Research ArticleApplication of Blockchain and Internet of Things to EnsureTamper-Proof Data Availability for Food Safety

Adnan Iftekhar ,1 Xiaohui Cui ,1 Mir Hassan ,2,3 and Wasif Afzal4

1Key Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education,School of Cyber Science and Engineering, Wuhan University, Wuhan, China2Wuhan National Laboratory for Optoelectronics, Huazhong University of Science and Technology, Wuhan, China3University of Glasgow, Glasgow, UK4Malardalen University, Vasteras, Sweden

Correspondence should be addressed to Xiaohui Cui; xcui@whu.edu.cn

Received 27 November 2019; Revised 27 April 2020; Accepted 6 May 2020; Published 30 May 2020

Academic Editor: Luis Patarata

Copyright © 2020 Adnan Iftekhar et al. -is is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

Food supply chain plays a vital role in human health and food prices. Food supply chain inefficiencies in terms of unfaircompetition and lack of regulations directly affect the quality of human life and increase food safety risks. -is work mergesHyperledger Fabric, an enterprise-ready blockchain platform with existing conventional infrastructure, to trace a food packagefrom farm to fork using an identity unique for each food package while keeping it uncomplicated. It keeps the records of businesstransactions that are secured and accessible to stakeholders according to the agreed set of policies and rules without involving anycentralized authority. -is paper focuses on exploring and building an uncomplicated, low-cost solution to quickly link theexisting food industry at different geographical locations in a chain to track and trace the food in the market.

1. Introduction

Food supply chains are vast and spread over the wholeworld. -ese supply chains connect three significant sectorsof an economy. -e agriculture sector includes crops andlivestock, the food-manufacturing industry, and the mar-keting sector, which provides distribution, wholesales, andretails. -e three significant sources of food are crops,livestock, and seafood. We are contaminating our crops withpesticides and excess fertilizers.-e use of growth hormonesand the administration of drugs is becoming normal toincrease milk and meat production. It is severely affectinghuman health and increases the risk of various cancers inhumans [1]. -e physical contaminants with foreign ma-terials, persistent organic pollutants, tampering BB (bestbefore) dates, altering documentation, misprinting ingre-dients, and inappropriate storage temperatures are somesignificant sources raising food safety and public health risksduring processing and shipment process [2].

Health consciousness is also an increasing phenomenonin recent years. As consumers are becoming more and morehealth-conscious, the demand for certain productionmethods to meet specific environmental, ethical, and nutri-tional needs is increasing [3]. Meat and milk producers areintroducing organic, natural, and grass-fed production in themarket [4]. -ese livestock farmers and producers areclaiming their meat and milk production more nutritionaland superior than traditionally produced meat and milk. Inthe organic meat andmilk production system, the animals areraised on 100% certified organic food with strictly controlledvaccination. -e basic similarities and differences betweentraditional, natural, grass-fed, and organic production sys-tems are summarized in Table 1. -ese organic and safetycertified products are becoming more prominent in marketsas a trend towards purchasing organic food is growing amongconsumers [5]. -is emergence of new markets and a globalincrease in food prices also increase the growth of food fraudsand food safety risks [6].

HindawiJournal of Food QualityVolume 2020, Article ID 5385207, 14 pageshttps://doi.org/10.1155/2020/5385207

Food fraud refers to a group of activities that performintentionally or unintentionally for economic gain. Spink andMoyer [2] defined food fraud as “Food fraud is a collectiveterm used to encompass the deliberate and intentional sub-stitution, addition, tampering, or misrepresentation of food,food ingredients, or food packaging, or false or misleadingstatements made about a product, for economic gain.” -eauthenticity of the food and food labeling is becoming asignificant concern for manufacturers, regulatory authorities,and consumers [7]. -e food manufacturers and distributorsare tampering to substitute or alter the product ingredientswith inferior ones in order to set an appropriate price for thetargeted market or yield more significant profits. -is phe-nomenon is proving a fatal threat to human health. In 2008, aChinese milk scandal found infant milk formula contami-nated with melamine, which affected 0.3 million infants, ofwhich about 54,000 were hospitalized, and six died due tokidney stones and other related problems [8]. In 2013, the beefburgers in Britain and Ireland were found to contain horsemeat [9]. In 2014, a Chinese crime syndicate was arrested whosold about more than 1 million USD meat of rats, minks, andother small mammals under cover of mutton [10]. Some foodsafety-related incidents in South Korea from 1998 to 2016 arealso documented [11]. -e recent pandemic of coronavirusdisease (COVID-19) started in December 2019, believed tooriginate from an unregulated wild animal’s meat market inWuhan. At the time of writing this paper, it has infected 2million peoples worldwide, which resulted in about 0.1million deaths and billions of dollar economic loss. -eChinese Journal of Food Hygiene published a study in 2011,claiming that more than 94 million people fall ill due tofoodborne illnesses, resulting in about 85,000 deaths each year[12]. -e UK Government published a very detailed reviewand recommendation report on the integrity and assurance offood supply chain networks to protect United Kingdomconsumers in July 2014. -e report presented a national foodcrime control framework to protect consumers from foodfraud and ensure food safety. -is report discussed all thefactors of food supply chains and highlighted the potentialfactors that cause food supply chain failure and fraud pos-sibilities [13].

When it comes to supply chain management in terms offood safety, the visibility of the supply chain is an importantissue. -e food supply chains are more complex than theother supply chains. It is a big challenge to make sure thepresence of associated data in the food supply chains from

origin to the destination. -ese data are essential to preventfoodborne illness risks, food integrity issues, and issuingvarious food certificates. Aung and Chang [14] described theimportance of traceability in the food supply chain con-cerning food safety and quality improvement.

Moreover, consumers are now more concerned aboutevidence that the products they are buying are produced inproper environmental facilities with acceptable specifica-tions. We also observed a high demand for certified food andmeat in supermarkets of Wuhan during the Covid-19pandemic. It is increasing the need for producers to shareessential information with food certifiers, whole-sellers,retailers, and consumers. NGOs and government regulatoryauthorities are also demanding more transparency, visibility,and traceability throughout the supply chain from source toretail. United Nation Global Compact defines traceability as“the ability to identify and trace the history, distribution,location, and application of products, parts, andmaterials, toensure the reliability of sustainability claims, in the areas ofhuman rights, labor (including health and safety), the en-vironment, and anticorruption” [15]. To fulfill the demandsof the consumers, the food industry is also getting moreconcerned about tracking their suppliers and supply chain.Researchers from Wuhan University and Huazhong Agri-culture University have formed an alliance with variousfood-manufacturing companies, agriculture and livestockfarm owners, and information technology providers inWuhan, Hubei, to develop a system to trace the food fromfarms to the fork [16]. We will refer to this alliance as aconsortium in this article.

