Youngjun Sung and Yoojae Won
Implementation of Quality Management System for Wild-Simulated Ginseng Using Blockchain
Abstract: A special government agency has been charged with implementing quality management to guarantee the quality of wild-simulated ginseng. However, these processes are carried out by use of documents, and this has resulted in information omission and high document management costs. To solve this problem, this study analyzed the existing quality management process by using a smart contract for the existing offline form and proposed a new quality management system for storing and managing all log data in the blockchain. This system reduced documentation management costs about quality management and recorded information in the previous step through the quality management steps, thus forming a step-by-step record chain. Experiments were conducted by implementing this system, which improved data integrity and reliability. Additionally, sensitive information, such as personal information, was included in the system by use of the off-chain technology.
Keywords: Blockchain , Smart Contract , Off-Chain , Quality Management System
There is a growing interest in food safety worldwide and an increase in the production of organic foods. Since consumers may find it difficult to distinguish these organic foods from regular foods, every country certifies food quality through the establishment of certain laws or standards . South Korea has established a food quality certification system, which has passed several quality management laws (e.g., laws related to pesticide-free and organic foods). Wild-simulated ginseng, a forest product, has been designated as a domestic quality management food. Wild-simulated ginseng is grown in an environment similar to that of ginseng in the wild, and no pesticides are used during its cultivation process. This differs from the cultivation process for general ginseng, and consumers often find it difficult to identify this difference; therefore, a special law requires government agencies to manage all ginseng production processes and certify their quality  (Table 1, Fig. 1).
The quality management process for wild-simulated ginseng manages all processes in a step-by-step manner from the start of cultivation to sale to the consumers. Wild-simulated ginseng that has passed all the steps can be distributed and sold with a certificate of acceptance provided by government agencies. However, quality management is an inefficient process for documenting and managing all processes for at least seven years. The process of sales requires more than 20 documents which must be maintained separately. This method is inefficient because it incurs significant document management costs. Even in the production record process, which is one of the quality management stages, producers who grow wild-simulated ginseng maintain handwritten records .
The period of inspection by the management agency is very long (3 years). Inspection cycles can cause information omission during the production process. Consequently, a transparent production process is not provided to consumers, and consumers have to trust and purchase the wild-simulated ginseng using only a single authentication method, a pass certificate. To solve this problem, we intend to provide a transparent quality management system using blockchain technology, which has been recently used in various fields such as supply chain management (SCM) and digital certificates [4,5].
In this study, we propose a quality management system that satisfies all legal requirements for the quality management of wild-simulated ginseng. The system utilizes a private blockchain to form a blockchain network for the quality management stakeholders. After network configuration, the existing quality management process is analyzed, and smart contracts are introduced into the step-by-step process to automate the offline process online and record step-by-step information on the blockchain. Lastly, due to the nature of the blockchain, the information recorded therein is also disclosed to third parties, so it is not suitable to record information in the same format as personal information. To this end, the personal information included in the form uses off-chain technology to prevent sensitive data from being recorded and exposed on the blockchain.
2. Related Work
2.1 Blockchain-based Data Management
2.1.1 Supply chain management
SCM technology manages flows so that the entire process from raw material procurement to product delivery can proceed smoothly. The technology tracks and manages information regarding products (e.g., supply, manufacturing, wholesale, and logistics) from a single value chain perspective. As supply chains develop, their management and control may become complex. However, as SCM is an important factor in increasing company competitiveness, it is important to strengthen it, for example, using models that evaluate the degree of trust between supply chain partners . It is also important to increase reliability within SCM, for example, using blockchain technology (a distributed ledger technology). Blockchain technology provides transparency, traceability, and security and can mitigate several SCM problems. Thus, it is important to increase the number of cases in which blockchain technology is applied to SCM .
