Received April 21, 2019, accepted May 17, 2019, date of publication May 20, 2019, date of current

version June 18, 2019.

Digital Object Identifier 10.1109/ACCESS.2019.2918000

Blockchain-Based Soybean Traceability

in Agricultural Supply Chain
KHALED SALAH 1, NISHARA NIZAMUDDIN1 , RAJA JAYARAMAN 2 , AND MOHAMMAD OMAR2
1 Department of Electrical and Computer Engineering, Khalifa University of Science and Technology, Abu Dhabi, United Arab Emirates
2 Department of Industrial and Systems Engineering, Khalifa University of Science and Technology, Abu Dhabi, United Arab Emirates

Corresponding author: Khaled Salah ([email protected])

ABSTRACT The globalized production and the distribution of agriculture production bring a renewed focus
on the safety, quality, and the validation of several important criteria in agriculture and food supply chains.
The growing number of issues related to food safety and contamination risks has established an immense need
for effective traceability solution that acts as an essential quality management tool ensuring adequate safety of
products in the agricultural supply chain. Blockchain is a disruptive technology that can provide an innovative
solution for product traceability in agriculture and food supply chains. Today’s agricultural supply chains
are complex ecosystem involving several stakeholders making it cumbersome to validate several important
criteria such as country of origin, stages in crop development, conformance to quality standards, and monitor
yields. In this paper, we propose an approach that leverages the Ethereum blockchain and smart contracts
efficiently perform business transactions for soybean tracking and traceability across the agricultural supply
chain. Our proposed solution eliminates the need for a trusted centralized authority, intermediaries and
provides transactions records, enhancing efficiency and safety with high integrity, reliability, and security.
The proposed solution focuses on the utilization of smart contracts to govern and control all interactions
and transactions among all the participants involved within the supply chain ecosystem. All transactions are
recorded and stored in the blockchain’s immutable ledger with links to a decentralized file system (IPFS)
and thus providing to all a high level of transparency and traceability into the supply chain ecosystem in a
secure, trusted, reliable, and efficient manner.

INDEX TERMS Blockchain, Ethereum, smart contracts, traceability, Soybean, agricultural supply chain,
food safety.

I. INTRODUCTION of information in the agricultural/food supply chain where

Monitoring the development of agricultural products and products are produced, processed and sent via several
efficient logistics management in food and agricultural sup- intermediaries makes it difficult to track and trace. Product
ply chain is critical to ensure product safety. The growing contamination and its implications to public health strongly
concerns about food safety and contamination risks have emphasize traceability as necessary policy tool towards mon-
renewed the focus for enhanced traceability across the sup- itoring food quality and safety [5]. Dabbene and Gay [6]
ply chain [1], [2]. In addition agricultural products being argue the use of precise data collection via information com-
traded across several countries require precise tracking and munication tools such as bar-codes and RFID enables data
conformance to country specific regulations [3], [4]. Trace- acquisition and better traceability in agricultural and food
ability of products in agricultural supply chain requires supply chains. The current practice of traceability in agri-
collection, communication and management of critical infor- culture supply chain largely suffers from data fragmentation
mation by precisely identifying the origin, various infor- and centralized controls which proves vulnerable to both data
mation exchange in the supply chain. The dynamic nature modification and management. In the event of contamination
identifying the source and swiftly isolating the product from
The associate editor coordinating the review of this manuscript and supply chain requires close coordination among multiple
approving it for publication was Sabu M. Thampi. stakeholders in agricultural supply chain. Individual stages in

2169-3536
2019 IEEE. Translations and content mining are permitted for academic research only.
VOLUME 7, 2019 Personal use is also permitted, but republication/redistribution requires IEEE permission. 73295
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
 K. Salah et al.: Blockchain-Based Soybean Traceability in Agricultural Supply Chain

FIGURE 1. Existing product flows in Soybean supply chains.

food supply chains often have good traceability but exchange chain participants. The Ethereum Virtual Machine (EVM)
of information between stages proves to be difficult and time is the runtime environment for Ethereum computations on
consuming [7]. which the user programs are executed. A smart contract is
Recent technology developments through the applica- a protocol that is intended to digitally verify and carry out
tion of blockchain technology can provide a meaningful credible transactions that are traceable and irreversible, with-
and practical solution ensuring traceability of agricultural out third party intervention [15].
produce and eliminates the need for a trusted centralized The overarching objective of this paper is to demon-
authority [8]. Blockchain technology has gained immense strate how blockchain and Ethereum smart contracts can
popularity among supply chain and logistics community efficiently trace and track and enable seemless integra-
due transparency and immutability of transactions, enhances tion of business transactions and workflows in the agricul-
trust among participating stakeholders. Due to its tamper- tural supply chain. We propose, implement, and analyze a
proof, trusted, secure and traceable nature, blockchain can blockchain-framework to provide traceability and visibility
be deployed effectively in the agriculture and food supply in the soybean supply chains. Figure 1 illustrates the existing
chain management. The overall structure and functioning of product flows in soybean supply chain identifying the actors
food supply chain is vast and complex involving multiple and their corresponding function in the supply chain. The
stakeholders ranging from farmers, manufacturers, proces- process flow shown in Figure 1 is being adopted from pre-
sors, and consumers [9]. Food and agricultural supply chain vious studies on modeling traceability framework [16], [17]
is getting a lot of attention from the research community due for bulk grain supply chains. The primary contribution of this
to problematic long supply chain, from raw materials to the paper can be summarized as follows:
end consumer makes it extremely hard and time-consuming • We present a blockchain-based solution and framework
in tracking back the origin of a product. Hence, there is a need for traceability and visibility in the soybean supply chain
to create a secure framework for tracking details about the using Ethereum smart contracts.
origin, farming methods adopted, and safety of the food prod- • We discuss and highlight key aspects of our blockchain
uct throughout the supply chain cycle without a third party or solution in terms of the overall system design and archi-
centralized control. Few other major issues to be solved in the tecture, featuring main interactions among the main par-
supply chain cycle includes provenance, protocol regulations ticipants, with entity relations and sequence diagrams.
across multiple distributors, processors and retailers [9]. • We present, implement, and test smart contract algo-
The blockchain is basically an immutable and decen- rithms that govern and ensure the proper interactions
tralized, shared public ledger of transactions which allows among key stakeholders in the soybean supply chain.
participants to keep track of transactions without cen- The reminder of the paper is organized as follows.
tral record-keeping [10]–[12]. Blockchain is a shared dis- Section 2 presents related literature of blockchain applica-
tributed ledger composed of add-on blocks that include tions in food and agricultural supply chain. In Section 3,
details of all transactions data, execution outcomes and is we discuss the design, system overview, and sequence
traceable [13], [14]. Each block is hashed and linked to the diagrams for blockchain based approach for soybean trace-
next block, making it a secure chain of immutable and ability. In Section 4, we describe implementation details
tamper-proof records [14]. The adoption and implementation including algorithms for soybean sale between various par-
of blockchain technology can also ensure payment security ticipants in the supply chain using smart contracts. Finally,
to sellers while sharing important criteria about the origin, Section 5 concludes the study and outlines briefly research
certification of organic or non GMO, crop yields, and alert challenges and future work.
potential contamination risk among others. Ethereum is a
programmable blockchain platform that has the ability to II. RELATED WORK
govern business logic including interactions, sequence of In this section, we review and highlight related work found
events, and access control to enforce the required work- in the literature on blockchain applications for food and
flow and execute agreed-on business logic among supply agricultural supply chains. While literature on blockchain

