Blockchain holds great promise in supply chain management. Blockchain can greatly improve supply chains by enabling faster and more cost-efficient delivery of products, enhancing products’ traceability, improving coordination between partners, and aiding access to financing.
To better understand this opportunity, we studied seven major U.S. corporations that are leaders in supply chain management and are trying to figure out how blockchain can help solve the challenges they face. These companies—Corning, Emerson, Hayward, IBM, Mastercard, and two others that wish to remain anonymous—operate in varied industries: manufacturing, retailing, technology, and financial services. Some of them are just beginning to explore blockchain, a few are conducting pilots, and others have moved even further and are working with supply chain partners to develop applications. This article describes what we’ve learned about the state of play, the advantages that blockchain can provide, and how the use of blockchain in supply chains will differ from its use in cryptocurrencies.
A blockchain is a distributed, or decentralized, ledger—a digital system for recording transactions among multiple parties in a verifiable, tamperproof way. The ledger itself can also be programmed to trigger transactions automatically. For cryptocurrency networks that are designed to replace fiat currencies, the main function of blockchain is to enable an unlimited number of anonymous parties to transact privately and securely with one another without a central intermediary. For supply chains, it is to allow a limited number of known parties to protect their business operations against malicious actors while supporting better performance. Successful blockchain applications for supply chains will require new permissioned blockchains, new standards for representing transactions on a block, and new rules to govern the system—which are all in various stages of being developed.
The Advantages of Blockchain
To illustrate the limitations of the current world of financial-ledger entries and ERP systems, along with the potential benefits of a world of blockchain, let us describe a hypothetical scenario: a simple transaction involving a retailer that sources a product from a supplier, and a bank that provides the working capital the supplier needs to fill the order. The transaction involves information flows, inventory flows, and financial flows. Note that a given flow does not result in financial-ledger entries at all three parties involved. And state-of-the-art ERP systems, manual audits, and inspections can’t reliably connect the three flows, which makes it hard to eliminate execution errors, improve decision-making, and resolve supply chain conflicts.
Execution errors—such as mistakes in inventory data, missing shipments, and duplicate payments—are often impossible to detect in real time. Even when a problem is discovered after the fact, it is difficult and expensive to pinpoint its source or fix it by tracing the sequence of activities recorded in available ledger entries and documents. Although ERP systems capture all types of flows, it can be tough to assess which journal entries (accounts receivable, payments, credits for returns, and so on) correspond to which inventory transaction. This is especially true for companies engaged in thousands of transactions each day across a large network of supply chain partners and products.
Making matters worse, supply chain activities are often extremely complicated—far more so than the exhibit depicts. For example, orders, shipments, and payments may not sync up neatly, because an order may be split into several shipments and corresponding invoices, or multiple orders may be combined into a single shipment.
One common approach to improving supply chain execution is to verify transactions through audits. Auditing is necessary for ensuring compliance with contracts, but it’s of limited help in improving decision-making to address operational deficiencies. Consider the problem a food company faces when its products reach the end of their shelf life in a retail store. A study that one of us (Vishal) worked on with a major manufacturer of packaged foods found that an audit or an inspection of inventory in a store can reveal the number of expired items, but it won’t explain the causes. Those can include glitches in any part of the supply chain, such as inefficient inventory management upstream, suboptimal allocation of products to stores, weak or sporadic demand, and inadequate shelf rotation (failure to put older products in front of newer ones). A record of all those activities can help reduce expirations.
Another way to strengthen supply chain operations would be to mark inventory with either RFID tags or electronic product codes that adhere to GS1 standards (globally accepted rules for handling supply chain data) and to then integrate a company’s ERP systems with those of its suppliers to construct a complete record of transactions. This would eliminate execution errors and improve traceability. However, the experiences of the companies we studied showed that integrating ERP systems is expensive and time-consuming. Large organizations may have more than 100 legacy ERP systems—a result of organizational changes, mergers, and acquisitions over time. Those systems often do not easily communicate with one another and may even differ in how they define data fields. One large company told us it had 17 ledgers in separate ERP systems associated with a single activity—trucking—and its suppliers and distributors had their own ledgers and ERP systems.
When blockchain record keeping is used, assets such as units of inventory, orders, loans, and bills of lading are given unique identifiers, which serve as digital tokens (similar to bitcoins). Additionally, participants in the blockchain are given unique identifiers, or digital signatures, which they use to sign the blocks they add to the blockchain. Every step of the transaction is then recorded on the blockchain as a transfer of the corresponding token from one participant to another.
Consider how the transaction in our example looks when represented on a shared blockchain. First, the retailer generates an order and sends it to the supplier. At this point, since no exchange of goods or services has taken place, there would be no entries in a financial ledger. However, with blockchain, the retailer records the digital token for the order. The supplier then logs in the order and confirms to the retailer that the order has been received—an action that again gets recorded on the blockchain but would not generate an entry in a financial ledger. Next the supplier requests a working-capital loan from the bank to finance the production of the goods. The bank verifies the order on the shared blockchain, approves the loan, and records the loan’s digital token on the same blockchain. And so on.
