Available via license: CC BY 4.0
Content may be subject to copyright.
Received January 21, 2021, accepted February 8, 2021, date of publication March 1, 2021, date of current version March 9, 2021.
Digital Object Identifier 10.1109/ACCESS.2021.3062845
A Trusted Blockchain-Based Traceability System
for Fruit and Vegetable Agricultural Products
XINTING YANG1,2,3, MENGQI LI1,2,3, HUAJING YU 1,2,3, MINGTING WANG1,2,3, DAMING XU1,2,
AND CHUANHENG SUN 1,2
1National Engineering Research Center for Information Technology in Agriculture, Beijing 100097, China
2National Engineering Laboratory for Agri-product Quality Traceability, Beijing 100097, China
3College of Information, Shanghai Ocean University, Shanghai 201306, China
Corresponding author: Chuanheng Sun (sunch@nercita.org.cn)
This work was supported in part by the National Natural Science Foundation of China under Grant 31871525, in part by the Beijing
Municipal Natural Science Foundation under Grant 4182023, and in part by the Academician and Expert Workstation of Yunnan Province
Project of Construction of Information-Based Cloud Platform for the Quality Traceability of Green and High-quality Dairy Products.
ABSTRACT Traditional traceability system has problems of centralized management, opaque information,
untrustworthy data, and easy generation of information islands. To solve the above problems, this paper
designs a traceability system based on blockchain technology for storage and query of product information
in supply chain of agricultural products. Leveraging the characteristics of decentralization, tamper-proof and
traceability of blockchain technology, the transparency and credibility of traceability information increased.
A dual storage structure of ‘‘database +blockchain’’ on-chain and off-chain traceability information
is constructed to reduce load pressure of the chain and realize efficient information query. Blockchain
technology combined with cryptography is proposed to realize the safe sharing of private information in the
blockchain network. In addition, we design a reputation-based smart contract to incentivize network nodes to
upload traceability data. Furthermore, we provide performance analysis and practical application, the results
show that our system improves the query efficiency and the security of private information, guarantees the
authenticity and reliability of data in supply chain management, and meets actual application requirements.
INDEX TERMS Blockchain, traceability, on-chain and off-chain, agricultural products.
I. INTRODUCTION
Fruit and vegetable agricultural products have excellent
production advantages in China, which is a large agricul-
tural country with superior climate conditions and abun-
dant species resources. According to data from the National
Bureau of Statistics of China [1], the total output of fruit
and vegetable agricultural products in 2019 was 995.03 mil-
lion tons, accounting for 54.48% of all agricultural prod-
ucts (1826.55 million tons). Fruit and vegetable agricultural
products have the characteristics of green, healthy and high
nutritional value [2], which are deeply loved by people. How-
ever, the short storage time and the low storage temperature
of storage requirements for fruit and vegetable agricultural
products, leading to food safety incidents are extremely prone
to occur [3].
In recent years, domestic and international safety incidents
of fruit and vegetable agricultural products have occurred
The associate editor coordinating the review of this manuscript and
approving it for publication was Adnan Abid.
frequently. Such as ‘‘poisonous ginger’’ incident in China [4],
Hami melon contamination by listeria in United States [5],
and the outbreak of E. coli in Germany [6], which have greatly
harmed the health of the majority of people. As a result,
the state attaches great importance to the traceability of food
supply chain, and countries strengthen management of trace-
ability by issuing relevant laws and regulations. The General
Food Law promulgated by the European Union in 2002 [7]
stipulates that a comprehensive traceability system must be
established in the food industry in order to recall targets in
a timely and accurate manner and transmit information to
consumers. The Food Safety Law implemented by China in
2009 [8], which provides that food producers and operators
should establish a food safety traceability system to ensure
food traceability. ‘‘Traceable’’ has become a challenge for all
food and food-related companies and the traceability system
has become an effective means of quality management in the
agricultural product supply chain [9]–[11].
The traceability of fruit and vegetable agricultural prod-
ucts involves many subjects. In accordance with the business
36282 This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/ VOLUME 9, 2021
X. Yang et al.: Trusted Blockchain-Based Traceability System for Fruit and Vegetable Agricultural Products
relationship, it can be divided into internal and external
entities of the supply chain [12], [13]. The internal entities
include production enterprises, processing enterprises, cold
chain logistics enterprises, sales enterprises, etc., and the
external entities embrace consumers and regulatory agencies,
etc. [14]. The entire supply chain has the characteristics
of many production points and sales points, long pro-
duction chains, and wide production areas, which makes
supervision and tracing of food safety particularly difficult
in practice [15]. In practical applications, data in traditional
traceability systems is centralized, and authoritative agen-
cies manage the central database of the traceability sys-
tem [16], [17]. Since the traceability data of each supply chain
node are managed by enterprise, the data are easy to tamper
with. Therefore, the reliability of information transmission
among different roles in the agricultural supply chain needs
to be increased.
Blockchain is a kind of distributed database which is
decentralized, tamper-proof, traceability, and maintained by
multiple parties [18]. It uses a cryptographic algorithm to
form a chain structure composed of data blocks in chrono-
logical order. Any party must receive the consent of all
other parties in advance according to agreed rules in order
to realize information sharing and information supervision
among different parties [19]. Moreover, blockchain integrates
many technologies, such as Peer to Peer (P2P) networks,
cryptographic technologies, smart contract, consensus mech-
anisms, timestamps, blockchain structure, etc. [20], [21].
Thus, it can achieve the self-verification and management of
data without relying on a third party. The use of blockchain
technology for the tracing of agricultural products can solve
the problems with the current traditional traceability sys-
tem. Blockchain are mainly divided into three categories:
Public chains, Consortium chains and Private chains [22].
Consortium chains refer to the blockchain which several
organizations participate in and manage together. In terms of
privacy, Consortium chains is intermediate between Public
chains and Private chains, with its data only being acces-
sible by members of the alliance. Additionally, Consortium
chains transaction efficiency is higher than Public chains.