-e core purpose of this work is to merge the traditionalsupply chain management practices with the blockchain totrace a food package from farm to fork with unique identityfor each food package while keeping it uncomplicated for theworkers. It will keep the business transactions tamper-proofand accessible to stakeholders according to the businesspolicies and agreed contracts of data sharing between thecompanies without involving any centralized authority formonitoring.

-e rest of the article is organized as follows: Section 2discussed the background of the blockchain and Internet of-ings (IoT) technology. Section 3 presents the existingliterature on the application of IoT and blockchain tech-nology in the food supply chains. Section 4 presents themethodology we used for this work. Section 5 introduces thedetails of our traceability system design and demonstrates

Table 1: Similarities and differences between different meat and milk production systems.

Traditional harvest system Natural harvest system Grass-fed harvest system Organic harvest system

Food Pasture in grazing system,high-concentrate diet

Minimally processed, noartificial ingredients, no

preservatives

Fed on grass and grains, noneforage supplements

Certified organic,animals are raised inhealthier facilities

Healthmanagement

Vaccinations, antibiotics,ionophores, growth

hormones may be used

Vaccinations, limitedantibiotics, no ionophores,

no growth hormones

Vaccinations, no antibioticsused, no ionophores, no growth

hormones

Only strictly controlledprimary vaccination

allowed

Marketing Mainly auction markets Some level of verification isrequired

Source and managementverification are required aspacker, wholesaler, retailer

Complete verificationrequired from birth to

retailer

2 Journal of Food Quality

the architecture and functions of it using a real-world usecase from the food industry in Hubei, China. We analyzedhardware cost in Section 6 and challenges and obstacles inSection 7.Wemake a brief conclusion of this work in Section8. Finally, the future directions are given in Section 9.

2. Background

2.1. Blockchain. Bitcoin originated from the world’s firstblockchain-based application [17]. It is a list of transactionsthat are accessible by many participants and secured usingdigital signatures and cryptography hash functions. -is listis distributed across many systems over peer-to-peer net-works in almost real time, which makes it practically im-possible to change in previous transactions or makes it veryeasy to detect any illegal change in records [18]. -eblockchain transactions store into sequentially orderedblocks. Each block holds several transactions, and a hashsignature that links this block to the previous block acts like apointer that forms a cryptographically secured chain ofblocks called the blockchain, illustrated in Figure 1. -eminers of the network repeatedly perform a function to solvea complicated mathematical puzzle to find a unique hashsignature for each block which is a mathematical proof thatthe block is mined [19].

-e main benefit of the blockchain is its quality ofimmutability, which makes it secure and as well as easy toaudit trial [20]. -e blockchain can be programmed to re-cord virtually anything that is expressible in code. -eenterprises are already adopting this technology, and othersare moving towards this technology. In the manufacturingbusiness, the supply chain is the most critical factor. In atypical supply chain scenario, multiple independent partiestake part in moving payload from point A to point B, andthey must track it to all destinations. -e grain supply chainusually passes through multiple storages at multiple desti-nations shipped by multiple logistics from farmer to the endconsumers. Figure 2 illustrates the stages of the grain supplychain from farms to retailers. A tamper-proof distributedledger can record the travel of a specific batch of productionthat where, when, and who shipped or stored it or if it needsto be shipped somewhere in a specific time.

-e blockchain is classified into the three categories aspublic, private, and permissioned [21]. Bitcoin is a classicexample of a public blockchain where all the participants canjoin, read, and write data without any permission from anyauthority. Any participant can be part of the consensusprocess called mining [22]. -e private blockchain is limitedwithin an organization where the participants are knownand trusted. -e permissioned blockchain is an example of agroup of companies or consortiums where participants arebound to some legal contract to get permission to access,read, and write the blockchain. -e consensus process ofpermissioned blockchains is based on pre-elected nodeswithin the consortium. A summary of this classification issummarized in Table 2.

-ere are many blockchain development platforms toutilize blockchains in making secure and transparenttransactions between the organizations [23]. A comparison

of some famous blockchain development platforms is givenin Table 3.

2.2. Internet of2ings. Digitization is continuously growingand disrupting many aspects of our everyday life. -e IoTplatform enables the industry to have real-time tracking ofassets and the environment. It also puts a critical impact onasset velocity, i.e., the assets linked to the inventory in thebusiness world. -e adoption of blockchain technology inthe IoT industry is at its rise due to its provenance in securityand tracking [24]. It is becoming essential to keep an eye onthe animal’s health and maintain batches with high effi-ciencies to increase the quality of the product and loweroperating income. -e demand for livestock identificationand traceability increased the need for quality control andcontrol of infectious diseases, medication, and its effects onthe environment and consumer health [25, 26]. -e recentadvances and increasing phenomena of using radio-fre-quency identification (RFID) in society increased thestandardization of RFID tag technology for specific purposes[27]. -is phenomenon also expands the use of RFID-en-abled biocapsules in the global livestock market. -esecapsules allow the farmer to capture the real-time data aboutbody temperature, daily drinking cycles, ruminal pH level,and amount of activity among a large batch of cattle. -esedata can be gathered and monitored in real time withbeacons across the farm. -e farmer can receive the in-formation anywhere through the web or mobile application.-ese data enable farmers to calculate the time and optimuminsemination period after the estrus, which also preventscattle from calving accidents. Furthermore, veterinary health

Block 10 Block 11 Block 12

Proof of work:000000741dfesPrevious block:0000006548sde

Proof of work:0000006584sdePrevious block:0000004584vds

Proof of work:0000004584vdsPrevious block:0000006749sef

Merkle root

Merkle treeMerkle treeMerkle tree

Merkle tree

Hash 34

Hash 3 Hash 4

Hash 12

Hash 1 Hash 2

Txn 3 Txn 4Txn 1 Txn 2

Figure 1: Blocks forming the blockchain using hash signature.