73296 VOLUME 7, 2019

K. Salah et al.: Blockchain-Based Soybean Traceability in Agricultural Supply Chain

applications in banking, finance and insurance industries ever blockchain based commodity trade of US soybeans to
have been increasing steadily, the literature on food and China’s Shangong Bohi Industry [32].
agriculture is scant and just started to gaining popularity. As evident from those related work, there is clearly a grow-
Tian in [18] propose a food supply chain traceability based ing trend in adopting blockchain technology for enhanced
on Hazard Analysis and Critical Control Points (HACCP) information security, transparency, authentication of various
using blockchain and Internet of Things. Previously, Tian criteria in food and agricultural supply chains. A significant
in [19] discussed the advantages and disadvantages of RFID portion of the literature discuss the conceptual application
and blockchain for agriculture food supply chain traceability. of blockchain in agricultural supply chains falling short of
Caro et al. [20] present AgriBlockIoT a blockchain based specific implementation framework or approach. Our paper
traceability solution integrating data from IoT devices along aims to bridge the gap and contribute to the growing liter-
the value chain. They developed a use case for tracking pro- ature of blockchain application by showing how blockchain
duce from farm to fork and present implementation compar- and Ethereum smart contracts can offer an efficient, trusted,
isons through both Ethereum and Hyperledger. Tse et al. [21] secure, and decentralized trace and track solution for soy-
discussed at high abstract level how to apply blockchain tech- bean supply chains. Our work presents salient features of the
nology to food supply chain, and compared the blockchain- proposed system, with adequate details on architecture, meta
based solution with tradition solutions. The authors also data, sequence diagrams of interactions and messaging, and
highlighted key aspects related to security, integrity, and trust. algorithms - which are generic enough to be applied to almost
Lin et al. [22] review blockchain concepts for Agri-ICT all use cases related to agricultural supply chain involving
systems and present a model ICT system for agriculture using multiple stakeholders.
blockchain technology.
Tripoli and Schmidhuber [23] discuss the application of III. A BLOCKCHAIN-BASED APPROACH
distributed ledger technologies (DLT) and smart contracts FOR SOYBEAN TRACEABILITY
for increasing efficiency, provide traceability in the agri- In this section, we describe our proposed solution that utilizes
culture. The authors identify both technical challenges and Ethereum blockchain and smart contracts to trace, track, and
barriers to the adoption to conclude that DLTs have signif- perform transactions in soybean supply chains. Our solution
icant potential in achieving sustainable development goals. eliminates the need a trusted centralized authority and pro-
Mao and Dianhui [24] present a blockchain based credit eval- vides transactions and records for food supply chain manage-
uation system via smart contracts for efficient management ment and safety with high integrity, reliability, and security.
in food supply chain. Galves et al. [25] review challenges
and potential use of blockchain for assuring traceability and A. GENERAL SYSTEM OVERVIEW
authenticity in the food supply chains. Mao et al. [26] propose The Ethereum smart contracts has the potential to transform
a consortium blockchain approach to efficient food trad- safety of agricultural and food products into an integrated
ing system. They proposed an optimized improved practi- smart system that guarantees the quality of product delivered
cal byzantine fault tolerance (iPBFT) algorithm to optimize to the end customer. Our proposed framework and solu-
trading portfolio of buyers in the food supply chain. Further tion will focus on the usage of smart contracts executed
Mao et al. [26] validate their proposed food trading system autonomously on the public Ethereum blockchain platform.
using consortium blockchain via a case study applied to The execution of the smart contract functions and code will
Shandong province in China. Lucena et al. [27] present an be carried out by the thousands of mining nodes that are
approach for grain quality measurement using blockchain globally distributed, and the execution outcome is agreed
and smart contracts. They present implementation solution by all of the mining nodes. It is worth noting that mining
for a real life case that resulted in 15% added valuation for nodes are what make up the blockchain network. A mining
genetically modified (GM) free for soy grain exports from node can be any computing machine that collects, validates,
grain exporter business in Brazil. Chinaka [28] studied how and executes transactions. The nodes also store data and
implementing blockchain based solution can facilitate value outcome of these transactions in a ledger that is replicated
transfer by translating farmer’s assets such as livestock, farm and synchronized by all mining nodes. In a way a mining
lands and produce applied to small scale agriculture in Africa. node has the same exact replica of all other mining nodes.
Schneider [29] designed a prototype blockchain system to In blockchain, smart contracts will receive transactions in
enhance transparency and automate processes in the agricul- form of function calls, and will also trigger events to enable
tural sector. Holmberg and Aquist [30] studied the challenges the participating entities to continuously monitor, track and
in implementing a blockchain based traceability solution in get suitable alerts when violations occur. Thereby, it even-
the dairy industry. Some recent examples of blockchain pilot tually helps in restoring the conditions to the optimal and
implementation in food and agriculture supply chain includes responding to the violations that occur within the food value
wheat trade facilitated via blockchain technology in Australia chain. Our solution focuses on soybean supply chain to be
through Agridigital using Ethereum. Food Traceability Pilots particular in this case. Figure 2 illustrates a general overview
by Walmart through IBM Hyperledger [31]. Louis Dreyfus, of the system architecture of the proposed food supply chain
a large commodity trader, successfully completed the first management solution. As shown in the figure, the main