In a traceability system, the main responsibility bodies of
the supply chain of agricultural products are related to the
cooperative relationship between participants in supply chain.
However, these responsibility bodies cannot be fully trusted.
In reality, the main responsibility bodies of the supply chain
are originally related to each other by horizontal interaction
or a vertical transaction relationship. Therefore, this paper
chooses the Consortium chain as basic network.
The main purpose of the current paper is to explain how
to apply blockchain technology to the traceability of agricul-
tural products. Using the features of distributed storage, hash
encryption, and programmable smart contract of blockchain
technology, we have designed and implemented a traceability
system for fruits and vegetables agricultural products based
on a trusted blockchain. We will describe the design process
of the system in detail, and clarify the key breakthrough
technologies of this system, including the on-chain and off-
chain storage structure and the combination of cryptogra-
phy to achieve privacy data protection. We will build the
blockchain environment based on Hyperledger Fabric for per-
formance testing and practical application of the agricultural
product traceability system to prove the practicability of the
system.
The main contributions of this paper include:
•We elaborated on the main shortcomings of current agri-
cultural product traceability and proposed solutions.
•We apply blockchain technology to the traceability of
agricultural products, and propose solutions to the prob-
lems of heavy load, slow query speed and privacy data
protection on the existing blockchain technology. The
detailed design of the on-chain and off-chain storage
structure and privacy data protection is a key part of the
research.
•We build blockchain environment based on Hyperledger
Fabric, and use the C# language to develop and imple-
ment traceability system to realize the process of storing
and querying agricultural product traceability informa-
tion. And through the system function test and system
actual application case evaluation system.
The remaining paper is organized as follows: Section II
provides a literature review on agricultural product traceabil-
ity; Section III describes the detailed design of the system;
Section IV describes the implementation of the system in
detail, analyzes the performance of the system, and gives the
application results of the system and the comparison of the
system and traditional traceability system. The conclusions
and suggestions for future work are presented in Section V.
II. RELATED WORK
Traceability is defined as the ability to access any or all
information relating to that which is under consideration,
throughout its entire life cycle, by means of recorded identifi-
cations [22]. For agricultural products, traceability means that
when quality problems occur, the raw materials or process-
ing links that have problems can be quickly and effectively
checked, product recalls are carried out when necessary, and
targeted penalties are implemented to improve the quality and
safety of agricultural products.
Traceability has become a challenge for all food and food-
related companies, and traceability system has become an
effective means of quality management in the agricultural
supply chain. In recent years, many scholars have carried
out exploratory research in the field of agricultural products
traceability. Qian et al. [24] combined the 2D barcode and
RFID technology to design and implement a Wheat Flour
Milling Traceability System (WFMTS) by pasting the QR
code label on the wheat package to link to its processing
information and pasting the RFID label to the storage box to
record the logistics information. The center database is used
to manage the information from raw materials to finished
products in the flour mill, and realize the whole process
VOLUME 9, 2021 36283
X. Yang et al.: Trusted Blockchain-Based Traceability System for Fruit and Vegetable Agricultural Products
monitoring from source, circulation to sales. However, the
information in the traditional traceability system is managed
by the enterprises in each link, and the transparency of the
information is low, and the information is easy to be tampered
with.
Blockchain as an emerging technology that has proper-
ties of decentralization, tamper-proof and traceability pro-
vides the possibility to solve the problems existing in the
current traditional agricultural product traceability system.
More and more scholars are beginning to pay attention to
the application of blockchain technology in the traceability
of agricultural products. In the application of combining
blockchain technology with Internet of Things technology,
Bumblauskas et al. [25] combined Internet of Things technol-
ogy and blockchain technology to track products from farm
to dining table in real time. Taking the egg supply chain of a
Midwestern company in the United States as an example, and
the implementation of blockchain technology in the supply
chain system from farm to consumer was summarized in
detail. Feng [26] built a real-time traceable food supply chain
traceability system based on HACCP, Blockchain and IoT
technologies, and introduced a new concept BigchainDB to
solve the problem of blockchain scalability. Liao and Xu [27]
established a blockchain traceability system based on intelli-
gent agriculture and wireless sensor networks for tea quality
and safety management, combined with food risk assessment
and safety traceability technology based on hazard factors
to efficiently control products quality and safety. From the
multi-chain structure level of blockchain. Zhao et al. [13]
designed a big data version fresh food traceability plat-
form based on the account blockchain (ABC) and trans-
action blockchain (TBC) dual-chain architecture, analyzed
user needs from the perspective of information ecology, and
proposed a risk compensation plan that traces participating
entities. Liu et al. [28] designed an anti-counterfeiting trace-
ability system (TSPPB) that uses dual blockchain and IPFS
systems. TSPPB utilized two sets of blockchains, includ-
ing public chain and private chain, and the TCDBB tech-
nology is used to solve the problem of product label copy
and spam. At the level of improved architecture and mecha-
nism. Li et al. [29] proposed the application of blockchain
technology to the database layer and communication layer
of the blockchain food safety traceability system through
analysis of the technical architecture level, and demonstrated
the design through the case of ham sausage effectiveness of
the program. Dwivedi et al. [30] designed a pharmaceutical
supply chain management system (PSCM) by combining
blockchain technology with traditional drug supply chain,
and proposed a mechanism to distribute all encryption keys
to all participants using smart contract technology. In the
design of blockchain storage model, many scholars used
on-chain and off-chain data storage to reduce pressure on
on-chain storage. Lin et al. [31] developed and implemented a
food safety traceability system based on blockchain and EPC
Information Service (EPCIS), and proposed an on-chain and
off-chain data storage structure to solve pressure problem of
blockchain data storage. Khaled et al. [32] proposed a method
of performing business transactions, where the Ethereum
blockchain and smart contract were used to track and trace
soybeans throughout the agricultural supply chain process,
coupled with the InterPlanetary File System (IPFS) for stor-
ing traceable data to reduce the amount of data stored on
the chain. Dong et al. [33] designed a reliable traceability
prototype system for the whole food supply chain based on
blockchain technology, and proposed a data storage mode
of ‘‘on-chain +cloud database’’. Thus ensuring the operat-
ing cost and efficiency of the blockchain system. Although
a large number of scholars have constructed a blockchain-
based traceability system, most of them have not considered
the issues of efficient information storage and privacy data
security.