Journal of Food Quality 3

services and owners are notified in real time with an alarmand message in case of any abnormality.

-e 2013 horse meat scandal found 100% horse meatcontent in beef products in some cases [28]. -e livestockfarm animals were RFID identified.-e beef originated fromDoly-Com Romania was identified as horse meat [29]. -ishorse meat was delivered to a cold storage company in Bredaby Draap Trading Ltd. Draap sold this meat to anothercompany in Europe named Spanghero. Spanghero sold thisbeef like meat and insisted that it received themeat labeled as“’Beef.” According to investigations reported by Frenchmedia, Spanghero tampered the documents regarding themeat [30]. -is was the failure of the traditional product anddata flow system as illustrated in Figure 3.

3. Related Work

-e blockchain is usable in medicine, economics, energy,and resource management. It is usable for exchanging al-most everything that has digital representation. Dubai isbecoming the first nation to use blockchains and the Internetto ensure food safety and consumer nutritional needs andpriorities. Dubai government is about to digitize all the fooditems from farm to fork using blockchains and Internet of-ings technology before Dubai expo 2020 [31]. In 2016,Walmart collaborated with IBM and Tsinghua University in-order to launch a food safety collaboration center in Beijingto improve the tracking of food items in the supply chain

with the help of blockchain technology. In 2017, Jingdong(jd.com) also joined this collaboration [32]. Jindong has alsomade a venture with an inner Mongolia-based beef pro-ducer, Kerchin, to apply blockchain technology in theproduction process. It enables its customers to track theinformation about the frozen meat, such as cow’s breed,weight, and diet as well as the location of farms by justscanning the QR code available on the food package. It is alsoabout to put more than 20 food items on the blockchain withKerching as a supply chain partner [33]. ZhongAn Tech-nology, a China-based technology company, developed aplatform to track and record the chicken farming. -ey arealso developing shared ledger-based business technologiesand strategic solutions [34]. Alibaba collaborates withBlackmores and many other Australia and New Zealand-based food producers and suppliers to create a blockchain-based platform that combats the rise of counterfeiters tar-geting Australian and New Zealand-based food items soldacross China on its platform [35]. IBM has teamed up withKrogen, McCormick and Company, McLane Company,Discall’s Tyson Foods, Golden State Foods, Unilever, Nestleand Dole, and many others to implement distributor ledgertechnology [36].

Alzahrani and Bulusu have proposed a block-supplychain based on blockchain and near-field communicationtechnology to tackle counterfeit products. -e authors makesure that only the node that has the product will offer a newblock, and the other nodes validate it. Second, the authors

Farm supplies andprimary storage

Grading, storage,and marketing

Production,warehousing,

and packaging

Distribution andwarehousing

Warehousing,and shelving

Farmer Grain market Processor Wholeseller Retailer

Figure 2: Storage and logistics are the backbone of an efficient supply chain.

Table 2: Classification of blockchains.

Public blockchain Private blockchain Permissioned blockchain

Read access No permission required fromany authority

Read access is private withinorganization participants

Public/participants are permissible undersome legal contracts

Write access No permission required fromany authority

Write access is private withinorganization participants

Participants are permissible under somelegal contracts

Consensusprocess

Anyone can join consensusprocess Pre-selected nodes within organization Pre-selected nodes within consortiums

Table 3: Comparison of blockchain development platforms.

Fabric VeChain Ripple R3Corda EthereumBlockchain type Permissioned Public Permissioned Permissioned Public and privateCryptocurrency None Ve-or Token Ripple None EtherSmart contract Yes Yes No Yes Yes

Security model Membershipservices

Block securityprotocol

Node validated and confirmedtransactions

Permissionedonly

Data are public and notencrypted

Integrationefforts Vary Easy Easy Vary Difficult

4 Journal of Food Quality

proposed a method to choose validators and a new con-sensus protocol based on Tendernint with the ability to selecta different set of validators for each validation event. -eysimulated that their new protocol is very efficient for largenetworks [37].

Tian analyzed the use of RFID in combination withblockchain technology and the possible advantages anddisadvantages of building an agri-food chain based on theRFID and blockchain [38]. He concluded that the RFID andblockchain enhanced the food safety and efficiency of thesupply chain. Moreover, a significant reduction in RFIDapplication prices will increase the use of RFID technologyin logistics over multiple times. In another article, hepresents a case study of building a food supply chain usingthe blockchain and IoTand demonstrates the hazard analysisand critical control points identification in real-time food-tracing scenario [39].

Ramundo et al. [40] present the potential of using thestate-of-the-art emerging technologies to build the foodsupply chain. -ey also evaluate the use of IoT platforms infood supply chains. -ey analyzed the use of IoT technol-ogies in the farm, processing, logistics, and distribution.-ey concluded that many companies adopt the alreadyexisting technologies but are continually experimenting with

new technologies, such as IoT, to stay at the top in the globalmarket.

A blockchain-based traceability system for the winesupply chain is proposed by Biswas et al. [41]. -e systemtraces the wine supply chain from grapes to bottle.

Tse et al. [42] presents the concept of blockchaintechnology and its potential use in information security ofthe food supply chain when compared to the traditionalcentralized supply chain.

Casino et al. briefly described the use of the blockchaininto various fields of supply chain management, trading,business, and transaction settlements [43].

Baralla et al. [44] describe the use of HyperledgerSawtooth to propose a framework to trace and secure thefood supply chain based in Europe. -e authors used atheoretical approach and concluded that blockchain tech-nology is highly useful for government officials to track andaudit the food supply chains.

Chen et al. [45] presented the challenges in the adoptionof blockchain technology for food supply chains. Olsen et al.[46] analyzed the cost benefits and limitations of applyingblockchain technology in the food industry. Johnson [47]concludes that the blockchain technology holds the strengthto regulate the food industry to prevent foodborne illness.Mondal et al. [48] proposed an architecture for the foodsupply chain based on the Internet of -ings to trace eachpacket of food within the supply chain in real time.

4. Methodology

-is work is based on a design-based research approach andthe concept of mindful use of information technology. -emindful use of technology focuses on using the most effectiveand cost-efficient features of a technology to contribute toproblem-solving [49]. Design-oriented methodology focuseson the analysis of practical problems from the real world bythe collaboration of researchers and practitioners to develop asolution using the existing design principles and technologicalinnovations [50]. -ese solutions are then further enhancedand improved by the required research and development todeploy in the production environment as a solution for de-fined problem. -e detailed workflow of our developedmethodology based on a design-based research approach isgiven in Figure 4.