VOLUME 7, 2019 73297

 K. Salah et al.: Blockchain-Based Soybean Traceability in Agricultural Supply Chain

FIGURE 2. A system overview for automating the Soybean traceability using Ethereum smart contracts.

participating entities include the Seed Company, the farmer, the content of the file is stored by multiple peers or
Grain Elevator, Grain Processor, Distributor, Retailer, the end nodes, which stores the file content with high integrity
Customer and the blockchain that has the EVM executing and resiliency [33]. The hash of the file content is only
the smart contract. Furthermore, in blockchain, every actor or stored within the smart contract.
participant has to have an Ethereum account, with a unique • Grain Elevator:
Ethereum Address (EA) which uniquely identifies the actor. It is an agrarian facility that stores grain. The Grain Ele-
The Ethereum account basically consists of the EA with pub- vator operator determines the grade, quality of the grain
lic and private keys which are used to cryptographically and and purchases the grain from the farmer. Some important
digitally sign and validate the integrity of the data within each factors to be considered while storing the grain are -
transaction, and associate each transaction with a specific EA temperature, moisture, and duration of storage [16].
or account. • Grain Processor:
The grain processor buys grains from the elevator,
B. SYSTEM DESIGN refines the grain, analyses the grain for moisture, elimi-
Each participating entity has a role, association, and interac- nates foreign material, and converts the untreated grain
tions with the smart contract. There are seven participating into the final product without the need for any further
entities and their role are summarized as follows: treatment or processing.
• Seed Company: • Distributor:
Seed Company is the entity that produces a huge vari- A distributor is generally a warehouse that purchases
ety of seeds identified by EAN-UCC global standard final products from the processor. It is an entity which
identifiers per lot of the product sold to specific farmer. is involved in the process of food product distribution to
The seed company acts as a strong ally in food security the general population.
efforts, as it facilitates farmers’ access to planting mate- • Retailer:
rials in the form of seed, fertilizers and other nutrients Retailer buys finished goods from the distributor usually
that support agricultural production. in batches with traceable identifiers and sells to con-
• Farmer: sumers in small quantities. For example, the retailer may
Farmer buys seeds from the seed company with traceable buy in bulk and sell in lower unit of measure to the cus-
standard identifiers of the batch of seeds and the com- tomer. The standard identifiers preserves a hierarchical
pany involved in the sale transaction, cultivates crops relationship enabling tractability.
and creates the smart contract. The farmer also takes • Customer:
the liability for regularly monitoring and recording the The customer is the end user who purchases and con-
growth details of the crops and saving it on the decen- sumes the product from the retailer.
tralized file system as images or MPEG files. A popular To ensure secure tracking of the produce using Ethereum
decentralized file system can be IPFS (Inter Planetary smart contracts and to get all the participants involved in the
File System), which is peer-to-peer file system in which entire process, and as shown in Figure 2, the seed company

73298 VOLUME 7, 2019

K. Salah et al.: Blockchain-Based Soybean Traceability in Agricultural Supply Chain

FIGURE 3. Entity relationship diagram.

produces the soybean seeds and saves the details of seed Figure 3 shows the entity-relationship diagram that illustrates
germination, chemical composition, viability, quality, and the smart contract attributes and functions and the relation-
dormancy. The soybean seeds sold by the seed company are ship between the participating entities and the smart contract.
identified using standardized identifiers such as serialized Such metadata and relations are key to implement smart
Global Trade Identification Numbers (GTIN) or equivalent contracts.
which contains the specific company prefix. The use of In blockchain and IPFS, all images, data, and records
standard identifiers enables digital connectivity and track- are digitally signed and attributed to a certain actor. This
ing potential of both the products and transaction related means that the actor for uploading images, in this case the
processes among the participating entities in the agricultural farmer uploading MPEG files, is the undisputed owner of
produce supply chain. The farmer buys the seeds from the such action and is responsible for inaccurate or fraudulent
seed company and carries out the farming. The details of the images or MPEG files. Blockchain via smart contracts can
crop growth are recorded by the farmer at timely intervals in be programmed in an automated way to impose penalties on
the decentralized file system through IPFS. The crop growth the farmer if acts dishonestly. Another option, is to install
images are time stamped and the IPFS hash of the file is stored the cameras in the field with capabilities and communication,
in the smart contract. The farmer stores the grain in the eleva- to automatically take images and send them to the blockchain
tor after checking relevant factors such as temperature, mois- to be recorded and stored. Such hardware cameras can be
ture and time, that causes changes in grains during storage, built and provided to farmers to install, and can be designed
as heat and humidity lead to quality deterioration resulting in in a way that cannot be hacked or tampered with, and there-
production loss [17], [34]. The grain is then, purchased by the fore their uploaded images can be audited, trusted and their
processor which involves refining, analyzing the quality of content can be disputed or refuted by any participant or
grain, eliminating moisture and finally preparing the finished stakeholder on the blockchain.
product. The distributor buys the finished product from the
processor in order to ship the products to potential buyers. C. SEQUENCE DIAGRAMS
The distributor becomes the grain processors’ direct point Each of the participating entities has an Ethereum
of contact for prospective buyers of food products as they address (EA) and participates by invoking functions within
stack up large quantities of products [35]. After this point the smart contract. Figure 4 outlines the sequence flow for a
distributor sells the products/goods to the retailer. Generally, scenario where a farmer creates a smart contract. Following
a retailer buys small quantities of an item from a distributor the offline agreement between the farmer and the seed com-
or a wholesaler and sells products directly to customers. pany, the farmer buys the seeds from the seed company and

VOLUME 7, 2019 73299

 K. Salah et al.: Blockchain-Based Soybean Traceability in Agricultural Supply Chain

FIGURE 4. Sequence diagram showing the interaction between seed company, farmer, smart contract and elevator.