III. DESIGN OF TRACEABILITY SYSTEM
A. SYSTEM FRAMEWOR K
This paper divides the traceability of agricultural products
into the links of production, processing, logistics, and sales.
The production link involves planting, transplanting, water-
ing, fertilizing and picking operations of fruits and veg-
etables agricultural products, and records key information
such as seedling information, planting process information,
environmental information, and product transaction infor-
mation. The processing link includes classifying, weighing,
packaging, pasting two-dimensional code and other opera-
tions for the picked fruits and vegetables, and recording the
product information, processing process, processing environ-
ment, product transaction and other key information. The
transportation link refers to the transportation of (such as
the Internet of Things) during the process of production,
processing, transportation, and sales. Using a traceability sys-
tem, detailed information about agricultural products can be
displayed to consumers, and the trust of consumers regarding
the safety of agricultural products can thereby be enhanced.
When quality and safety accidents occur with agricultural
products, law enforcement agencies can trace back to the
problem link and determine where the main responsibility
for the accident lies. Blockchain traceability refers to the
use of blockchain technology in traceability systems; such
traceability is achieved using blockchain characteristics of
decentralization, non-tampering, and traceability to ensure
the authenticity and transparency of traceability information
in agricultural products traceability systems, and achieves
effective and reliable traceability. The blockchain-based fruit
and vegetable agricultural products traceability system uses
the data storage scheme to manage the growth information,
processing information, logistics information and sales infor-
mation of fruits and vegetables agricultural products, so as
to monitor the whole process of agricultural products pro-
duction, processing, transportation and sales. The structure
of agricultural products blockchain traceability system was
mainly divided into storage layer, service layer, interface
layer and application layer, as shown in Figure 1.
36284 VOLUME 9, 2021
X. Yang et al.: Trusted Blockchain-Based Traceability System for Fruit and Vegetable Agricultural Products
FIGURE 1. Blockchain-based traceability system architecture.
Among them, the storage layer includes MySQL, the local
database of the system, and the CouchDB database that comes
with the blockchain system. The local database stores the
public information of each link. The blockchain system stores
the encrypted ciphertext of the private information and the
hash value of the public information to meet the needs of
fast data query and small storage space on the blockchain,
which ensures the security of private information. At the same
time, the authenticity comparison is provided when checking
whether the traceability information has been tampered with.
The service layer includes data analysis, reputation-based
smart contract, key management, key authorization, PBFT
consensus mechanism, etc. The interface layer is composed
of smart contracts, which mainly implement the data upload
blockchain system and data query functions. The data was
automatically executed when the label is printed, and the
query is automatically responded to when the terminal makes
a request. The application layer provides business functions
corresponding to different users of the system platform, such
as uploading data from enterprises in various links, querying
traceability information by consumers, and supervision by
government departments.
B. ON- CHAIN AND OFF-CHAIN DATA STORAGE
TECHNOLOGY
The existing storage mode of blockchain traceability sys-
tem involves directly writing traceability information of each
node of agricultural products into blockchain. With increas-
ing numbers of nodes, an increasing amount of transaction
data is obtained, and the storage load pressure of blockchain
consequently increases [34]. Due to the unique chain-type
structure of blockchain, the query efficiency is very low;
FIGURE 2. On-chain and off-chain data collaboration storage.
members of the same blockchain network access all the data
on the chain ledger. To overcome these shortcomings, this
paper improved the storage mode of a blockchain traceabil-
ity system for agricultural products and designed a method
involving the double storage of traceability information under
the chain of ‘‘database +blockchain’’. As shown in Figure 2,
after traceability data uploaded to the system, the system clas-
sifies the data. The product public information is stored in the
local database. The encrypted ciphertext and the hash value
of public information is uploaded to blockchain. Considering
the issue of storage space, the SHA256 algorithm with higher
VOLUME 9, 2021 36285
X. Yang et al.: Trusted Blockchain-Based Traceability System for Fruit and Vegetable Agricultural Products
FIGURE 3. Traceability information privacy protection data flow diagram.
security is used in the encryption of public information. For
the input of any length string, the SHA256 algorithm will
generate a 64-bit hexadecimal value. The block structure
of the traceability information data structure stored on the
blockchain includes block head and block body. The block
head mainly includes current block number, the hash value
of previous block, timestamp, and other information. The
block body mainly contains transaction-related information.
The storage format of the traceability information is shown
in Table 1. ‘‘Key’’ is the corresponding ID in the ‘‘Value’’
parameter, which acts as an index and a unique identification
field. The parameter ‘‘Value’’ is the value written to the
blockchain, including Type, ID, PrivateData and InfoHash.
The parameter ‘‘Type’’ is the name of the structure body that
is defined by the ‘‘Value’’ parameter. The parameter ‘‘ID’’ is
the unique identification field of the source message record,
which corresponds to the unique identification field of the
source traceability information stored in the local database.
The parameter ‘‘PrivateData’’ is the ciphertext of private
information encrypted by CBC algorithm. The parameter
‘‘InfoHash’’ is the value of source traceability information
after this information has been hashed.