We need to record the circulation of products and as-sociated data in the whole food supply chain to trace the foodfrom farm to fork. -e traditional supply chain environ-ments are based on conventional database technologies. Itdoes not fulfill our purpose, as there is no continuous flow ofinformation throughout the chain in these environments.

Figure 3 is a high-level illustration of the movement ofthe products and associated data in traditional supply chainenvironments. Every supplier in the traditional supply chainmoves the products and related data on its own. -eavailability of the data from farmers to the end consumer isscarce in these cases. -e distributed databases also have noconsensus mechanism, and an intruder or administrator cantamper the information.

Prod

uct

Dat

a

Figure 3: -e traditional flow of products and associated data.

Journal of Food Quality 5

In new emerging technologies, the blockchain is provinga potential candidate [18]. In blockchain-based supplychains, a single ledger is shared among all the entities of thesystem.-e end buyer has the same data shared by the initialsupplier plus the additional data being added at each stage ofthe production. Figure 5 is showing high-level details of ourpotential design to flow the products and associated data inblockchain-based supply chains. -e blockchain makes surethe availability of the tamper-proof data from each stage ofproduction to all the stakeholders.

To identify the product throughout the supply chain, theInternet of-ings (IoT) is almost standard [51]. Still, it failedto stop the sale of horse meat as beef in the traditional supplychain environment due to the lack of continuous data flowthroughout the supply chain [30]. -e emerging of IoT withthe blockchain is becoming a hot area of technologicaldevelopment.

-is methodology provides a heaven for small andmedium enterprises that cannot invest much to obtain ordeveloped new technologies. -is work fits best under de-sign-based research methodology as the research is intendedto solve a real-life problem by the construction of a newartifact using the already existing technology, and ourconsortium has members from all required fields. -e nextsection briefly describes the blockchain and IoT technologywith its core components important to this work.

5. Traceability System Designand Implementation

United Nations’ Global Compact Office defines the threetraceability models for tracing products in supply chains: thesegregation, mass balance and book and claimmodel, drawnin Figure 6. -e product segregation model makes sure theseparation of certified products from noncertified productsthroughout the supply chain. -e mass balance modelallowed the mixing of certified materials with noncertifiedmaterials in a controlled manner, and certified input shouldnot be less than certified output. -e book and claim modelrelies on the link between the volumes of the certified

material produced at the beginning of the supply chain andthe number of accredited products sold at the end of thevalue chain [15]. -is work aims to implement a productsegregation model using blockchain technology and inte-grate it with already existed infrastructure at organizationswithout disturbing the traditional business practices to alarge extent. We choose the Hyperledger Fabric as ablockchain platform. Hyperledger Fabric is a specializedplatform for enterprises to create their own blockchain. -eimportant features of Hyperledger Fabric are described inthe coming section. Hyperledger Fabric is an open sourceplatform and described itself as “an open source collabo-rative effort created to advance cross-industry blockchaintechnologies. It is a global collaboration, hosted by the LinuxFoundation, including leaders in finance, banking, Internetof -ings, supply chains, manufacturing, and technology.”

5.1. Hyperledger Fabric, an Enterprise-Ready BlockchainSolution. -e food industry is deploying informationtechnology to capture the market share and increase thecustomer’s trust. -e distribution companies are veryconscious of their potential customers. Any food safetydisaster or substandard product can affect their reputation,and they can lose customers. A group of organizations in-cludes food manufacturers, food distribution, and livestockfarms association and joins a consortium to improve datasharing, production quality, supply chain, customers trust,and market share [16].-emembers of a consortium are notlimited to industry only. It may also contain some regulatoryauthorities such as the Food and Drug AdministrationAuthority and even the animal diet producers and moni-toring organizations. -e typical benefits of the consortiuminclude but not limited to standardization, collaboration,and efficiencies.

Some most essential components and services providedby Hyperledger Fabric, an open-source blockchain platform,for enterprises which made it a strong candidate of choicefor this work are described below [52, 53]. HyperledgerFabric is a programmable network that encapsulates the

Literature review

Practicalproblem

definition

Analysis of practicalproblems by researchers

and professionals alliance

Development ofsolution informedby existing design

technologicalinnovation

Iterative cycles ofdevelopment,enhancement

and refinementof solution

Testenvironmentdeployment

Productionenvironmentdeployment

Is solutioncompliance?

Is solutionfound?

No

No

YesYesInitiate project

NoIs solutionapplicable? Yes

Update /maintenance

Figure 4: Research methodology.

6 Journal of Food Quality

business logic implementation and application of thebusiness network by way of smart contracts or chaincodes[54].

5.1.1. Permissioned Network. In a public blockchain net-work, the users downloaded the software and started tomake transactions without disclosing their identities. Inbusiness networks, it is not an appropriate way of thingswork. -e anonymity is not acceptable in business net-works. -e members in business networks always have aknown identity and assigned roles. -e Hyperledger Fabricis a permission-based network, and it assigns knownidentities and roles to carry out transactions. All the usersand components in the Hyperledger Fabric network areneeded to be identified on the network. -ese entities areassigned by the identities on the network by the Mem-bership Service Providers (MSPs) and Certification Au-thority (CA) by the Hyperledger Fabric using the PublicKey Infrastructure (PKI) to authorize and validate the usersand components.

5.1.2. Confidential Transactions. In the business world,confidentiality from unrelated parties is a critical feature inmany scenarios. Sometimes, the business networks want tokeep their transactions very confidential from the unnec-essary parties and only reveal them to the counterparty.Hyperledger Fabric provides the channel functionality toachieve transaction privacy between selected parties only.Each channel has one ledger, and there may be multiplechannels between consortium members on the samenetwork.

5.1.3. Consensus and Policy Support. Members of the con-sortium make many policies, decisions, rules, and regula-tions to run the consortium. Typically, consortium uses adecentralized approach for making decisions. Many ad-ministrators from member organizations decide with themajority to make any changes into the network, which af-fects the members of the consortium or business network.Such a decentralized decision system needs governance anddecision-making models. Hyperledger Fabric technologysupports decentralized administration by way of policies.