2.1.2 Digital certificates
Certificates, a type of official document used for fact-checking, can be used in various ways (e.g., certificate of completion, confirmation, certificate, and diploma). Offline certificates pose the risk of being lost and have several other disadvantages that users must manage separately. Conversely, digital certificates have several advantages, such as a lower risk of loss and convenient management. However, digital certificates also pose certain risks, such as forgery and falsification; hence, a separate management process is required to verify their authenticity . Blockchain technology can provide a solution to this problem. In addition to the distributed storage and management of existing digital certificates in a blockchain, blockchains and smart contracts can be used to automatically issue and manage digital certificates when certain conditions are met. A digital certificate solution utilizing blockchain technology is proposed in this paper .
2.1.3 Internet of Things management
The Internet of Things (IoT), an interconnected smart device, shares sensed data through a network and enables the monitoring, sharing, and adjustment of interactions between things that are connected through the Internet. Several methods have been proposed to utilize IoT management by applying blockchain technology .
Consider an example of a smart car that records and manages all state data and errors that occur in the vehicle (e.g., information regarding tires, oil, and temperature) in a blockchain . Recent research has focused on the use of blockchains to manage the operational integrity of black box data inside cars, and cases of blockchains being used to maintain data integrity are growing .
2.2 System Configuration
Blockchain technology distributes and manages data in blocks to multiple nodes in a P2P network . The nodes can select other nodes for distributing blocks using a consensus algorithm before distribution . The selected node creates a new block by writing the hash value of the previously created block in the block header. After this block generation process, all blocks (except the genesis block) form a hash chain that stores the hash value of the previous block. The hash chain maintains data integrity and prevents malicious users from forging and manipulating data. The hash chain is formed through the process shown in Fig. 2.
A given blockchain can be broadly divided into a public blockchain and a private blockchain based on its configuration, and the relevant differences are listed in Table 2. A public blockchain is an open network where anyone can participate; although free participation is possible, it has the disadvantage of slow block generation. In a private blockchain, each participant receives a different type of permission, and only a small number of nodes with consensus authority can proceed with consensus formation, which allows faster block generation but forfeits the benefit of decentralization.
2.2.2 Smart contract
The smart contract concept was first proposed by Nick Szabo , where contracts between contracting parties are written in code and automatically executed when the conditions are met. Buterin  implemented the smart contract concept using the Solidity language provided by Ethereum, which allows for the use of various blockchains for various purposes (e.g., electronic voting and real estate transactions) .
The nature of blockchain allows for the opening of the data stored in the block to the network participants, so this system is unsuitable for storing sensitive information or large files with extensions such as img, mp3, and mp4 in the block.
The off-chain mechanism provides functions for managing private data and large-capacity data in the blockchain. Once data are stored outside the block, metadata referring to the corresponding raw data can be recorded and managed in the blockchain. The storage methods for the on-chain and off-chain mechanisms are shown in Fig. 4.
3. Proposed Quality Management System
3.1 Existing Quality Management Process
From cultivation to the market, wild-simulated ginseng production quality management is executed by two agencies (Agency1 and Agency2). The applications for each course and the issuance of the results are all completed through offline documents. A certificate attesting the previous step has been passed must be attached when applying for permission to conduct the next step.
Before commencing wild-simulated ginseng production, the ginseng must be inspected by Agency2 to determine its suitability for production. Production commences only after Agency2 has obtained inspection results and filed a favorable report. The production lasts at least seven years, and every three years, Agency2 inspects production progress and records its findings. To be allowed into the market produced wild-simulated ginseng must be inspected for quality. During this process, all the aforementioned records are investigated. Only wild-simulated ginseng that has passed all these inspections may be awarded a certificate of acceptance and thereby allowed into the market. Even after distribution, the two agencies inspect whether it is the accredited ginseng that was distributed. Figs. 5 and 6 present brief flowcharts depicting this process.
3.2 Quality Management using Blockchain
To apply blockchain to the existing quality management process, we analyzed the existing process and proposed the use of smart contacts. In the proposed system, the smart contract is used as an interface for processing existing offline documents online and recording log data onto the blockchain.