C. TR ACEABILITY INFORMATION PRIVACY PROTECTION
PROCESS
The data involved in the entire supply chain is not only prod-
uct traceability information, but also contains private data
that only relevant companies can view, such as transaction
information. For competing enterprises, data privacy is an
important issue. This paper design a traceability informa-
tion privacy protection data flow that private information is
encrypted by deploying smart contract and uploaded to the
TABLE 1. Data structure of the blockchain used in the system.
blockchain together with the hash value of public informa-
tion. As shown in Figure 3, private information such as trans-
action data is encrypted by the Cipher Block Chaining (CBC)
mode of the AES encryption algorithm. The required Key1 is
randomly selected by the smart contract generate and upload
encrypted ciphertext to the blockchain. In order to ensure
the security of the Key1, this paper used the Elliptic Curves
Cryptography (ECC) to encrypt the Key1. The encrypted
Public Key authorized the viewing node. The Public Key of
the authorized viewing node and the Encrypted Key1 form a
key-value pair, stored in the world state of the smart contract
and written to the blockchain. When the relevant enterprise
nodes view the private data on the blockchain, the current
node private key is used to decrypt the Encrypted Key1 on
the blockchain to obtain the original Key1, and the Key1 is
used to decrypt the private information and view the private
information.
D. SMART CONTRACT FOR SYSTEM
Smart contract are essentially time-driven, stately computer
programs deployed on shared distributed database [35]. It can
36286 VOLUME 9, 2021
X. Yang et al.: Trusted Blockchain-Based Traceability System for Fruit and Vegetable Agricultural Products
TABLE 2. Smart contract interface.
be automatically executed when trigger conditions are met,
and node behavior can be transmitted and verified in an
informationized manner. Smart contract are called Chaincode
in Hyperledger Fabric, which greatly reduces the degree
of manual participation and ensures the decentralization of
blockchain and the tamper-proof of data. Alliance members
in the Consortium chains negotiate to participate in book-
keeping, and define incentive methods based on business
needs. The agricultural products traceability system based on
blockchain designed in this paper records traceability data
of fruit and vegetable products from production, process-
ing, transportation, and sales. In order to encourage alliance
members to upload traceability data, this paper designs a
reputation-based smart contract. After the node uploads the
traceability data that meets the requirements, the contract
logic will be triggered to increase the reputation value of the
node. Table 2 presents the interface of the smart contract.
Algorithm 1 Upload Information to Blockchain
Input: ID, Private Information, Hash Value of Public Infor-
mation, Public Key
1ID is the unique identification field of the source
message record
2 if Private Information!=NULL then
3Randomly generate Key1.
4PrivateData=CBC (Private Information, Key1).
5EncryptedKey =ECC (Key1, Public Key).
6InfoHash=Hash Value of Public Information.
7Add ID, PrivateData, EncryptedKey, InfoHash
in blockchain.
8 end
9 else
10 InfoHash=Hash Value of Public Information.
11 Add ID,InfoHash in blockchain.
12 end
Algorithm 1 elaborates the process of upload information
to blockchain. The system categorizes the information and
uploads it to the smart contract. After the smart contract
judges the information, the corresponding operation is exe-
cuted on the chain to realize the secure sharing of private
information in the blockchain network.
Algorithm 2 illustrates the process of granting reputation
value. When alliance member nodes upload traceability data
that meets requirements, the smart contract awards the node
reputation value in line with the consortium network incentive
mechanism, and incentivizes network nodes to upload trace-
ability data to ensure the integrity of the traceability chain and
the sharing of information between upstream and downstream
links.
Algorithm 2 Grant Reputation Value
Input: Peer, traceability data
1Peer is the member of organization of the Hyperledger
Fabric.
2Initialize reputation value when peer enters the con-
sortium chain.
3 if Peer upload traceability data then
4 if All nodes have reached a consensus after
verification then
5Trigger the smart contract issuing the reputation
value.
6 end
7 else
8Preview an error after returning the contract to its
previous state.
9 end
10 end
E. TR ACEABILITY ANTI-COUNTERFEITI NG PROCESS
The flow chart of fruit and vegetable agricultural products
blockchain traceability system is shown in Figure 4. The
traceability information is collected by Internet of Things
device or manually entered. Users upload the traceability
information of production, processing, logistics and sales
to the system. After classification by system, the trace-
ability information is divided into private information and
public information. The private information is uploaded to
blockchain after CBC encryption, and the public information
is stored in the local database. The SHA256 algorithm is used
to hash the public information. The hash value obtained is
stored in the blockchain system, and the block number is
returned. The block number is updated to the public infor-
mation record corresponding to the database. If the agricul-
tural products information needs to be modified, the hash
value of the public information needs to be rewritten into the
blockchain to update its block number. Consumers can obtain
public information and block number from the database by
scanning the QR code, hash the public information obtained,
and compare the consistency with the hash value stored on the
blockchain through the block number to determine whether
the product traceability information has been tampered with.
IV. IMPLEMENTATION OF THE TRACEABILITY SYSTEM
A. DEVE LOPMENT ENVIRONMENT
The blockchain-based traceability system for fruit and veg-
etable agricultural products mainly divided into two parts,
VOLUME 9, 2021 36287
X. Yang et al.: Trusted Blockchain-Based Traceability System for Fruit and Vegetable Agricultural Products
FIGURE 4. Anti-counterfeiting flowchart of system traceability.
one is the construction of consortium chain and the other is
the development of business system. Based on the Linux sys-
tem environment, the blockchain system builds a Hyperledger
Fabric distributed network environment based on multiple
Docker containers, using CouchDB as database. The shell
script is mainly used to build the network. The smart con-
tract is implemented in Go language, and the related depen-
dent NodeSDK development, implement Restful interface for
resource call. The application system is developed in .Net/C#
language, adopted the MVC (Model-View-Controller) archi-
tecture, the Bootstrap framework and a novel JavaScript
framework. The blockchain system Hyperledger Fabric runs
under the Linux environment and uses the Ubuntu operating
system with Inter Xeon Gold 6130 processor, 16GB memory,
100GB hard disk, and 100Mb/s bandwidth. The blockchain
network is deployed in a Docker container, as well as the
installation, testing, and instantiation of smart contracts need
to be performed in the Docker environment. As shown in Fig-
ure 5, the blockchain system in this paper adopts the deploy-
ment mode of single machine multi-node. The production
enterprises, processing enterprises, logistics enterprises, sales
enterprises and regulatory authorities in the supply chain are
added to the blockchain as five organizations, named org1-
org5 respectively. Each organization contains a peer node,
corresponding to a CA certification, used to track identity and
digital certificate management. The entire network contains
an orderer node, which is used to sort data and generate
blocks.