5.1.4. Identity Management. Hyperledger Fabric uses PKI(Public Key Infrastructure) for managing identities. Itprovides two tools: the active directory integration andFabric-CA Server for identity management. -e typicalprocess of generating identities is given in Figure 7. -eproof of identity is furnished by the identity owner to theregistration authority. -e registration authority validatedthe user information and passed it to the certification au-thority. -e certification authority creates an x509 certificateand gives it back to the owner and validation authority toprove its validity. -e other components, such as peers and

Traceability models

Product segregation Mass balance Book and claim

Identity preservationBulk commodity

Figure 6: Traceability models to trace products in supply chains.

1. Certificate of origin2. Certificate of attributes3. Certificate of facilities4. Processed data and pictures

5. Shipment date6. Order number7. Animal IDs8. Processed data

9 Batch data10. Production data

11. Item number12. Temperature13. Barcodes

14. Distribution center15. Shipment number15. Temperature

16. Receive date17. Temperature18. Packaging barcode20. Shipment number21. Order number

22. Receive date23. Order number24. Barcodes26. Invoice number27. Customer data

A B C D E F G

Figure 5: Flow of products and associated data on the blockchain.

Journal of Food Quality 7

orderer nodes, also require the identity to participate in thenetwork.

5.1.5. Application Development and Integration. It is rela-tively easy to integrate the Hyperledger Fabric blockchainwith the existing enterprise system. Every organization in thenetwork can develop the interaction system according to itsneeds. Fabric front end applications can be developed in-dependently using RESTful APIs as middleware, or thecustom middleware can also be designed using one of theSDKs provided by Hyperledger Fabric as shown in Figure 8.

5.2. System Architecture. Figure 9 is showing the systemarchitecture of this solution. -is system is initially installedat three locations. -e two peer nodes are installed at thelivestock farms organization premises. -e food-manufacturing organization and the food distribution or-ganization each holds two peer nodes. -e food-manufacturing organization additionally holds the two or-derer nodes and Kafka clusters. -is system is currentlydemonstrating Kafka as an ordering service. HyperledgerFabric also supports Raft-based consensus. -e Raft-basedordering service provides crash fault tolerance such as theKafka ordering service implementation, without the need tomanage external dependencies. Additionally, when usingRaft, ordering service nodes can be provided by differentorganizations across the world in various data centers. -eanchor peers of each organization communicate with theorderer and other peer nodes using the Internet while theKafka brokers are connected with orderer nodes over asimple layer 2 switch.

Figure 10 is demonstrating the general application servercomponents at each location. -e application server basedon NodeJS, which hosts Fabric SDK and NodeJS Express.-e NodeJS Express is providing RESTful APIs and variousfunctions to send, receive, and access data from existing IoTand software infrastructure at the organizations. -e ITsupport teams from all the organizations engaged duringthis development and deployment. -is application serverprovides a generic interface that can use to generate a newuser interface or integrate it into an already existing ap-plication interface. Every current organization or futureorganization joining this venture can integrate this solutioninto their existing system according to their needs with veryminimum efforts due to its simplicity.

5.3. IdentityManagementatLivestock. Our farms are alreadycertified as organic product producers. -e animals at thefarm are marked and injected with RFID tags. All therelated parameters of the animals are structured. -is tagcontains the necessary information stored in it, such as thedate of birth, breed, and ownership information. -eunique identification (ID) of the tag is the identity of theanimal, or the manual ID can be assigned. -e veterinarianscans and puts all the information regarding diseasecontrol, vaccination, and weight in the system. It is alsotracking and recording the conditions of the productionsite, which includes the environmental conditions in-cluding water quality, water temperature, air quality, en-vironment temperature and humidity, labor conditions,and the quality of the processes. -e data from the animaltags and environmental sensors including veterinary ser-vices are collected by installed antennas in the farms andsend to the workstation at the facility as well as to theenterprise HQ owning the farms as demonstrated in Fig-ure 11. -e system stores those data in the traditionaldatabases for internal use alongside posting these trans-actions on the blockchain. -ese certified organic farmswork on two models. -ey raised their batch of animals oronly provided services to food-manufacturing organiza-tions. In both scenarios, the related data are captured andstored according to the organic certification authority.-ese data are transferred to the customers with theproduct as product specifications. -e blockchain appli-cation is integrated in such a way that it does not affect anyexisting system.

-e system uses a chaincode to store data on theblockchain. It only transfers the required data at a fixedfrequency from the current traditional database to theblockchain using that chaincode. A new version of chain-code deployed for each new batch of animals. -e chaincodeis required to put or read the data from the blockchain.-esechaincodes are transferred to the customer with the productas product specifications.

-e farms also deliver organic meat instead of liveanimals. In this case, a separate chaincode is used tomanage the unique identity of each meat package. -efarm generates the set of new identities for each packageof meat against the animal identity as illustrated inFigure 12. -ese identities are stored on the blockchainfrom the system using a separate chaincode developedexplicitly for this purpose. -is new chaincode reads theanimal’s identities against each transaction using the

John Doe Registrationauthority

Certificationauthority

Validationauthority

Figure 7: Typical process to issue x509 certificates.

8 Journal of Food Quality

Applicationserver

Applicationserver

Endorser peer Endorser peer

Anchor peer

Applicationserver

Endorser peer

Anchor peer

Anchor peer

Organic foods ChFarms

Orderer

Ordererpeer

Ordererpeer

Broker0 Broker1

Broker3Broker2

Ka�a 1

Broker0 Broker1

Broker3Broker2

Ka�a 0

NodeJS express

Zee Distributors

IoTdevice

IoTdevice

IoTdevice

Internet of �ings

Figure 9: An overview of the system.

Traditional webapplication RESTful API

Desktop

application

Custom middleware

Existing enterprisesystem

HyperledgerFabric

Grunt

Figure 8: Hyperledger Fabric application development.

Journal of Food Quality 9

Environmental sensors

Bloused and ear-taged animal

Veterinary services

RFID antennaand scanners

Internet

Enterprise HQ

Base antenna

On location work stationconnected to

enterprise system

Figure 11: IoT-enabled farming.