A blockchain is valuable partly because it comprises a chronological string of blocks integrating all three types of flows in the transaction and captures details that aren’t recorded in a financial-ledger system. Moreover, each block is encrypted and distributed to all participants, who maintain their own copies of the blockchain. Thanks to these features, the blockchain provides a complete, trustworthy, and tamperproof audit trail of the three categories of activities in the supply chain.
Blockchain thus greatly reduces, if not eliminates, the kind of execution, traceability, and coordination problems that we’ve discussed. Since participants have their own individual copies of the blockchain, each party can review the status of a transaction, identify errors, and hold counterparties responsible for their actions. No participant can overwrite past data because doing so would entail having to rewrite all subsequent blocks on all shared copies of the blockchain.
The bank in our example can also use the blockchain to improve supply chain financing. It can make better lending decisions because by viewing the blockchain, it can verify the transactions between the supplier and the retailer without having to conduct physical audits and financial reviews, which are tedious and error-prone processes. And including lending records in the blockchain, along with data about invoicing, payments, and the physical movement of goods, can make transactions more cost-effective, easier to audit, and less risky for all participants.
Furthermore, many of these functions can be automated through smart contracts, in which lines of computer code use data from the blockchain to verify when contractual obligations have been met and payments can be issued. Smart contracts can be programmed to assess the status of a transaction and automatically take actions such as releasing a payment, recording ledger entries, and flagging exceptions in need of manual intervention.
It’s important to note that a blockchain would not replace the broad range of transaction-processing, accounting, and management-control functions performed by ERP systems, such as invoicing, payment, and reporting. Indeed, the encrypted linked list or chainlike data structure of a blockchain is not suited for fast storage and retrieval—or even efficient storage. Instead, the blockchain would interface with legacy systems across participating firms. Each firm would generate blocks of transactions from its internal ERP system and add them to the blockchain. This would make it easy to integrate various flows of transactions across firms.
Let’s now take an in-depth look at how the companies we studied are applying blockchain to tackle needs that current technologies and methods can’t address.
The U.S. Drug Supply Chain Security Act of 2013 requires pharmaceutical companies to identify and trace prescription drugs to protect consumers from counterfeit, stolen, or harmful products. Driven by that mandate, a large pharmaceutical company in our study is collaborating with its supply chain partners to use blockchain for this purpose. Drug inventory is tagged with electronic product codes that adhere to GS1 standards. As each unit of inventory flows from one firm to another, its tag is scanned and recorded on the blockchain, creating a history of each item all the way through the supply chain—from its source to the end consumer. Some early success in piloting this approach in the United States has led the company to conduct more pilots in other locations and to move toward broad implementation in Europe. Meanwhile, IBM is working on a similar effort to create a safer food supply chain. It has founded the IBM Food Trust and entered into a partnership with Walmart to use blockchain for tracing fresh produce and other food products.
These kinds of applications require minimal sharing of information: Purchase orders, invoices, and payments do not need to be included on the same blockchain. As a result, companies that are wary of sharing competitive data are more willing to participate on the platform.
The benefits are clear. If a company discovers a faulty product, the blockchain enables the firm and its supply chain partners to trace the product, identify all suppliers involved with it, identify production and shipment batches associated with it, and efficiently recall it. If a product is perishable (as fresh produce and certain drugs are), the blockchain lets participating companies monitor quality automatically: A refrigerated container equipped with an internet of things (IoT) device to monitor the temperature can record any unsafe fluctuations on the blockchain. And if there are concerns about the authenticity of a product that a retailer returns, the blockchain can allay them, because counterfeit goods would lack a verification history on the blockchain. (We’ll talk later about how companies are trying to prevent corrupt actors from introducing counterfeit goods into both supply chains and their blockchains.) Companies across industries are therefore exploring this application of blockchain—motivated either by regulations requiring them to demonstrate the provenance of their products or by downstream customers seeking the capability to trace component inventory.
Increasing efficiency and speed and reducing disruptions.
Emerson, a multinational manufacturing and engineering company, has a complex supply chain. It involves thousands of components across many suppliers, customers, and locations. Michael Train, the president of Emerson, told us that such supply chains often have to contend with long, unpredictable lead times and lack of visibility. As a result, a small delay or disruption in any part of the supply chain can lead to excess inventory and stock-outs in other parts. He believes that blockchain could help overcome these challenges.
Here’s a simple illustration of the problem and how blockchain could address it. Consider product A, which uses components C1 and C2, and product B, which uses components C1 and C3. If the manufacture of product B is held up because of a disruption in the production of component C3, the optimal move is to temporarily allocate inventory of C1 to product A until the disruption is resolved. However, if all products and components are manufactured by different companies with limited visibility into one another’s inventory, what could easily happen is that excess inventory of C1 piles up at the company making product B even if the maker of product A has a stock-out of C1.