B. CONSE NSUS MECHANISM
Consensus mechanism is a vital part of the blockchain, which
ensures that the participating nodes in the blockchain net-
work store consistent information and the latest blocks are
FIGURE 5. Blockchain network environment construction.
accurately added to the blockchain. The current mainstream
consensus mechanisms include Proof of Work (POW), Proof
of Stake (POS), Delegated Proof of Stake (DPOS), practical
Byzantine fault tolerance (PBFT), etc. [36]. The core idea of
POW is to ensure data consistency and consensus security by
introducing the computing power competition of distributed
nodes, which can solve the problem of sybil attacks caused by
the free entry and exit of nodes. In order to avoid the power
consumption caused by the high dependence on the comput-
ing power of the nodes, the researchers proposed a proof of
stake mechanism POS that does not rely on computing power
and a delegated proof of stake mechanism DPOS. PBFT
consensus mechanism allow distributed system to reach a
consensus when there are a few malicious nodes. It uses
cryptographic algorithms such as signatures, signature ver-
ification, and hashing to ensure tamper-proof modification,
anti- counterfeiting, and non-repudiation during message
transmission. And optimized the BFT algorithm, reducing its
algorithm complexity from exponential level to polynomial
level. PBFT consensus mechanism is not suitable for large-
scale Public chain scenarios owing to consensus cost, and it
is more suitable for Consortium chains or Private chains with
fewer nodes. According to the characteristics of agricultural
product traceability, this paper chooses Consortium chain as
the basic network, and the consensus mechanism chooses the
PBFT mechanism apply to Consortium chain.
C. MODU LE DISPLAY
The system operation page includes eight modules which
are my farm, base management, production operations, crop
traceability, value-added services, system settings, big data
analysis, and blockchain management modules were shown
in Figure 6. Figure 6(a) shows the resume creation page.
Users can set the traceability information displayed to con-
sumers according to the traceability requirements, and the
hash value of the information will be stored in the blockchain.
Figure 6(b) is the blockchain management page. Users can
36288 VOLUME 9, 2021
X. Yang et al.: Trusted Blockchain-Based Traceability System for Fruit and Vegetable Agricultural Products
FIGURE 6. Blockchain-based traceability system operation page:
(a) Resume creation page. (b) Blockchain management page.
view the relevant information of the blockchain in real time,
such as the number of nodes, the number of contracts, the
number of blocks, and the number of transactions.
D. SYSTEM PERF ORMANCE ANALYSIS
This paper has carried on the comprehensive test to system
performance, by changing send rate to test system through-
put and network latency. An open source blockchain perfor-
mance evaluation tool Hyperledger Caliper is used, which
allowed users to perform performance tests on the blockchain
network through predefined use cases and obtain a set of
performance test results. The currently supported perfor-
mance indicators include transaction success rate, transaction
throughput, transaction latency, resource consumption, etc.
The performance of system is mainly reflected in two aspects:
transaction latency and transaction throughput under different
send rate. Transaction latency refers to the response time
required for a single user to send a request, including query
latency and write latency. Transaction throughput refers to the
number of successfully sent data per second, also falls into
two categories: query throughput and write throughput. This
paper tested transaction latency and transaction throughput
of the system. The test results of read request are shown in
Figure 7. It can be found from Figure 7(a) that when the send
rate of the read request is less than 250tps, the send rate of
the read request is linearly related to the read throughput.
When the send rate exceeds 250tps, the read throughput of
system reaches saturation. The system read latency is tested
by increasing the send rate from 0tps to 400tps in a total time
of 60 s, each increment of 50tps. The transaction latency test
result of the read request is shown in Figure 7(b). The results
show that the send rate has little effect on the read latency of
the system, and the average response time of the system is
stable at 0.02s. The test results of the input request are shown
in Figure 8. It can be found from Figure 8(a) that when the
send rate of the input request is less than 125tps, the send rate
of the input request is linearly related to the input throughput.
When the send rate exceeds 125tps, the input throughput of
the system reaches saturation. The system input latency is
tested by increasing the sending rate from 0tps to 200tps
in a total time of 60 s, each increment of 50tps. It can be
shown from Figure 8(b) that there is no obvious correlation
between the send rate and the input latency of the system,
and the average response time of the system has always been
kept at a low input latency of about 0.12s. In summary, the
fruit and vegetable agricultural products traceability system
based on blockchain technology proposed in this paper is
a low-latency, high-throughput system that satisfies actual
production needs.
E. APPLICATION OF THE TRACEABILITY SYSTEM
The system has been successfully applied to an apple com-
pany in Yantai City, Shandong Province, China. Through on-
the-spot investigation on the apple supply chain of enterprise,
the apple traceability system collection method was designed
from the planting and warehousing, processing, transporta-
tion and sales links, and the ‘‘database +blockchain’’ pro-
posed on the chain was used. The dual storage design of
traceability information, and the query method of the exter-
nal database index, have realized the efficient storage and
query of the traceability information of the apple blockchain.
Figure 9 shows the product label and product information
query method in practical applications. Figure 9(a) shows the
label information of apple. Users can scan the QR code on the
package and send a query request to the background server
to obtain the relevant traceability information of the product.
Figure 9(b) and Figure 9(c) are the schematic diagrams of
consumers inputting traceability source code on the web to
query product information and using mobile phone to scan
QR code to query product traceability information and data
storage certificate information. After the system apply in the
company, compared with the traditional traceability system
used before, the cost of equipment and system software has
increased by 7.3% and 24.9% respectively. The system appli-
cations requires to rent multiple server equipment. The cost
of developing and maintaining the system has increased to
a certain extent, and the training fee for workers to learn to
use software has increased slightly. In addition, the use of IoT
devices to automatically upload data to the system reduces the
waste of human resources and reduces labor costs by 5.3%.