Batch A: Ingredients [x0[1], a, b, c]

Batch B: Ingredients [Z[1], a, b, c]Batch C: Ingredients [Z[1], a, b, c]

Ingredients [x0[2], a, b, c]Ingredients [x1, a, b, c]

x = [x0, x1, ......., xn]

ID: y

ID: x

y: [z]

Figure 12: Product identification management throughout the supply chain.

SQL / moongoes

Express

NodeJS

User interface

Fabric SDK

Application server Hyperledger Fabricpeer nodes

Iot system

IOT

Existing ERPsystem

Figure 10: Application server.

10 Journal of Food Quality

previous-stage chaincode and stores this set of identitiesfor this product. Both chaincodes transfer with theproduct to the customer as product specifications. Itmakes each meat package traceable with the completehistory of the animal.

5.4. Identity Management at Food-Manufacturing Plant.-e food package can be traced and tracked into the supplychain from farms to a food processing company to thedistributor, distributor to the whole seller, and whole sellerto the retailer. It is possible with the help of bar code, QRcode, or RFID tag on each box of food. -e blockchain caneasily handle the identities and regulate who gets access tothe information behind each product. Each box of meatcontains an RFID or QR code which provides informationabout it, and it is possible to track that animal and itscomplete production process from birth to meat processingas shown in Figure 12. -e shipping company is providingreal-time data from its shipping trucks using its IoT plat-form, which are accessible by both the shipper and receiver.-ese data are also updating through a smart contract atspecific intervals on the blockchain.

5.5. Food-Manufacturing Plant. -e food manufacturerreceives its raw material shipment with its product speci-fication chaincode.-is chaincode enables the manufacturerto read the raw material specifications from the blockchain.-e manufacturer transfers these data from the blockchainto its traditional database management system for its in-ternal operations and records. If the company receives a liveanimal, it deploys the smart contract to update the set ofidentities against each package of food the same as a live-stock farm.

-e organization deploys a chaincode to record itsproduction on the blockchain. -is chaincode queries theraw material using the raw material specification chaincodeit receives with its raw material. -e organization systemassigns a unique identification number to finished productagainst the ingredients identification group. Each piece ofmeat package has its ID attached in slaughterhouse againstanimal ID. Each box of meat is scanned, and the informationis entered into the database. A set of new sub IDs created bythe system against each ID, assigned to each batch ofproducts where the meat is in use, as shown in Figure 13.-at identification also prints on the finished product in theform of QR or barcode. -ese identities are transferred fromtraditional databases to the blockchain using an explicitlydeveloped chaincode for a batch of production. -esechaincodes later deliver as a final product specification to thedistributor with the product delivery. A mobile phone ap-plication is also under development for consumers wherethey can trace the whole supply chain by providing theidentification of the finished product.

6. Hardware Cost

-is solution is cost efficient for small and medium orga-nizations that cannot invest in building a whole new system

from scratch. -e Hyperledger Fabric is an open-sourceplatform that receives regular updates. We used existingorganizations’ resources. All the nodes deployed on the 4thgeneration intel core i7 processor consists of 32GB of RAMand 2TB of storage media.

-e system does not interfere with existing infrastruc-ture but provides a separate layer of data sharing that can beintegrated into the system using a generic applicationstructure as a reference. In each stage of production, theorganizations are using their traditional database systems ortransfer the data from the blockchain to the system for theirbusiness tasks. -e blockchain is used only as a tool to makea record of transactions that cannot be altered and trans-ferred to the customer.

-is blockchain-based data sharing system is expandableas any organization can join or leave the blockchain con-sortium at any time. -e new coming organizations do notneed to change or develop the system according to theconsortium need. -ey can integrate it without much extradevelopment cost.

7. Challenges and Obstacles

-ere are many blockchain applications for the food supplychain under trial or implementation worldwide. Companiesand organizations are facing several challenges and obstaclesin blockchain adoption. Two major challenges and obstaclesare described in this section.

7.1. Interoperability and Universal Acceptance. -ere is notany formal road map to adopt the blockchain as globaltool for transactions. -e blockchain-based traceabilityfor the food supply chain is not even yet ready to im-plement as a nationwide uniform system. In manycountries, the rules, regulations, standards, and labelingare conflicted between the states. -ere is no technologyexist to connect different blockchain systems. -is lack ofinteroperability and data sharing is currently the biggestobstacle. -e blockchain system compared to the tradi-tional electronic traceability system is more homogeneousas it stores the transactions instead of state values whichmakes interoperability and sharing of data easier than thetraditional traceability systems. -e states and countriesneed to develop a standard ontology which defines whatthe recorded data elements and values mean in the foodsupply chain. -ey also need common messagingstandards.

7.2. Laws and Regulations. -e attempts to regulate theblockchain have been another area of controversy. -e roleof regulators and state authorities are not clear. -efunctioning of blockchain may also conflict with regula-tory requirements. For instance, information shared in aledger cannot be modified or altered. -is feature is incontradiction with the right to forgetting in manycountries.

Journal of Food Quality 11

8. Conclusion

-is paper applies blockchain technology in the food in-dustry in a holistic and mindful approach to measuring andcertifying the quality of the final product cost-effectively. Ithelps to provide transparent information about the foodfrom farm to fork. -is information is also useful for FoodControl Authorities to prevent potential food safety hazards.It also enhances healthy competition between companies toimprove product quality continuously. It provides decen-tralized, credible, and transparent information, being storedon the blockchain parallel to the traditional database systembehind the scene with existing enterprise systems withoutany noticeable change in conventional business operationsand training general employees.

9. Future Work

We are working with food certification authorities andNGOs to unify the type of data (ontology) that are requiredto certify and record for organic products within China. Itwill help to create a unified chaincode for different stages ofthe supply chain.-is will provide a cost-efficient and robustsystem with a unifying standard API.

Many governments are also showing interest in usingblockchain technology in governance. It needs to develop asystem where the different blockchains can transfer the dataon each other or can be merged. No business can be fullycentralized or decentralized without compromising in areas

such as security, privacy, performance, and scalability.Governments and private organizations are also concernedabout their data privacy and protection. It needs to develop amechanism to process the data on their servers at theirpremises and bring only desired results from the servers tothe blockchain for analysis and predictions. -e biggesthurdle for governments to be part of this chain is the laws toprotect the data that can be used to affect fair competitionand other national food security threats.

Data Availability

Data sharing is not applicable for this article.

