduction technology is
no longer necessary and the judgment result is
effective in suppressing forgery.
The generation section concatenates the judgment
value and “transaction information” U1 unique to
each product, which is independent of the product
features. Concatenation is performed by an exclusive
disjunction. A hash of the concatenated values is then
generated as the identification key, Key1. Transaction
information identifies products and utilizes, for
example, the manufacturer, serial number,
temperature, number of verifications, and random
numbers. There are no obstacles even if outsiders
possess the same type of product. This is because the
judgment value is not output to the outside of the
A Supply Chain Management System to Prevent Counterfeiting and Trace Different Transactions Instead of using PUF Device
103
Figure 1: Overview of the system.
device and is kept secret by transaction information,
including random numbers. Therefore, the
identification key becomes random and unpredictable
for each transaction.
The registration section links the information to
be verified with the information to identify it. First,
“verification information” is generated by hashing to
use the Key1 output from the generation section.
Then, “identification information” is utilized to
identify the information corresponding to the product
for which judgment is conducted from among a large
amount of verification information. The identification
information applies to the manufacturer and serial
number pair that is part of the transaction information.
Finally, the generated verification and identification
information are mapped and registered in the
blockchain. Owing to the features of the blockchain,
all values are made public. If Key1 is made public as
is and a malicious logistics provider forges a
verification device to output the value, an end user
may be fooled. To prevent fraud, Key1 is hashed and
the identification key is unpredictable.
The following shows the specific flow of the
registration device.
(1) A supplier inputs an input signal P1, which
depends on the product’s features and transaction
information U1.
(2) The judgment section determines whether the
measurement results are valid.
(3) The generation section generates Key1 from the
judgment value and transaction information.
(4) The registration section generates verification
information from Key1 and registers it in the
blockchain with identification information.
3.3 Verification Device
The verification device is a device handled by the
deliverer who transports the product. In much the
same way as the registration device, a product-
specific key is generated by inputting product features
and transaction information, which are different for
each product. At this point, the end user, who is
notified by the supplier, can input the transaction
information and make a judgment using the product
to confirm its legitimacy. This device consists of four
elements: measurement, judgment, generation, and
output.
The measurement section measures a feature
value of the target as an input signal P2 and outputs a
measurement value.
The judgment section outputs a judgment value
based on the acquired measurement values. It
determines that the product is legitimate if it is within
the pre-set range and illegitimate if it is outside the
range.
The generation section concatenates the judgment
value and “transaction information” U2 unique to
each product, which is independent of the product’s
features. Concatenation is performed by an exclusive
disjunction. A hash of the concatenated values is then
generated as the identification key, Key2. When an
end user purchases a product, the information used in
the registration device is notified by the supplier as
transaction information.
The output section provides Key2 generated by
the generation section to the identification device.
In the proposed method, the verification device
consists of four parts: measurement, judgment,
generation, and output. It can be implemented freely
ICE-B 2021 - 18th International Conference on e-Business
104
according to the application, such as by installing a
generation and output section in the identification
device described below. The measurement, judgment,
and generation sections are the same as those in the
registration device. Thus, if an input signal P2 and
transaction information U2 input to the verification
device are identical to the input signal P1 and the
transaction information U1 input to the registration
device, it is obvious that Key2 without noise is always
equal to Key1. There is also no property that the key
is slightly different each time as in SCM using PUF.
Since there is no need to implement techniques such
as a fuzzy extractor, it is not necessary to consider the
increase in processing time and implementation cost.
The specific flow of the verification device is as
follows.
(1) A user provides input signal P2 and transaction
information U2 to the device.
(2) The judgment section determines whether the
measurement results are valid.
(3) The generation section generates Key2 from the
judgment value and transaction information.
(4) The output section outputs Key2 to the
identification device.
3.4 Identification Device
The identification device is a device that is handled
by an end user. It checks whether the verification
information (hash of Key1) registered in the
blockchain matches the value of hashed Key2 from
the verification device presented by the deliverer.
This process confirms the authenticity of the product.
This device consists of three elements: acquisition,
verification, and registration.
The acquisition section obtains verification
information from the blockchain based on the
identification information, and obtains Key2 from the
verification device. The identification information is
extracted using part of the transaction information
provided by the supplier to the end user in this process.
The verification section checks for consistency
between the verification information and the
information in Key2. Specifically, Key2 is hashed,
and whether the hash matches the verification
information is examined. If it matches, the section
outputs the success information to the registration
section, indicating that the verification is successful.