According to financial statistics in 2019 and 2020, the total
cost of the product increased by 18.2%, and the sales revenue
increased by 34.6%. Although many factors influencing the
increase in sales revenue, the application of the blockchain-
based traceability system guarantees the authenticity and
VOLUME 9, 2021 36289
X. Yang et al.: Trusted Blockchain-Based Traceability System for Fruit and Vegetable Agricultural Products
FIGURE 7. System test results of the read request: (a) Effect of the send rate on blockchain network read throughput. (b) Effect of the send rate on
blockchain network read transaction latency.
FIGURE 8. System test results of the input request: (a) Effect of the send rate on blockchain network input throughput. (b) Effect of the send rate
on blockchain network input transaction latency.
TABLE 3. Comparison of traditional traceability system and
blockchain-based traceability system.
credibility of data, which is an significant reason for winning
the trust of consumers.
F. SYSTEM COMPARISON
Due to complicated links involved in the supply chain pro-
cess, Table 3 compares to six aspects of the traditional
apple traceability system with the blockchain-based apple
traceability system in terms of storage mode, data sharing
scope, data transparency, government supervision, data cred-
ibility and query rate. The traditional apple traceability sys-
tem adopts centralized management. Product manufacturers,
processing manufacturers, logistics parties, sales companies,
and terminal consumers of government regulatory depart-
ments are isolated from each other. They are isolated islands
of information, with low transparency of data and low gov-
ernment supervision. Product information is easy to be tam-
pered, and the credibility of the information is low. The
blockchain-based apple traceability system adopts all links
of companies and government regulatory departments as
nodes to join the consortium chain. It has the characteristics
of multi-centralization, does not rely on an organization or
individual, and realize information transparency, ensure the
integrity and accuracy of the information, and strengthen
government supervision. At the same time, real-time tracking
is performed to ensure the verifiability and reliability of
product quality, and to solve the ‘‘trust problem’’ in apple
supply chain management. The non-calculable modification
of the data using the blockchain ensures that the traceable
data cannot be tampered with, assures the credibility of the
data, and has ideal application functions such as tracing the
main body responsibility, tracking product flow, and sharing
regulatory information, which effectively protects consump-
tion rights. In order to solve the problem of low query rate of
the blockchain, this paper proposed an on-chain and off-chain
36290 VOLUME 9, 2021
X. Yang et al.: Trusted Blockchain-Based Traceability System for Fruit and Vegetable Agricultural Products
FIGURE 9. Product label and product information query method: (a) The tag of apple. (b) Query
results of traceability code input on the web. (c) Mobile phone scan query results.
storage structure, which can alleviate the storage pressure
brought to the blockchain system due to the continuous
increase of data under the condition of ensuring that the
data cannot be tampered, so that the query rate significantly
improved.
V. CONCLUSION
In this paper, we designed and implemented the traceabil-
ity system of fruits and vegetables agricultural products
based on the non-tampering and traceable characteristics of
blockchain, and discussed the storage and query design of
the system. To overcome the problems of high data load pres-
sure and poor private security of the blockchain traceability
system as the data grows, an on-chain and off-chain data
storage method using ‘‘database +blockchain’’ is proposed.
The public information displayed to consumers is stored in
the supply chain to the local database, whose hash value
by SHA256 algorithm was upload to the blockchain system.
The private information encrypted by the CBC encryption
algorithm is stored into the blockchain for sharing with
relevant companies. The storage method proposed in this
paper combines the actual situation, taking into account the
need for encryption of corporate private information as well
as the need for public supervision of supply chain public
information, and reduce the pressure of data load on the
chain. By storing the block number of the public informa-
tion on the database, the association between the blockchain
and the database is realized. The consumer obtains the
public information from the database by scanning the QR
code, and the system verifies the information according to
the corresponding block number stored in the database to
determine whether the product information has been tam-
pered with. With the development of blockchain, in order
to meet actual business needs, multi-chain is the future
development direction. For future research, we will fur-
ther explore the cross-chain technology between multiple
chains and a new type of consensus mechanism suitable for
traceability.
REFERENCES
[1] NBSC National Bureau of Statistics of China. (2019). National Data.
[Online]. Available: https://data.stats.gov.cn/
[2] G. Francois, V. Fabrice, and M. Didier, ‘‘Traceability of fruits and
vegetables,’’ Phytochemistry, vol. 173, May 2020, Art. no. 112291, doi:
10.1016/j.phytochem.2020.112291.
[3] J. Hu, X. Zhang, L. M. Moga, and M. Neculita, ‘‘Modeling and
implementation of the vegetable supply chain traceability system,’’
Food Control, vol. 30, no. 1, pp. 341–353, Mar. 2013, doi: 10.1016/j.
foodcont.2012.06.037.
[4] W. Li, S. M. Pires, Z. Liu, X. Ma, J. Liang, Y. Jiang, J. Chen,
J. Liang, S. Wang, L. Wang, Y. Wang, C. Meng, X. Huo, Z. Lan,
S. Lai, C. Liu, H. Han, J. Liu, P. Fu, and Y. Guo, ‘‘Surveillance
of foodborne disease outbreaks in China, 2003–2017,’’ Food Con-
trol, vol. 118, Dec. 2020, Art. no. 107359, doi: 10.1016/j.foodcont.2020.
107359.
[5] A. N. Desai, A. Anyoha, L. C. Madoff, and B. Lassmann, ‘‘Chang-
ing epidemiology of listeria monocytogenes outbreaks, sporadic cases,
and recalls globally: A review of ProMED reports from 1996 to
2018,’’ Int. J. Infectious Diseases, vol. 84, pp. 48–53, Jul. 2019, doi:
10.1016/j.ijid.2019.04.021.