FIGURE 5. Sequence diagram showing interactions between elevator, smart contract, processor, and distributor.

an event SeedsRequestedByFarmer is invoked and is made active entities. Now, when the crop is harvested, there is an
available to all active participants (i.e., the farmer and the offline agreement between the farmer and the grain elevator
seed company). The seed company executes the function, to store the produced crop. The farmer is given the details
sellSeeds() that includes the attributes such as the Seed about the moisture, humidity, weight of the grain stored in
Company Ethereum Address (Seed Company EA), Ethereum the elevator and the duration of storage in the elevator. The
Address of Farmer (Farmer EA), Quantity, LotAttributes etc. farmer agrees to it and sells the grain to store in the elevator.
The crop growth details are updated by the farmer at regular Figure 4 shows the functions buyGrain() and sellToElevator()
time points onto the file system through IPFS. The farmer executed by the Grain Elevator and the Farmer respectively.
saves the image of the crop in the IPFS and stores the Figure 5 represents the message sequence diagram in
IPFS hash in the smart contract. The crop growth updating which the grain processor buys grain stored in the eleva-
process continues until the harvesting stage, with the image tor. The function buyGrain() is executed by the proces-
of the crop being recorded at frequent intervals. As shown sor by passing parameters such as the Ethereum addresses
in Figure 4, the updateGrowthImage() function records the of both requesting grain processor (Processor EA) and
growth of the crop. Each time an image is uploaded in that of the Grain Elevator (Elevator EA), Quantity and
the IPFS, an IPFS hash is stored in the smart contract and DateOfPurchase. The event GrainRequestedByElevator trig-
event CropGrowthImageUpdated is broadcasted among all gers the associated elevator, which then executes the

73300 VOLUME 7, 2019

K. Salah et al.: Blockchain-Based Soybean Traceability in Agricultural Supply Chain

FIGURE 6. Sequence diagram showing interactions among distributor, smart contract, retailer and customer.

sellGrainToProcessor() function. Event GrainSoldToEleva- central authority in the supply chain. Beginning with transac-
tor is broadcasted along with the network with the parameters tion of seeds sold between the seed company and the farmer,
showing the buyer and seller EA, Quantity and DateOfSales. to the next echelon the total volume of soybean produce
The distributor entity then requests to the grain processor in sold between subsequent entities is logged and all transac-
order to buy finished products from the processor. Figure 5 tions can be verified. For example, the volume of grain sold
shows buyProductFromProcessor() function executed by the between entities with the agreed conditions cannot be altered
interested distributor. The distributor is usually a warehouse or changed. In addition, grains of different quality criteria
that buys stores and ships products in bulk amounts to whole- cannot be mixed together for sale as total volume is known
salers or retailers. Event ProductRequestedByDistributor is to all stakeholders. The conditions of the agricultural field
triggered to invoke the processor to sell the grain to the and crop growth are difficult to monitor while the presented
distributor. The farmer executes the sellProductToDistribu- approach wherein the farmer periodically uploads images of
tor() function with the function parameters consisting of the the plant and land conditions via IPFS offers a digital record
Ethereum addresses of grain processor, distributor, quantity that can be used to validate the conditions agreed upon.
sold and date of sales. The event ProductSoldToDistributor is Continuous monitoring for quality compliance is further
activated to notify the actively interacting entities (i.e., Grain ensured by the use of traceable identifiers per lot and the
Processor and Distributor) at that particular time point. ability to trace all corresponding transactions between the
Figure 6 shows the message sequence diagram in which stakeholders. The quality of the shipment and conditions can
distributor, retailer, and the customer collaborate with the also be monitored using IoT-enabled containers and packages
smart contract. The distributor interacts with the interested equipped with sensors, cameras, GPS locator, and 4G com-
retailers to sell the goods and retailers in turn request for the munication. These sensors can continuously during the ship-
goods from the distributor in limited quantities. As shown ment process relay and send notifications on the conditions
in Figure 6, retailer executes buyProductFromDistributor() of the crop, product, and shipped items. With blockchain,
function and the event ProductRequestedByRetailer is acti- such information and notification cannot be altered or tam-
vated. The distributor executes sellProductToRetailer() func- pered with and immediately available and accessible by all
tion and event ProductSoldToRetailer notifies all participants stakeholders in a trusted and decentralized manner, with no
about the sale of goods. Now, the end customer buys the intermediaries. With this, additional attributes can be added to
product from the local retailer by executing buyProductFrom- address the traceable precise physical location of the product
Retailer() function, and event EndProductRequestedByCus- or stakeholder location by using standard identifiers such as
tomer is triggered by the smart contract. Finally, the retailer global location identifiers or by geotagging the stakeholder
sells the product to end customer by executing sellEndProd- location which can be sent by GPS sensors installed within
uct() function. The smart contract broadcasts the sale of the the shipping or storage containers.
product with the EndProductSold event. It is worth noting that it is possible that a stakeholder can
cheat or can transact and record fraudulent data. In this case,
D. TRACEABLE FUNCTIONALITY the blockchain does record the data as such with validated
The advantage of using traceable functionality in soybean attribution to the originating of the data (i.e. the true stake-
supply chain using our proposed blockchain-based solution holder). If at a later stage, the data was caught to be incor-
utilizing smart contracts is the availability of verifiable and rect, the judges and all participants can with 100% certainty
non-modifiable information to all stakeholders without a attribute the data to a particular actor or stakeholder. In this