The smart contract is primarily divided into five stages, where all stages (except the first) record the address of the previous stage so that the results can be referred to in the previous management process. In the first step, we allowed the recording of commonly required personal information on the form. However, because personal information is unsuitable for recording on a blockchain that is disclosed to network participants, we recorded the personal information off-chain, and any metadata that could be referred to in the record was recorded on the blockchain. In the subsequent steps, an automated contract is generated in the section where the producer requests from the institution and the section records the production process in the existing process, and the result is recorded after satisfying the pass conditions such as pass in the previous step. Fig. 7 depicts the flow of the proposed system.
3.3 Network Design
Blockchain for the quality management system consists of a private blockchain, not a public blockchain. In the proposed system, the network is opened up and composed of permitted participants rather than free participation to speed up to block generation and restrict some privileges. In the proposed system, only authorized nodes to operate nodes in the network, and participating participants have read, write, and consensus rights on blocks by establishing a consortium of producers and two agencies. Other participants can only read the information on the blockchain. The detailed node configuration is as follows.
First, each network participant constitutes an off-chain node (in addition to the blockchain node.) When storing information using on-chain metadata, metadata that can identify only the node to share information with is stored. For this purpose, an off-chain group is selected. For example, if there are n nodes in the blockchain network, at least three of them will constitute one private group. The group will have the same off-chain so that the producer can inquire about the personal information recorded. Private data are managed with an off-chain mechanism so that the other nodes within the same blockchain network cannot be inquired about unless they are in the same group. Fig. 8 shows the blockchain network.
4. Implementation and Experiments
4.1 Implementation Environment
We established a blockchain network environment to test whether the proposed quality management system could solve problems in the existing process. The detailed implementation of the environment of this study is presented in Table 3.
Four experimental nodes were built in the same VMware-based virtual environment. We used a blockchain platform provided by Hyperledger Besu, an open-source Ethereum client, and Tessera, an off- chain feature provided by Hyperledger Besu [17,18]. Hyperledger Besu is a blockchain client that satisfies both the public and private blockchain properties. Only the relevant producers and agencies can participate in network blockchain consensus. Thus, using Besu, a private blockchain is configured, and only authorized users can reach consensus and write. “Read” permits the disclosure of quality management information recorded in the blockchain to general consumers.
Three nodes could share private data by utilizing Tessera, and one node was implemented as a general Ethereum client that did not have a separate off-chain environment. Solidity was used as the smart contract language, and a web application that could utilize smart contract was used to build a web server based on Node.js to provide a user interface that would facilitate effective quality management records and inquiries.
4.2 Registration Personal Information
First, we deployed a smart contract (by Tessera) to store personal information. Smart contracts for recording personal information (e.g., producer’s name, home address, and phone number) were not accessible by other participants. This contract and all its transaction information were stored off-chain and only accessible by the owner’s private group and the agencies.
Different result values were produced depending on whether an off-chain was used to search the recorded data using a transaction hash. For a node that did not belong to a given group, by not using an off-chain, the retrieved data were considered different, even if the same transaction hash and block hash were retrieved. This is because the blockchain on-chain records metadata for referencing the off-chain data. Thus, it is important to prevent leakage of personal information by creating and managing only previously trusted nodes as a group (as shown in Figs. 9 and 10).
If a private transaction is searched inside the blockchain without using a node or an off-chain that is not in the same group, a general transaction receipt is returned. The transaction is written for the Ethereum environment, and its contents and receipt are the same. However, this information alone cannot reveal what was recorded. Even if a third party checks its contents, they will not find personal information unless they have an off-chain that has the same configurations as the owner’s private group.
Conversely, if an off-chain that can identify metadata is configured in the transaction, information can be retrieved easily. The transaction recorded in the off-chain is returned by referring to the metadata recorded in the on-chain. The recorded data are stored in the information recorded in the off-chain, and the recorded personal information can be queried in the system.
Fig. 11 depicts a web application for recording personal information. When personal information is recorded (e.g., producer’s name, home address, and phone number), it is stored separately in the off-chain, and only institutions and personal information subjects involved in the quality management process can inquire about this information.