[6] P. Luber, ‘‘The case of the European escherichia coli outbreak from
sprouts,’’ in Global Safety of Fresh Produce. Amsterdam, The Netherlands:
Elsevier, 2014, pp. 356–366.
VOLUME 9, 2021 36291
X. Yang et al.: Trusted Blockchain-Based Traceability System for Fruit and Vegetable Agricultural Products
[7] Regulation 178/2002 of the European Parliament and of the Council of 28
January 2002 Laying Down the General Principles and Requirements of
Food Law, Establishing the European Food Safety Authority and Laying
Down Procedures in Matters of Food Safety, Eur. Commission, Brussels,
Belgium, 2002.
[8] Food Safety Law of the People’s Republic of China, Order No. 21 of the
President of the People’s Republic of China C.F.R., Standing Committee
NPC, Beijing, China, 2009.
[9] M. M. Aung and Y. S. Chang, ‘‘Traceability in a food supply chain: Safety
and quality perspectives,’’ Food Control, vol. 39, pp. 172–184, May 2014,
doi: 10.1016/j.foodcont.2013.11.007.
[10] X. Yang, J. Qian, C. Sun, and Z. Ji, ‘‘Key technologies for establishment
agricultural products and food quality safety traceability systems,’’ Trans.
Chin. Soc. Agricult. Machinery, vol. 45, no. 11, pp. 212–222, 2014, doi:
10.6041./j.issn.1000-1298.2014.11.033.
[11] Y. Zhang, W. Wang, L. Yan, B. Glamuzina, and X. Zhang, ‘‘Develop-
ment and evaluation of an intelligent traceability system for waterless live
fish transportation,’’ Food Control, vol. 95, pp. 283–297, Jan. 2019, doi:
10.1016/j.foodcont.2018.08.018.
[12] F. Casino, V. Kanakaris, T. K. Dasaklis, S. Moschuris, and
N. P. Rachaniotis, ‘‘Modeling food supply chain traceability based
on blockchain technology,’’ IFAC-PapersOnLine, vol. 52, no. 13,
pp. 2728–2733, 2019, doi: 10.1016/j.ifacol.2019.11.620.
[13] L. Zhao, X. Bi, and A. Zhao, ‘‘Frame reconstruction of mobile traceabil-
ity information system for fresh foods based on blockchain,’’ Food Sci.,
vol. 41, no. 3, pp. 314–321, 2020, doi: 10.7506/spkx1002-6630-20181119-
217.
[14] K. Demestichas, N. Peppes, T. Alexakis, and E. Adamopoulou,
‘‘Blockchain in agriculture traceability systems: A review,’’
Appl. Sci., vol. 10, no. 12, p. 4113, Jun. 2020, doi: 10.3390/
app10124113.
[15] X. Yang, M. Wang, D. Xu, N. Luo, and C. Sun, ‘‘Data storage and
query method of agricultural products traceability information based on
blockchain,’’ Trans. Chin. Soc. Agricult. Eng., vol. 35, no. 22, pp. 323–330,
2019, doi: 10.11975/j.issn.1002-6819.2019.22.038.
[16] H. Yu, B. Chen, D. Xu, X. Yang, and C. Sun, ‘‘Modeling of rice supply
chain traceability information protection based on block chain,’’ Trans.
Chin. Soc. Agricult. Machinery, vol. 51, no. 8, pp. 328–335, 2020, doi:
10.6041./j.issn.1000-1298.2020.08.036.
[17] P. Zhu, J. Hu, Y. Zhang, and X. Li, ‘‘A blockchain based solution for
medication anti-counterfeiting and traceability,’’ IEEE Access, vol. 8,
pp. 184256–184272, 2020, doi: 10.1109/ACCESS.2020.3029196.
[18] Y. Lu, ‘‘Blockchain and the related issues: A review of current research
topics,’’ J. Manage. Anal., vol. 5, no. 4, pp. 231–255, Oct. 2018, doi:
10.1080/23270012.2018.1516523.
[19] K. Christidis and M. Devetsikiotis, ‘‘Blockchains and smart contracts for
the Internet of Things,’’ IEEE Access, vol. 4, pp. 2292–2303, 2016, doi:
10.1109/ACCESS.2016.2566339.
[20] Y. Lu, ‘‘The blockchain: State-of-the-art and research challenges,’’
J. Ind. Inf. Integr., vol. 15, pp. 80–90, Sep. 2019, doi: 10.1016/j.jii.2019.
04.002.
[21] Z. Zheng, S. Xie, H. Dai, X. Chen, and H. Wang, ‘‘An overview
of blockchain technology: Architecture, consensus, and future trends,’’
in Proc. IEEE Int. Congr. Big Data (BigData Congress), Jun. 2017,
pp. 557–564.
[22] X. Li, F. Lv, F. Xiang, Z. Sun, and Z. Sun, ‘‘Research on key tech-
nologies of logistics information traceability model based on con-
sortium chain,’’ IEEE Access, vol. 8, pp. 69754–69762, 2020, doi:
10.1109/ACCESS.2020.2986220.
[23] P. Olsen and M. Borit, ‘‘How to define traceability,’’ Trends
Food Sci. Technol., vol. 29, no. 2, pp. 142–150, Feb. 2013, doi:
10.1016/j.tifs.2012.10.003.
[24] J.-P. Qian, X.-T. Yang, X.-M. Wu, L. Zhao, B.-L. Fan, and B. Xing,
‘‘A traceability system incorporating 2D barcode and RFID technology
for wheat flour mills,’’ Comput. Electron. Agricult., vol. 89, pp. 76–85,
Nov. 2012, doi: 10.1016/j.compag.2012.08.004.
[25] D. Bumblauskas, A. Mann, B. Dugan, and J. Rittmer, ‘‘A blockchain
use case in food distribution: Do you know where your food has
been?’’ Int. J. Inf. Manage., vol. 52, Jun. 2020, Art. no. 102008, doi:
10.1016/j.ijinfomgt.2019.09.004.