VOLUME 7, 2019 73301

 K. Salah et al.: Blockchain-Based Soybean Traceability in Agricultural Supply Chain

scenario, blockchain can identify fraud. To overcome this Algorithm 2 Grain Processor Buys Grain From Elevator
type of fraud, blockchain can be programmed through smart Input: ‘gp’ is the list of registered Processors
contracts to have additional functions to invalidate shipment, Ethereumaddress(EA) of GrainProcessor,
or the whole supply chain process, and put some action to Ethereumaddress(EA) of Elevator Quantity,
impose penalties on fraudulent stakeholders, or take alterna- DatePurchased, GrainPrice
tive and corrective actions. This will constitute new corrective 1 Contractstate is BuyFromElevator
data and actions that will be generated and linked to the 2 State of the grain processor is GrainRequested
fraudulent data, and thus ensuring accurate and undisputed 3 Grain Elevator state is CropBoughtFromFarmer
traceability and auditability. 4 Restrict access to only gp ∈ GrainProcessor
5 if GrainSale is agreed and GrainPrice = paid then
IV. IMPLEMENTATION FRAMEWORK 6 Contract state changes to GrainRequestAgreed.
In this section, we describe the algorithms that define 7 Change State of the grain processor to
the working principles of our proposed blockchain-based WaitForGrainFromElevator.
approach. As discussed earlier, the farmer creates the 8 Grain Elevator state is SellGrainToProcessor
smart contract. The farmer then agrees to the purchase 9 Create a notification message stating sale of grain to
terms (offline) with one of the registered seed companies. requesting processor
Algorithm 1 describes the process that govern the sale of 10 end
seeds by the seed company to the farmer. After the initial state 11 else
of the contract is established, the smart contract checks to 12 Contract state changes to GrainRequestFailed.
confirm that the requesting farmer is already registered and 13 State of grain processor is RequestFailure.
the price of seed is paid. If the scenario is successful, then 14 Grain Elevator state is CancelRequestOfProcessor
the state of the contract changes to SeedRequestSubmitted, 15 Create a notification message stating request failure
the farmer state changes to WaitForSeeds and state of seed 16 end
company changes to AgreeToSell. The contract notifies all the 17 else
active entities in the chain about the state changes otherwise 18 Revert contract state and show an error.
the state of contract and other active participants reverts to 19 end
initial state and transaction terminates.

Algorithm 1 Seed Company Sells Seeds to Farmer

Input: F is the list of registered farmers to check two conditions as shown in Algorithm 2: (i) if the
Ether eumaddress(EA) of farmer. requesting grain processor is a registered entity and (ii) if the
Ethereumaddress(EA) of SeedCompany sale of grain is agreed and purchase price is paid. If these
Quantity, SeedType, SeedBrand,SeedPrice two conditions are true or satisfied, the contract state changes
1 Contractstate is Created to GrainRequestAgreed, processor state changes to WaitFor-
2 State of the farmer is SeedsRequested GrainFromElevator, elevator state changes to SellGrainTo-
3 Seed Company state is Ready Processor, and all the active entities are notified with a
4 Restrict access to only f ∈ F i.e., registered Farmers message on the sale of grain to the processor. In the other
5 if farmer = registered and SeedPrice = paid then case, if the above mentioned two conditions are not satisfied,
6 Contract state changes to SeedRequestSubmitted. contract state changes to GrainRequestFailed, processor state
7 Change State of the farmer to WaitForSeeds. changes to RequestFailure, elevator state changes to Cancel-
8 Seed Company state is AgreeToSell RequestOfProcessor.
9 Create a notification message stating sale of seeds The grain processor then sells the finished product to the
10 end distributors. Next, we elaborate the state of the system and the
11 else entities where the retailer buys the product from distributor.
12 Revert contract state and show an error. Date of product manufacture, Quantity Sold, and Date of
13 end Purchase are some of the important parameters to keep a
check. The distributors and retailers will be identified with
their Ethereum addresses and state of the contract as shown
Algorithm 2 describes the process of selling the grain to in Algorithm 3. At this stage, the contract state is Product-
Grain Processor by Elevator. The elevator stores the pur- SoldToDistributor, and distributor state is ProductReceived-
chased crop from farmer and stores it in mass quantities. FromProcessor. The state of the retailer is ReadyToPurchase.
Most important criteria to consider in this stage are moisture The contract restricts the access to only registered retailers
content, bin number, date of purchase and shipment date. and checks if sale agreement is accepted and product payment
At this stage, the contract state is BuyFromElevator. The is completed. If these conditions are met, the contract exe-
state of the grain processor is GrainRequested and the grain cutes the transaction where the distributor ships the product
elevator state is CropBoughtFromFarmer. The contract has to the retailer. Here, the state of the contract changes to