The registration section links the “verified
information” to the verification information when it
is successfully verified and registers it in the
blockchain. The information with the verified
information is restricted so that it cannot be verified
again when verified later. This prevents the same
transaction information from being used in a
malicious manner.
The specific flow of the identification device is as
follows.
(1) The acquisition section obtains Key2 from the
verification device. Additionally, it searches the
blockchain based on the input identification
information, and obtains verification information
if the verified information is not attached.
(2) The verification section hashes Key2 obtained in
(1). It verifies whether the hash value matches the
verification information and outputs the result.
(3) The registration section adds the verified
information to the identification information if the
result is valid. This is then registered in the
blockchain.
4 EVALUATION
To evaluate the proposed method, we discuss possible
attacks in Section 4.1. Section 4.2 presents a
comparison with conventional SCM using barcodes,
RFID, mere substances, and PUF. Furthermore, the
implementation cost is described in Section 4.3.
4.1 Attacks on the Proposed Method
4.1.1 Fraud by Logistics Providers
The simplest example of a supply chain is a supplier,
a logistics provider, and an end user. It is assumed that
there is no fraud in the commercial transactions
between a supplier and an end user because they can
simply terminate the contract if they are dissatisfied
with the other party. The main possible source of an
attack is that the logistics provider may swap the
authentic item with a counterfeit and the end user
receives a counterfeit. In general, an end user does not
have a large-scale verification device. The
verification is carried out using a verification device
owned by a logistics provider for authenticity
judgment or normal judgment. The following is an
example of a diamond transaction. If a diamond is
simply swapped with an object that is not a diamond,
such as zircon or zirconia, it can be easily detected as
unjustified. However, even if the object is a fake that
does not fit into the range, it is possible to make the
fake real by forging the device.
The proposed method counters this attack by
generating a different key for each transaction. The
term “transaction” in this context does not refer to the
A Supply Chain Management System to Prevent Counterfeiting and Trace Different Transactions Instead of using PUF Device
105
entire commercial transaction between a supplier and
an end user. Instead, it refers to transactions between
companies, such as transactions between a supplier
and a logistics provider, and transactions between a
logistics provider and an end user. The following
describes the countermeasure method in detail.
The success or failure of the normal judgment for
input signal P2 is noted, but the judgment value itself
is not output. It is also difficult to obtain the value
from the outside by analyzing the device. The
judgment value is then secreted into the device using
transaction information U2, including random
numbers, and output as the identification key Key2.
The random number is known only to the supplier and
the end user. Key2 is difficult to predict unless the
judgment value is leaked to the outside and
randomness is maintained as long as the random
number is not known. In summary, the judgment
value is not output to the outside of the device but is
kept secret in the transaction information to generate
Key2, which is difficult to predict and random. This
means that even if PUF is not applicable to the target,
the function is equivalent to that of PUF. Furthermore,
a logistics provider does not have the advantage of
storing the Key2 value; because they generate a
unique identification key for each transaction, it is
meaningless and cannot be used for other transactions.
This feature is not found in PUF. In addition, by
adding verified information to the transaction once it
is used, the system prevents unauthorized double use.
Unless a famous brand adds verified information to
its own products, it will be possible for an unknown
brand to sell its products fraudulently. Using the
identification information in the blockchain, an
unknown brand can falsely sell products that are
identical to the quality of a famous brand. The
products are indistinguishable from those of famous
brands. It is therefore necessary to add the verified
information to the verification information used to
limit double use. However, even if the information is
verified once, it can be verified again. The number of
verifications included in the transaction information
is added by one, and the verification information for
that is generated. By registering this information, a
new verification can be performed.
4.1.2 Blockchain-based Attacks
A blockchain is a public ledger. Therefore, a third party
can view the blockchain to obtain information about
transactions. This subsection describes the study of the
possibility of fraud using the proposed method.
The only information to be registered in the
blockchain is identification and verification
information. The identification information is used to
obtain verification information for the corresponding
transaction from the blockchain. It is created using
part of the transaction information. In the proposed
method, when a product is registered, a supplier
notifies all transaction information only to the end
user through a secure channel. The system works
properly only when the product meets the required
standard, and the correct transaction information is
entered. Even if an outsider obtains the identification
and verification information by browsing the
blockchain, he/she will not be able to know the
transaction information, such as random numbers. It
is not possible to generate the correct Key2. Thus,
there is no room for an outsider to show the
authenticity of the product using the proposed
method. In addition, it is very difficult to falsify the
identification and verification information published
on blockchain, and attacks using such information are
hard. This is because blockchains are virtually
impossible to tamper with in terms of computational
complexity.