Conflicts of Interest

-e authors have no affiliation with any organization with adirect or indirect financial interest in the subject matterdiscussed in the manuscript.

Acknowledgments

-e authors would like to acknowledge the support providedby the National Key R&D Program of China (no.2018YFC1604000).

References

[1] H. Malekinejad and A. Rezabakhsh, “Hormones in dairyfoods and their impact on public health-A narrative review

Add event to thesystem and blockchainagainst original RFID

Add event to thesystem and blockchain

RFID[n] tags QRcode application

RFID[n] tagsapplication

Slicing Preparing fooditems

RFID tagreader

RFID tagreader

Finished product withunique identification

Start

y

Yes

x0, x1, y

x 0, x

1, x n

x = [x0, x1, xn]

No

Batch A: [x0, a, b, c][x1, a, b, c]

Batch B: [y, a, b, c]

Furtherprocessing

Figure 13: Flowchart explaining the identity generation process.

12 Journal of Food Quality

article,” Iranian Journal of Public Health, vol. 64, no. 6, p. 742,2015.

[2] J. Spink and D. C. Moyer, “Defining the public health threat offood fraud,” Journal of Food Science, vol. 76, no. 9,pp. R157–R163, 2011.

[3] F. A. Shafie and D. Rennie, “Consumer perceptions towardsorganic food,” Procedia—Social and Behavioral Sciences,vol. 49, pp. 360–367, 2012.

[4] R. Machen, “Natural, grass-fed and organic beef (2010 TAMUbeef cattle short course),” 2010.

[5] M. B. Basha, C. Mason, M. F. Shamsudin, H. I. Hussain, andM. A. Salem, “Consumers attitude towards organic food,”Procedia Economics and Finance, vol. 31, pp. 444–452, 2015.

[6] E. Holbrook, “Dining on deception,” Risk Management,vol. 60, no. 4, pp. 28–31, 2013.

[7] S. Charlebois, A. Schwab, R. Henn, and C. W. Huck, “Foodfraud: an exploratory study for measuring consumer per-ception towards mislabeled food products and influence onself-authentication intentions,” Trends in Food Science &Technology, vol. 50, pp. 211–218, 2016.

[8] X. Pei, A. Tandon, A. Alldrick, L. Giorgi, W. Huang, andR. Yang, “-e China melamine milk scandal and its impli-cations for food safety regulation,” Food Policy, vol. 36, no. 3,pp. 412–420, 2011.

[9] P. J. O’mahony, “Finding horse meat in beef products—aglobal problem,” QJM: An International Journal of Medicine,vol. 106, no. 6, pp. 595–597, 2013.

[10] C. Buckley, Rat Meat Sold as Lamb Highlights Fear in China,-e New York Times, New York, USA, 2013.

[11] M. S. Park, H. N. Kim, and G. J. Bahk, “-e analysis of foodsafety incidents in South Korea, 1998–2016,” Food Control,vol. 81, pp. 196–199, 2017.

[12] X. Mao, J. Hu, and X. Liu, “Epidemiological burden ofbacterial foodborne diseases in China—preliminary study,”Chin Journal of Food Hygiene, vol. 23, pp. 132–136, 2011.

[13] C. Elliott, Elliott Review into the Integrity and Assurance ofFood Supply Networks-Final Report: A National Food CrimePrevention Framework, Department for Environment, Food &Rural Affairs Food Standards Agency, London, UK, 2014.

[14] M. M. Aung and Y. S. Chang, “Traceability in a food supplychain: safety and quality perspectives,” Food Control, vol. 39,pp. 172–184, 2014.

[15] B. S. Norton, A Guide to Traceability: A Practical Approach toAdvance Sustainability in Global Supply Chains, United Na-tions Global Compact Office, New York, USA, 2014.

[16] X. Cui, Wuhan University Food Safety and Food QualityBlockchain Alliance by Cui Xiaohui, Wuhan University,Wuhan, China, 2018, http://www.foodblockchain.com.cn/.

[17] A. Narayanan, E. Felten, A. Miller, and S. Goldfeder, Bitcoinand Cryptocurrency Technologies: A Comprehensive Introduc-tion, Princeton University Press, Princeton, NJ, USA, 2016.

[18] M. Swan, Blockchain: Blueprint for a New Economy, O’ReillyMedia, Inc., Newton, MA, USA, 2015.

[19] J. Kroll, I. C. Davey, and E. W. Felten, “-e economics ofbitcoin mining, or bitcoin in the presence of adversaries,” inProceedings of the Twelfth Workshop on the Economics ofInformation Security (WEIS 2013), pp. 1–21, Washington, DC,USA, June 2013.

[20] E. Kiktenko, N. Pozhar, M. Anufriev et al., “Quantum-securedblockchain,” Quantum Science and Technology, vol. 3, no. 3,2018.

[21] D. Massessi, “Public vs. private blockchain in a nutshell,” 2018,https://medium.com/coinmonks/public-vs-private-blockchain-in-a-nutshell-c9fe284fa39f.

[22] A. Baliga, “Understanding blockchain consensus models,”White Paper, 2017.

[23] V. Clincy andH. Shahriar, “Blockchain development platformcomparison,” in Proceedings of the 2019 IEEE 43rd AnnualComputer Software and Applications Conference (COMPSAC),vol. 1, pp. 922-923, Milwaukee, WI, USA, July 2019.

[24] A. Reyna, C. Martin, J. Chen, E. Soler, and M. Diaz, “Onblockchain and its integration with IoT: challenges and op-portunities,” Future Generation Computer Systems, vol. 88,pp. 173–190, 2018.

[25] J. McKean, “-e importance of traceability for public health andconsumer protection,” Revue Scientifique et Technique (Inter-national Office of Epizootics), vol. 20, no. 2, pp. 263–371, 2001.

[26] C. Saa, M. J. Milan, G. Caja, and J. J. Ghirardi, “Cost eval-uation of the use of conventional and electronic identificationand registration systems for the national sheep and goatpopulations in Spain,” Journal of Animal Science, vol. 83,no. 5, pp. 1215–1225, 2005.

[27] N. Adhiarna and J.-J. Rho, “Standardization and globaladoption of Radio Frequency Identification (RFID): strategicissues for developing countries,” in Proceedings of the FourthInternational Conference on Computer Sciences and Conver-gence Information Technology, Seoul, Korea, November 2009.