73302 VOLUME 7, 2019

K. Salah et al.: Blockchain-Based Soybean Traceability in Agricultural Supply Chain

Algorithm 3 Distributor Ships Product to Retailer Algorithm 4 Customer Buys From Retailer
Input: ‘r’ is the list of registered Retailers Input: Ethereumaddress(EA) of Retailer,
Etherenumaddress(EA) of Distributor, Ethereumaddress(EA) of Customer,
Ethereumaddress(EA) of Retailer, DatePurchased, Product ID.
DateM anufactured, Quantity Sold, SalesID
DatePurchased 1 Contractstate is SaleRequestAgreedSuccess
1 Contractstate is ProductSoldToDistributor 2 Retailer state is ProductDeliveredSuccessful
2 Distributor state is ProductReceivedFromProcessor 3 Customer state is ReadyToBuy
3 i Retailer state is ReadyToPurchase 4 Restrict access to only Customers
4 Restrict access to only r e Jietatler 5 if ProductPayment = successful then
5 if Sale = agreed and ProductPayment = successful then 6 Contract state changes to ProductSoldToCustomer.
6 Contract state changes to 7 Retailer state is ProductSaleSuccessjul
SaleRequestAgreedSuccess. 8 Customer state is SuccessfulPurchase
7 Distributor state changes to PmductSoldToRetailer. 9 Create a ’purchase success’ notification message.
8 Retailer state is ProductDeliveredSuccessful 10 end
9 Create a ’success’ notification message. 11 else
10 end 12 Contract state changes to SaleOfPmductDenied.
11 else 13 Retailer state is ProductSaleFailure
12 Contract state changes to SaleRequestDenied. 14 Customer state is FailedPurchase
13 Distributor state changes to RequestFailed. 15 Notify with a ’purchase failure’ message.
14 Retailer state is ProductDeliveryFailure 16 end
15 is Create a request failure notification message. 17 else
16 end 18 is Revert contract state and show an error.
17 else 19 end
18 Revert contract state and show an error.
19 end

The contract notifies with an event about the sales failure to

everyone in the network.
SaleRequestAgreedSuccess, and the distributor state changes
to ProductSoldToRetailer, and Retailer state changes to Pro- V. CONCLUSION
ductDeliveredSuccessful. For a successful product delivery In this article, we have proposed a solution and generic frame-
done, the contract sends out a notification message stating the work leveraging Ethereum blockchain and smart contracts
successful delivery to the retailer. Else, for a failure scenario, to trace, track, and perform business transactions removing
the contract state changes to SaleRequestDenied, state of dis- intermediaries and central point of processing for soybean
tributor becomes RequestFailed and retailer state changes to traceability across agricultural supply chain. We have pre-
ProductDeliveryFailure and the failure notification message sented details and aspects related to the system architec-
is sent out to all participants. ture, design, entity-relation diagram, interactions, sequence
Finally, we describe the algorithm for purchases made diagrams, and implementation algorithms. We showed how
by the customer from the retailer in Algorithm 4. The cus- our solution can be applied for tracing and tracking soybean
tomer is the final entity in the food processing and track- supply chain. However, the presented aspects and details
ing model. The customer state is ReadyToBuy initially. The are generic enough and can be applied to provide trusted
state of the contract and retailer are SaleRequestAgreed- and decentralized traceability to any crop or produce in the
Success and ProductDeliveredSuccessful respectively. Here, agricultural supply chain. To date, blockchain technology
the smart contract restricts access to only Customers to still faces key challenges related to scalability, governance,
make a purchase from the registered retailers. The impor- identity registration, privacy, standards, and regulations. As a
tant parameters considered to track the product are Cus- future work, we plan to look at addressing some of these key
tomer Ethereum address, Retailer Ethereum address, Date challenges and develop solutions addressing them. We also
Purchased, Sales ID, Product ID. Upon successful payment of plan to integrate within our proposed solution automated
the product price, the state of contract changes to ProductSol- payments and proof of delivery - in which parties are paid
dToCustomer, retailer state changes to ProductSaleSuccessful using cryptocurrency in an automated and centralized manner
and customer state changes to SuccessfulPurchase. If the by the smart contracts upon successful physical delivery of
payment made is not correct, the state of contract changes crops and products. A blockchain-based proof of delivery of
to SaleOfProductDenied, retailer state changes to Product- physical assets with automated payments in cryptocurrency
SaleFailure and customer state changes to FailedPurchase. as well as dispute handling was previously proposed in [36].

VOLUME 7, 2019 73303

 K. Salah et al.: Blockchain-Based Soybean Traceability in Agricultural Supply Chain

REFERENCES [26] D. Mao, Z. Hao, F. Wang, and H. Li, ‘‘Innovative blockchain-based