[26] T. Feng, ‘‘A supply chain traceability system for food safety
based on HACCP, blockchain & Internet of Things,’’ in Proc.
Int. Conf. Service Syst. Service Manage., Dalian, China, 2017,
pp. 1–6.
[27] Y. Liao and K. Xu, ‘‘Traceability system of agricultural product based
on block-chain and application in tea quality safety management,’’
J. Phys., Conf. Ser., vol. 1288, p. 12062, Aug. 2019, doi: 10.1088/1742-
6596/1288/1/012062.
[28] J. Liu, T. Yang, and W. Wang, ‘‘Traceability system using public and
private blockchain,’’ J. Cyber Secur., vol. 3, no. 3, pp. 17–29, 2018, doi:
10.19363/j.cnki.cn10-1380/tn.2018.05.03.
[29] M. Li, D. Wang, X. Zeng, Q. Bai, and Y. Sun, ‘‘Food safety tracing
technology based on block chain,’’ Food Sci., vol. 40, no. 3, pp. 279–285,
2019, doi: 10.7506/spkx1002-6630-20171026-299.
[30] S. K. Dwivedi, R. Amin, and S. Vollala, ‘‘Blockchain based secured
information sharing protocol in supply chain management system with
key distribution mechanism,’’ J. Inf. Secur. Appl., vol. 54, Oct. 2020,
Art. no. 102554, doi: 10.1016/j.jisa.2020.102554.
[31] Q. Lin, H. Wang, X. Pei, and J. Wang, ‘‘Food safety traceability system
based on blockchain and EPCIS,’’ IEEE Access, vol. 7, pp. 20698–20707,
2019, doi: 10.1109/ACCESS.2019.2897792.
[32] K. Salah, N. Nizamuddin, R. Jayaraman, and M. Omar, ‘‘Blockchain-
based soybean traceability in agricultural supply chain,’’ IEEE
Access, vol. 7, pp. 73295–73305, 2019, doi: 10.1109/ACCESS.2019.
2918000.
[33] Y. Dong, X. Zhang, J. Xu, X. Wang, J. Kong, and P. Sun,
‘‘Blockchain-based traceability model for grains and oils whole supply
chain,’’ Food Sci., vol. 41, no. 9, pp. 30–36, 2019, doi: 10.7506/
spkx1002-6630-20190418-227.
[34] J. F. Galvez, J. C. Mejuto, and J. Simal-Gandara, ‘‘Future chal-
lenges on the use of blockchain for food traceability analysis,’’ TrAC
Trends Anal. Chem., vol. 107, pp. 222–232, Oct. 2018, doi: 10.1016/
j.trac.2018.08.011.
[35] A. Vacca, A. Di Sorbo, C. A. Visaggio, and G. Canfora, ‘‘A systematic
literature review of blockchain and smart contract development: Tech-
niques, tools, and open challenges,’’ J. Syst. Softw., vol. 174, Apr. 2021,
Art. no. 110891, doi: 10.1016/j.jss.2020.110891.
[36] Y. Yuan, X. Ni, S. Zeng, and F. Wang, ‘‘Blockchain consensus algo-
rithms: The state of the art and future trends,’’ Acta Automatica
Sinica, vol. 44, no. 11, pp. 2011–2022, 2018, doi: 10.16383/j.aas.2018.
c180268.
XINTING YANG received the B.S. degree from
Qingdao Agricultural University, Qingdao, China,
in 1997, the M.S. degree from Zhejiang University,
in 2000, and the Ph.D. degree from the Research
Center for Eco-Environmental Sciences, Chinese
Academy of Sciences, China, in 2008. In July
2000, he worked with the Beijing Agricultural
Information Technology Research Center and suc-
cessively served as a Research Intern, an Assis-
tant Researcher, an Associate Researcher, and a
Researcher, where he has been the Deputy Director, since 2012. His research
interests include blockchain, agricultural traceability, and machine learning.
MENGQI LI received the B.S. degree in net-
work engineering from Anhui Sanlian University,
Anhui, China, in 2018. She is currently pursuing
the M.S. degree with Shanghai Ocean University.
Her research interests include blockchain, cryp-
tography, and agricultural traceability.
36292 VOLUME 9, 2021
X. Yang et al.: Trusted Blockchain-Based Traceability System for Fruit and Vegetable Agricultural Products
HUAJING YU received the B.S. degree in soft-
ware engineering from the Jiangsu University of
Science and Technology, Jiangsu, China, in 2019.
He is currently pursuing the M.S. degree with
Shanghai Ocean University. His research interests
include blockchain and smart contract.
MINGTING WANG received the B.S. degree
in electronic information science and technology
from Fuyang Normal University, Anhui, China,
in 2017, and the M.S. degree from Shanghai Ocean
University, Shanghai, China, in 2020. Her current
research interests include blockchain and consen-
sus mechanism.
DAMING XU received the B.S. and M.S. degrees
from the Taiyuan Institute of Technology, Shanxi,
China, in 2013, and the M.S. degree from
the Taiyuan University of Technology, in 2016.
In July 2016, he worked with the Beijing Agri-
cultural Information Technology Research Center,
where he successively served as an Engineer. His
research interests include blockchain and agricul-
tural traceability.
CHUANHENG SUN received the B.S. degree
from Qingdao Agricultural University, Qingdao,
China, in 2002, the M.S. degree from Zhejiang
University, in 2005, and the Ph.D. degree from
China Agricultural University, Beijing, China,
in 2012. In July 2005, he worked with the Beijing
Agricultural Information Technology Research
Center, where he successivelyserved as a Research
Intern, an Assistant Researcher, an Associate
Researcher, and a Researcher. His research inter-
ests include blockchain, agricultural product identification, and agricultural
traceability.
VOLUME 9, 2021 36293