4.3 Application and Record
In the subsequent steps, an automated contract is generated when the production process has passed the current step to pave way for the next step. This is done to proceed with the specific legal quality management process that is used as a means of authentication. Fig. 12 shows the log in which the pass and time were recorded in the previous step.
To search the log in the previous step, the contract address must be known in the previous step. Except for the first personal information registration where the previous step does not exist, all other steps must record the address in the previous step. The abstracted code is shown in Fig. 13.
Essentially, smart contracts, except for the first step, require the admin address to be hosted in the contract and the previous contract address (previous CA) to refer to the results of the previous step. When deploying the initial contract, admin and previous CA are initialized using the constructor provided by Solidity.
Subsequently, the detailed data to be recorded in the blockchain and the functions that the smart contract should show are different, but the two things need to be recorded in common. First, when the producer succeeds in reporting or application, the relevant information should be recorded. The time of the transaction, the address of the wallet that sent the transaction, and the contract address are recorded in the blockchain before quality management. The second is the information that the management organization needs to record in the blockchain and records whether it has passed the relevant step. Similar to the previous step, the records should be executed only by the governing body.
4.4 Pass Confirmation
In the existing quality management process, when all certification processes are completed, a pass certificate is issued, thus allowing distribution and sales. As the proposed system can inquire about data recorded in the blockchain (e.g., whether the application has been passed) by referring to the address of each step, if the contract address that was passed in the final step is used, then the log data recorded in the blockchain can be used for the previous steps. Therefore, the results can be searched continuously. This function makes it more difficult to falsify accurate management records and logs during each process. Fig. 14 shows the results of the final pass inquiry.
Through the certificate, the processing results can be checked when inquiring about the link to each account address. However, the address in which personal information is stored cannot be viewed by a third party. To inquire about this information, a blockchain node with an off-chain that maintains the same group must be used. By accessing from outside the node, we secured an authentication method using the characteristics of the blockchain. The personal information subject transmits a transaction signed with the private key for self-authentication and compares it with the information in the log recorded when creating personal information. The log recorded when registering personal information is shown in Fig. 15.
If the address recorded in the log where personal information was first recorded and the address that signed the transaction were the same, authentication was complete, and the log was recorded, and personal information was inquired. Conversely, if the recorded address and the transaction address were different, it was considered that authentication failed and the corresponding personal information could not be viewed, so a failure log was recorded. The authentication success and failure logs are shown in Figs. 16 and 17, respectively.
In this paper, we propose a quality management system for wild-simulated ginseng using blockchain that satisfies legal procedures by analyzing the quality management process for wild-simulated ginseng. The system is based on private blockchain, and institutions and producers related to wild-simulated ginseng production participate in the network and record information. In addition, this study automates the existing process by introducing a smart contract to the existing offline document process and converts it online to reduce document management costs and facilitate convenient quality management procedures. This system increases the integrity of information and provides transparent information to increase the reliability of the quality of wild-simulated ginseng from consumers, and at the same time, it can protect the personal information written in the document by using off-chain technology.
In the future, we plan to actually operate the proposed system for quality management of wild-simulated ginseng to measure the performance of the system and identify additional requirements for the system. Therefore, we try to find a way to supplement the shortcomings of this paper and develop the proposed model into a practical system.
This work was supported by an Institute for Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korean government (MSIT) (No. 2019-0-01343, Training Key Talents in Industrial Convergence Security).
This study was carried out with the support of the R&D Program for Forest Science Technology (Project No. 2020184C10-2122-AA02) under the Korea Forest Service (Korea Forestry Promotion Institute).
He received his B.S. and M.S. degrees from the Department of Computational Statistics at Chungnam National University, South Korea, in 1985 and 1987, respectively. He received his Ph.D. from the Department of Computer Science Engineering at Chungnam National University, South Korea, in 1998. He worked on wireless Internet information security at the Electronics and Telecommunications Research Institute from February 1987 to February 2001 and mobile security at AhnLab from March 2001 to August 2004. His work included incident handling, and he was in charge of management planning at the Korea Internet & Security Agency from September 2004 to February 2014. He is currently a professor in the Department of Computer Science Engineering at Chungnam National University.