approach for sustainable and credible environment in food trade: A case
[1] M. M. Aung and Y. S. Chang, ‘‘Traceability in a food supply chain: Safety
study in Shandong Province, China,’’ Sustainability, vol. 10, no. 2, p. 3149,
and quality perspectives,’’ Food Control, vol. 39, pp. 172–184, May 2014.
2018.
[2] T. Bosona and G. Gebresenbet, ‘‘Food traceability as an integral part
[27] P. Lucena, A. P. Binotto, F. da Silva Momo, and H. Kim, ‘‘A case
of logistics management in food and agricultural supply chain,’’ Food
study for grain quality assurance tracking based on a blockchain busi-
Control, vol. 33, no. 2, pp. 32–48, 2013.
ness network,’’ (2018), arXiv:1803.07877. [Online]. Available: https://
[3] J. Hobbs, ‘‘Liability and traceability in agri-food supply chains,’’ in Quan- arxiv.org/abs/1803.07877
tifying the Agri-Food Supply Chain. Springer, 2006, pp. 87–102. [28] M. Chinaka, ‘‘Blockchain technology—Applications in improving finan-
[4] D. Mao, Z. Hao, F. Wang, and H. Li, ‘‘Novel automatic food trading cial inclusion in developing economies: Case study for small scale agricul-
system using consortium blockchain,’’ Arabian J. Sci. Eng., vol. 44, no. 4, ture in Africa,’’ Ph.D. dissertation, Sloan School Manage., Massachusetts
pp. 3439–3455, Apr. 2018. Inst. Technol., Cambridge, MA, USA, 2016.
[5] L. U. Opara and F. Mazaud, ‘‘Food traceability from field to plate,’’ [29] M. Schneider, ‘‘Design and prototypical implementation of a blockchain-
Outlook Agricult., vol. 30, no. 2, pp. 239–247, 2001. based system for the agriculture sector,’’ M.S. thesis, Fac. Bus., Econ.
[6] F. Dabbene and P. Gay, ‘‘Food traceability systems: Performance eval- Inform., Univ. Zurich, Zürich, Switzerland, 2017.
uation and optimization,’’ Comput. Electron. Agricult., vol. 75, no. 2, [30] A. Holmberg and R. Aquist, ‘‘Blockchain technology in food supply
pp. 139–146, 2011. chains: A case study of the possibilities and challenges with an implemen-
[7] J. Storoy, M. Thakur, and P. Olsen, ‘‘The TraceFood framework— tation of a blockchain technology supported framework for traceability,’’
Principles and guidelines for implementing traceability in food value M.S. thesis, Fac. Health, Sci. Technol., Karlstad Univ., Karlstad, Sweden,
chain,’’ J. Food Eng., vol. 115, no. 2, pp. 41–48, 2013. 2018.
[8] M. A. Khan and K. Salah, ‘‘IoT security: Review, blockchain solutions, [31] R. Kamath, ‘‘Food traceability on blockchain: Walmart’s pork and mango
and open challenges,’’ Future Gener. Comput. Syst., vol. 82, pp. 395–411, pilots with IBM,’’ J. Brit. Blockchain Assoc., vol. 1, no. 2, p. 3712, 2018.
May 2018. [32] (2018). Bloomberg. [Online]. Available: https://www.bloomberg.
[9] L. Lucas. Financial Times. (2018). From Farm to Plate, Blockchain Dishes com/news/articles/2018-01-22/dreyfus-teams-with-banks-for-first-
Up Simple Food Tracking. Accessed: Jun. 12, 2018. [Online]. Available: agriculture-blockchain-trade
https://www.ft.com/content/225d32bc-4dfa-11e8-97e4-13afc22d86d4 [33] N. Nizamuddin, H. Hasan, and K. Salah, ‘‘IPFS-blockchain-based authen-
[10] A. Bogner, M. Chanson, and A. Meeuw, ‘‘A decentralised sharing app ticity of online publications,’’ in Proc. Int. Conf. Blockchain (ICBC) (Lec-
running a smart contract on the Ethereum blockchain,’’ in Proc. 6th Int. ture Notes in Computer Science). Seattle, WA, USA: Springer, 2018.
Conf. Internet Things, 2016, pp. 177–178. [34] C. H. V. Bazoni, E. I. Ida, D. F. Barbin, and L. E. Kurozawa, ‘‘Near-infrared
[11] K. Salah, M. Rehman, N. Nizamuddin, and A. Al-Fuqaha, ‘‘Blockchain spectroscopy as a rapid method for evaluation physicochemical changes of
for AI: Review and open research challenges,’’ IEEE Access, vol. 7, stored soybeans,’’ J. Stored Products Res., vol. 73, pp. 1–6, Sep. 2017.
pp. 10127–10149, 2019. [35] D. Folinas, I. Manikas, and B. Manos, ‘‘Traceability data management for
[12] H. Hasan and K. Salah, ‘‘Combating deepfake videos using blockchain and food chains,’’ Brit. Food J., vol. 108, no. 8, pp. 622–633, 2006.
smart contracts,’’ IEEE Access, vol. 7, no. 1, pp. 41596–41606, Dec. 2019. [36] H. Hasan and K. Salah, ‘‘Blockchain-based proof of delivery of physical
[13] R. Beck, J. S. Czepluch, N. Lollike, and S. Malone, ‘‘Blockchain-the gate- assets with single and multiple transporters,’’ IEEE Access, vol. 6, no. 1,
way to trust-free cryptographic transactions,’’ in Proc. ECIS, May 2016, pp. 46781–46793, Dec. 2018.
p. 153.
[14] M. E. Peck, ‘‘Blockchains: How they work and why they’ll change the
world,’’ IEEE Spectr., vol. 54, no. 2, pp. 26–35, Sep. 2017.
[15] K. Toyoda, P. T. Mathiopoulos, I. Sasase, and T. Ohtsuki, ‘‘A novel
blockchain-based product ownership management system (POMS) for
anti-counterfeits in the post supply chain,’’ IEEE Access, vol. 5,
pp. 17465–17477, 2017.
[16] M. Thakur and K. A.-M. Donnelly, ‘‘Modeling traceability information in KHALED SALAH received the B.S. degree in
soybean value chains,’’ J. Food Eng., vol. 99, no. 2, pp. 98–105, 2010. computer engineering with a minor in computer
[17] M. Thakur and C. R. Hurburgh, ‘‘Framework for implementing traceability science from Iowa State University, USA, in 1990,
system in the bulk grain supply chain,’’ J. Food Eng., vol. 95, no. 2, and the M.S. degree in computer systems engi-
pp. 617–626, 2009. neering and the Ph.D. degree in computer science
[18] F. Tian, ‘‘A supply chain traceability system for food safety based on from the Illinois Institute of Technology, USA,
HACCP, blockchain & Internet of Things,’’ in Proc. Int. Conf. Service Syst. in 1994 and 2000, respectively. He is currently a
Service Manage. (ICSSSM), Jun. 2017, pp. 1–6. Full Professor with the Department of Electrical
[19] F. Tian, ‘‘An agri-food supply chain traceability system for China based and Computer Engineering, Khalifa University of
on RFID & blockchain technology,’’ in Proc. 13th Int. Conf. Service Syst. Science and Technology, United Arab Emirates.
Service Manage. (ICSSSM), Jun. 2016, pp. 1–6. He joined Khalifa University, in 2010, where he is currently teaching
[20] M. P. Caro, M. S. Ali, M. Vecchio, and R. Giaffreda, ‘‘Blockchain-based bachelor’s and master’s degree courses in the areas of cloud computing,
traceability in agri-food supply chain management: A practical imple- computer and network security, computer networks, operating systems, and
mentation,’’ in Proc. IoT Vertical Top. Summit Agricult.-Tuscany (IOT performance modeling and analysis. Prior to joining Khalifa University,
Tuscany), 2018, pp. 1–4.
he was the Department of Information and Computer Science, King Fahd
[21] D. Tse, B. Zhang, Y. Yang, C. Cheng, and H. Mu, ‘‘Blockchain application
University of Petroleum and Minerals (KFUPM), KSA, for ten years. He has
in food supply information security,’’ in Proc. IEEE Int. Conf. Ind. Eng.
over 190 publications. He holds three patents. He has been giving a number of
Eng. Manage. (IEEM), Dec. 2017, pp. 1357–1361.
international keynote speeches, invited talks, tutorials, and research seminars
[22] Y. P. Lin, J. R. Petway, J. Anthony, H. Mukhtar, S.-W. Liao, C.-F. Chou, and
Y.-F. Ho, ‘‘Blockchain: The evolutionary next step for ICT e-agriculture,’’
on the subjects of Blockchain, the IoT, fog and cloud computing, and cyber-
Environments, vol. 4, no. 2, p. 50, 2017. security. He is a member of the IEEE Blockchain Education Committee.
[23] M. Tripoli and J. Schmidhuber. (Aug. 2018). Emerging Opportunities He was a recipient of the Khalifa University Outstanding Research Award
for the Application of Blockchain in the Agri–Food Industry. [Online]. from 2014 to 2015, the KFUPM University Excellence in Research Award,
Available: FAO.org from 2008 to 2009, and the KFUPM Best Research Project Award from
[24] D. Mao, F. Wang, Z. Hao, and H. Li, ‘‘Credit evaluation system based 2009 to 2010. He was also a recipient of the Departmental Award for
on blockchain for multiple stakeholders in the food supply chain,’’ Int. J. Distinguished Research and Teaching in prior years. He is the Track Chair
Environ. Res. Public Health, vol. 15, no. 2, p. 1627, 2018. of the IEEE GLOBECOM 2018 on cloud computing. He is an Associate
[25] J. F. Galvez, J. C. Mejuto, and J. Simal-Gandara, ‘‘Future challenges on Editor of IEEE BLOCKCHAIN NEWSLETTER. He serves on the Editorial Boards
the use of blockchain for food traceability analysis,’’ TrAC Trends Anal. of many WOS-listed journals including IET Communications, IET Networks,
Chem., vol. 107, pp. 222–232, Oct. 2018. Elsevier’s JNCA, Wiley’s SCN, Wiley’s IJNM, J. UCS, and AJSE.