4.2 Comparison with Conventional
Methods
The proposed method and conventional methods are
compared from four perspectives, as shown in Table 1.
Conventional methods include SCM with external
tags using barcodes or RFID, SCM utilizing mere
substances, and SCM using PUF.
The method using PUF is superior in that it
provides an exact match between the shipment and
the product received. This is not necessarily
important, however. The fact that the product is
verified as legitimate is generally sufficient.
Furthermore, fraud is possible if a verification
device's output indication for a shipped PUF device is
forged. Using external tags or mere substances is also
flawed in both respects and cannot dispel concerns of
end users. It is difficult to determine whether a
product is legitimate if an attacker removes a barcode
or RFID tag and attaches it to a counterfeit product,
or replaces only contents. In the case of mere
substances, an end user cannot know an exact result
as long as an attacker can produce a verification
device to tamper with an output. On the other hand,
the proposed method does not fix a judgment value
but outputs it randomly for each transaction, so it is
possible to determine whether a product is legitimate.
This method does not use PUF. However, it provides
advantages of unpredictability and randomness like
PUF. It can also be applied to products that can use
PUF.
ICE-B 2021 - 18th International Conference on e-Business
106
Table 1: Comparison with conventional methods (1=lowest; 3=highest).
Method
Exact match with
the shipment
Legitimate product Cost
Distinction per
transaction
Proposed method 2 3 1 3
External tag
(Barcode/RFID)
1 1 3 1
Mere material 2 2 1 1
PUF 3 2 1 1
External tags are the best in terms of cost, and
although they are not as secure, they are relatively
easy to install in existing systems. The other three
methods are not so simple, as they consist of
somewhat complex systems to detect counterfeit
products. However, there is a trade-off between high
security and cost (simplicity), and supply chain
members must pursue what users want.
The distinguishing feature between each
transaction is found only in the proposed method. For
example, the PUF-based method does not distinguish
between each transaction in the process of product
flow from user A to user B to user C. The proposed
method can generate individualized unique keys
using transaction information in the process of
product flow from user A to user B and from user B
to user C. Double use such as unauthorized resale or
diversion can thus be prevented, and suppliers can
understand how the products they sell are resold. This
is important for ensuring traceability and safety for
users.
4.3 Simulation
A blockchain substrate called Ethereum was used to
simulate the proposed method from the viewpoint of
ease of development and payment in virtual currency.
Ethereum generates a fee every time a smart contract
is executed. This provides incentives to miners, who
are responsible for approving transactions and
keeping the blockchain secure. The fee is managed in
units called gas. The behaviors of the registration,
verification, and identification devices were checked
in a test environment. Remix (2021), a web browser
integrated development environment (IDE) for
developers of the dedicated language Solidity, was
employed. Cost calculations were performed using
the rate on February 15, 2021. Etherscan (2021)
showed that the average gas price was 178.715 Gwei.
CoinGecko (2021) showed that the dollar rate was
1804.98 USD/ETH. The implementation cost of each
device was calculated as shown in Table 2.
“Transaction cost” used in Remix (Table 2) is
expressed as the sum of the commonly used
transaction cost and execution cost. The cost was high
due to the steep rise in the gas price and Ethereum rate.
The former involves the limitations of the current
processing power of Ethereum, that is, scalability
issues. The latter involves a complex combination of
factors, but the increase in the number of users and
the scalability problem can be cited as factors.
However, this is not the essence of the proposed
method. This is because other programs have
calculated similarly high costs. Although not
optimistic, the Ethereum Foundation is already
pushing for migration and integration into
Ethereum 2.0. This is expected to solve the
continuous rise in gas prices and make it possible to
advance to faster technology with lower costs.
Therefore, it is important to improve the system to an
advanced level, in parallel, while paying attention to
cost.
Table 2: Cost of each device.
Device Transaction cost [gas] Cost [USD]
Registration
device
84673 27.31
Verification
device
35050 11.31
Identification
device
34675 11.19
5 CONCLUSIONS
We proposed a method for determining authenticity
using normal judgments for supply chain
management. It has the feature of being able to
perform the same function as PUF for devices and
materials for which PUF has not yet been established.