[28] BBC News, Findus Beef Lasagne Contained up to 100%Horsemeat, BBC News, London, UK, 2013, https://www.bbc.com/news/uk-21375594.

[29] BBC News, Horsemeat Scandal: France Blames ProcessorSpanghero, BBC News, London, UK, 2013, https://www.bbc.com/news/world-21464052.

[30] -e Local News Sweden, Food Safety Watchdog May ReportFindus, -e Local News Sweden, Stockholm, Sweden, 2013,https://www.thelocal.se/20130210/46112.

[31] IBM Blockchain Blog, One Nation’s Move to Increase FoodSafety with Blockchain, IBM Blockchain Blog, Armonk, NY,USA, 2018, https://www.ibm.com/blogs/blockchain/2018/02/one-nations-move-to-increase-food-safety-with-blockchain/.

[32] IBM News, Walmart, IBM and Tsinghua University Explorethe Use of Blockchain to Help Bring Safer Food to Dinner Tablesacross China, BBC News, London, UK, 2016, https://www-03.ibm.com/press/us/en/pressrelease/50816.wss.

[33] N. Kshetri, “5G in E-commerce activities,” IT Professional,vol. 20, no. 4, pp. 73–77, 2018.

[34] Y. Cen, H. Wang, and X. Li, “Improving business processinteroperability by shared ledgers,” in Proceedings of the 6thInternational Conference on Informatics, Environment, Energyand Applications, Jeju, Republic of Korea, March 2017.

[35] N. Edwards, “Blockchain meets the supply chain,” MHDSupply Chain Solutions, vol. 47, no. 4, 2017.

[36] D. Pyrzenski, I’ll Only Eat Blockchain Cereal with a FoodSafety Label on the Box, IBM Blog, Armonk, NY, USA, 2018,https://www.ibm.com/blogs/blockchain/2017/09/ill-only-eat-blockchain-cereal-with-a-food-safety-label-on-the-box/.

[37] N. Alzahrani and N. Bulusu, “Block-supply chain: a new anti-counterfeiting supply chain using NFC and blockchain,” inProceedings of the 1st Workshop on Cryptocurrencies andBlockchains for Distributed Systems, Munich, Germany, June2018.

[38] F. Tian, “An agri-food supply chain traceability system forChina based on RFID & blockchain technology,” in Pro-ceedings of the 13th International Conference on Service Sys-tems and Service Management (ICSSSM), Kunming, China,June 2016.

[39] F. Tian, “A supply chain traceability system for food safetybased on HACCP, blockchain & Internet of things,” in

Journal of Food Quality 13

Proceedings of the International Conference on Service Systemsand Service Management, Dalian, China, June 2017.

[40] L. Ramundo, M. Taisch, and S. Terzi, “State of the art oftechnology in the food sector value chain towards the IoT,” inProceedings of the 2nd IEEE International Forum on Researchand Technologies for Society and Industry Leveraging a BetterTomorrow, Bologna, Italy, September 2016.

[41] K. Biswas, V. Muthukkumarasamy, and W. L. Tan, “Block-chain based wine supply chain traceability system,” in Pro-ceedings of the Future Technologies Conference, Vancouver,Canada, November 2017.

[42] D. Tse, B. Zhang, Y. Yang, C. Cheng, and H. Mu, “Blockchainapplication in food supply information security,” in Pro-ceedings of the International Conference on Industrial Engi-neering and Engineering Management (IEEM), Singapore,December 2017.

[43] F. Casino, T. K. Dasaklis, and C. Patsakis, “A systematicliterature review of blockchain-based applications: currentstatus, classification and open issues,” Telematics and Infor-matics, vol. 36, pp. 55–81, 2019.

[44] G. Baralla, A. Pinna, and G. Corrias, “Ensure traceability inEuropean food supply chain by using a blockchain system,” inProceedings of the 2nd International Workshop on EmergingTrends in Software Engineering for Blockchain, Montreal,Canada, May 2019.

[45] S. Chen, J. Yan, B. Tan, X. Liu, and Y. Li, “Processes andchallenges for the adoption of blockchain technology in foodsupply chains: a thematic analysis,” in Proceedings of theiConference 2019 Proceedings, Washington DC, USA, March2019.

[46] P. Olsen, M. Borit, and S. Syed, “Applications, limitations,costs, and benefits related to the use of blockchain technologyin the food industry,” Report no: 4/2019, Nofima, Tromsø,Norway, 2019.

[47] W. G. Johnson, “Blockchain meets genomics: governanceconsiderations for promoting food safety and public health,”Journal of Food Law and Policy, vol. 15, no. 1, pp. 74–97, 2019.

[48] S. Mondal, N. Kriti, D. Kumar, and P. Chahal, “Blockchaininspired RFID based information architecture for food supplychain,” IEEE Internet of 2ings Journal, vol. 6, no. 3,pp. 5803–5813, 2019.

[49] E. B. Swanson and N. C. Ramiller, “Innovating mindfully withinformation technology,” MIS Quarterly: Management In-formation Systems, vol. 28, no. 4, pp. 553–583, 2004.

[50] K. Ashford-Rowe, “Applying a design-based research ap-proach to the determination and application of the criticalelements of an authentic assessment,” in Proceedings of theSecond Emerging Technologies Conference 2008, Wollongong,Australia, 2008.

[51] J. Gubbi, R. Buyya, S. Marusic, and M. Palaniswami, “Internetof things (IoT): a vision, architectural elements, and futuredirections,” Future Generation Computer Systems, vol. 29,no. 7, pp. 1645–1660, 2013.

[52] E. Androulaki, A. Barger, V. Bortnikov et al., “Hyperledgerfabric: a distributed operating system for permissionedblockchains,” in Proceedings of the 13th EuroSys Conference,EuroSys 2018, Porto, Portugal, April 2018.

[53] C. Cachin, “Architecture of the hyperledger blockchain fab-ric,” in Proceedings of the Leibniz International Proceedings inInformatics, LIPIcs, Dagstuhl, Germany, May 2017.

[54] S. Rouhani and R. Deters, “Security, performance, and ap-plications of smart contracts: a systematic survey,” IEEEAccess, vol. 7, pp. 50759–50779, 2019.

14 Journal of Food Quality