73304 VOLUME 7, 2019

K. Salah et al.: Blockchain-Based Soybean Traceability in Agricultural Supply Chain

NISHARA NIZAMUDDIN received the B.Sc. and MOHAMMAD OMAR is currently a Full Pro-
M.Sc. degrees in computer science from VIT Uni- fessor and the Founding Chair of the Department
versity, India, in 2010 and 2016, respectively. She of Engineering Systems and Management (cur-
was a Software Developer with BOSCH, India, for rently renamed Industrial and Systems Engineer-
two years. She is currently a Researcher with the ing). Prior to joining the Masdar Institute/KUST,
Department of Electrical and Computer Engineer- he has was an Associate Professor and a Graduate
ing, Khalifa University of Science and Technol- Coordinator with Clemson University, Clemson,
ogy, United Arab Emirates. She conducts research SC, USA. He was a part of the Founding Faculty
on projects involving Blockchain, cloud comput- Cohort of Clemson University research park in
ing, databases, and cybersecuirty. She recently Greenville, SC, USA. He has over 100 publications
published a number of research articles on blockchain applications and in the areas of product lifecycle management, knowledge-based manufactur-
infrastructure, ethereum smart contracts, and cybersecurity. ing, and automated testing systems, in addition to authoring several books
and book chapters. He holds four U.S. and international patents. He was
named a Tennessee Valley Authority Fellow of two consecutive years during
the Ph.D. degree, in addition to being a Toyota Manufacturing Fellow. His
professional career includes a Postdoctoral service at the Center for Robotics
and Manufacturing Systems CRMS, and a Visiting Scholar at the Toyota
Instrumentation and Engineering Division, Toyota Motor Company, Japan.
His group graduated seven Ph.D. dissertations and over 35 M.Sc. theses.
RAJA JAYARAMAN is currently an Associate Four Ph.D. students are currently on academic ranks in U.S. universities.
Professor with the Department of Industrial His work has been recognized by the U.S. Society of manufacturing engi-
and Systems Engineering, Khalifa University of neers SME through the Richard L. Kegg Award. He has also received
Science and Technology, Abu Dhabi, United the SAE Foundation Award for Manufacturing Leadership. In additios,
Arab Emirates. His expertise is in multi-criteria he has received the Murray Stokely Award from the College of Engineering,
optimization techniques applied to diverse applica- Clemson University. He has also led an NSF I/UCRC Center and a part of a
tions including supply chain and logistics, health- the DoE GATE Center of Excellence in Sustainable Mobility Systems. His
care, energy, environment, and sustainability. His current research interests include capabilities in composite fabrication and
research interests include systems engineering manufacturing analytics at a laboratory Masdar City Campus. His current
approach and process optimization techniques to research group supported two Postdoctoral Scholar’s Career Planning to
characterize, model, analyze complex systems for efficient supply chains, become an Assistant Professor at the Texas A&M (TAMUQ), in 2013, and
maintenance operations planning, and healthcare delivery. His Postdoctoral the University of Sharjah, in 2015.
research was centered on technology adoption and the implementation of He currently serves as an Editor-in-Chief for the Journal of Material Sci-
innovative practices in healthcare supply chains and service delivery. He has ence Research (Part of the Canadian Research Center), and as an Associate
led several successful research projects and pilot implementations in the area Editor for the Journal of Soft Computing (a Springer), handling the areas
of supply chain data standards adoption for the U.S. healthcare systems. of decision science, knowledge-based systems, in addition to his member-
His research has been published in top-rated journals including Annals of ship on several editorial boards and conference organizations. Furthermore,
Operations Research, IISE Transactions, Energy Policy, Applied Energy, he serves on the Advisory Board of the Strata PJSC (part of Mubadala
Knowledge-Based Systems, the Journal of Theoretical Biology, Engineering Aerospace).
Management Journal, and others.

VOLUME 7, 2019 73305