The use of blockchain improves traceability within
the entire SCM and increases the difficulty of data
tampering. The weakness of the public ledger was
A Supply Chain Management System to Prevent Counterfeiting and Trace Different Transactions Instead of using PUF Device
107
overcome using an algorithm based on hash functions.
We will continue to study more secure and efficient
requirements for practical use, with the goal of
reducing the cost of implementation.
REFERENCES
OECD, and EUIPO (2016). Trade in counterfeit and pirated
goods: Mapping the economic impact, OECD
Publishing, 68, Retrieved March 5, 2021, from
https://www.oecd-ilibrary.org/governance/trade-in-
counterfeit-and-pirated-goods_9789264252653-en.
OECD, and EUIPO (2019). Trends in trade in counterfeit
and pirated goods, OECD Publishing, 11, Retrieved
March 5, 2021, from https://www.oecd-
ilibrary.org/trade/trends-in-trade-in-counterfeit-and-
pirated-goods_g2g9f533-en.
Dietrich, F., Ge, Y., Turgut, A., Louw, L., and Palm, D.
(2021). Review and analysis of blockchain projects in
supply chain management, Procedia Computer Science,
180, 724-733.
Pournader, M., Shi, Y., Seuring, S., and Koh, S. L. (2020).
Blockchain applications in supply chains, transport and
logistics: a systematic review of the literature,
International Journal of Production Research, 58 (7),
2063-2081.
Hackius, N., and Petersen, M. (2017). Blockchain in
logistics and supply chain: trick or treat?, In
Digitalization in Supply Chain Management and
Logistics: Smart and Digital Solutions for an Industry
4.0 Environment. Proceedings of the Hamburg
International Conference of Logistics (HICL), 23, 3-18.
Tijan, E., Aksentijević, S., Ivanić, K., and Jardas, M.
(2019). Blockchain technology implementation in
logistics, Sustainability, 11 (4), 1185.
Sun, W., Zhu, X., Zhou, T., Su, Y., and Mo, B. (2019).
Application of blockchain and RFID in anti-
counterfeiting traceability of liquor, 2019 IEEE 5th
International Conference on Computer and
Communications.
Toyoda, K., Mathiopoulos, P. T., Sasase, I., and Ohtsuki, T.
(2017). A novel blockchain-based product ownership
management system (POMS) for anti-counterfeits in
the post supply chain, IEEE Access, 5, 17465-17477.
Devadas, S., Suh, E., Paral, S., Sowell, R., Ziola, T., and
Khandelwal, V. (2008). Design and implementation of
PUF-based “unclonable” RFID ICs for anti-
counterfeiting and security applications, 2008 IEEE
International Conference on RFID, 58-64.
Hori, Y., Hagiwara, M., Kang, H., Kobara, K., and
Katashita, T. (2015). Device-specific information
generation device, device-specific information
generation system and device-specific information
generation method, Japanese Patent P2015-154291A.
[in Japanese].
Aniello, L., Halak, B., Chai, P., Dhall, R., Mihalea, M., and
Wilczynski, A. (2019). Towards a supply chain
management system for counterfeit mitigation using
blockchain and PUF, arXiv preprint arXiv:1908.09585.
Negka, L., Gketsios, G., Anagnostopoulos, N. A.,
Spathoulas, G., Kakarountas, A., and Katzenbeisser, S.
(2019). Employing blockchain and physical unclonable
functions for counterfeit IoT devices detection,
Proceedings of the International Conference on Omni-
Layer Intelligent Systems, 172-178.
Dodis, Y., Reyzin, L., and Smith, A. (2004). Fuzzy
extractors: How to generate strong keys from
biometrics and other noisy data, In International
conference on the theory and applications of
cryptographic techniques, 523-540.
Koike, M. (2010). Authenticity verification system,
information generation device, authenticity verification
device, information generation program, and
authenticity verification program, Japanese Patent
P2010-81039A. [in Japanese].
Remix (2021). Ethereum IDE, Retrieved March 5, 2021,
from https://remix.ethereum.org/.
Etherscan (2021). Ethereum average gas price chart,
Retrieved March 5, 2021, from
https://etherscan.io/chart/gasprice.
CoinGecko (2021). Ethereum price, ETH price index, chart,
and info, Retrieved March 5, 2021, from
https://www.coingecko.com/en/coins/ethereum.
ICE-B 2021 - 18th International Conference on e-Business
108