One solution is for the companies in question to agree to centralize their data on production and inventory-allocation decisions in a common repository. But imagine the level of integration that would entail: All involved companies would have to trust the others with their data and accept centralized decisions, regardless of whether they are partners or competitors. That’s not realistic.
A more practical solution is for participating companies to share their inventory flows on a blockchain and allow each company to make its own decisions, using common, complete information. Companies would utilize a kanban system to place orders with one another and manage production. Kanban cards would be assigned to the produced items, and the blockchain would record digital tokens representing the kanban cards. This would enhance the visibility of inventory flows across companies and make lead times more predictable.
Emerson is not the only company that thinks blockchain could increase the efficiency and speed of its supply chain. So does Hayward, a multinational manufacturer of swimming pool equipment. (Disclosure: Vishal has done a small amount of consulting for Hayward. He has also been hired to advise a start-up that’s developing blockchain applications for the palm oil industry.) According to Don Smith, Hayward’s senior vice president of operations, it is possible to treat finished goods, process capacity, work-in-process inventory, and raw materials like digital currency. If you do, he says, machine time and inventory at various stages can be reliably assigned to customer orders. Blockchain makes this possible by solving the double-spend problem—the erroneous allocation of the same unit of capacity or inventory to two different orders.
Walmart Canada has already begun using blockchain with the trucking companies that transport its inventory. A shared blockchain makes it possible to synchronize logistics data, track shipments, and automate payments without requiring significant changes to the trucking firms’ internal processes or information technology systems.
Part of the appeal of using blockchain to enhance supply chain efficiency and speed is that these applications, much like those for improving traceability, require participating companies to share only limited data—in this case, just inventory or shipment data. Moreover, these applications are useful even within large organizations with multiple ERP systems.
Improving financing, contracting, and international transactions.
When inventory, information, and financial flows are shared among firms through a blockchain, significant gains in supply chain financing, contracting, and doing business internationally are possible.
Consider the matter of financing. Banks that provide working capital and trade credit to firms face a well-known problem of information asymmetry regarding a borrower firm’s business, the quality of its assets, and its liabilities. For example, a company might borrow money from several banks against the same asset, or request a loan for one purpose and then use it for another. Banks design their processes to control such risks, which increases transaction costs, slows down access to capital, and reduces the capital available to small firms. Such frictions are detrimental not only to banks but also to firms that need cheap working capital.
Another activity ripe for improvement is accounts payable management, an elaborate process that involves invoicing, reconciling invoices against purchase orders, keeping track of terms and payments, and conducting reviews and approvals at each step. Even though ERP systems have automated many of these steps, considerable manual intervention is still needed. And since neither of the transacting firms has complete information, conflicts often arise.
A third area of opportunity is cross-border trade, which involves manual processes, physical documents, many intermediaries, and multiple checks and verifications at ports of entry and exit. Transactions are slow, costly, and plagued by low visibility into the status of shipments.
The retailing and financial services companies we studied are conducting pilot blockchain projects or developing platforms in all three areas. By connecting inventory, information, and financial flows and sharing them with all transacting parties, a blockchain enables companies to reconcile purchase orders, invoices, and payments much more easily and to track the progress of a transaction with counterparties. When the supplier receives an order, a bank with access to the blockchain can immediately provide the supplier with working capital, and when merchandise is delivered to the buyer, the bank can promptly obtain payments. Since there is a readily available audit trail and reconciliations can be automated, using smart applications that rely on the blockchain data, conflicts between the bank and the borrowing firm are eliminated.
Creating a Workable Technology
The companies we studied have found that using blockchain in supply chain management will require the creation of new rules, because the needs of supply chains differ from those of cryptocurrency networks in important ways. The blockchain protocol for the Bitcoin network is a marvelous system that simultaneously achieves several goals. It provides a remarkably secure, irrevocable record of financial transactions, minimizes the double-spend problem, and provides proof of ownership of a digital coin. And it does so without relying on a centralized authority and while allowing participants to remain anonymous and enter and exit the network freely. To achieve all this, however, the Bitcoin network sacrifices speed, consumes a large amount of energy to mine bitcoins, and has some vulnerability to hacking.
Supply chains do not need to make the same trade-offs because they operate in a different way and have different characteristics. Let’s examine those in depth.
Supply chains require private blockchains among known parties, not open blockchains among anonymous users. So that members of a supply chain can ascertain the source and quality of their inventory, each unit of it must be firmly coupled with the identity of its particular owner at every step along the way. Consequently, only known parties can be allowed to participate in such a blockchain, which means that companies must receive permission to join the system.
Moreover, permission must be granted selectively. That’s because the open and decentralized structure of blockchain poses a risk to data privacy. When companies post transactions on a blockchain, that data can be accessed by any participant. As the volume of data swells, it could potentially be misused to gather competitive intelligence, trade stocks, or predict market movements. For security reasons, therefore, the blockchain participants need to be vetted and approved.
Building a trusted group of partners with which to share data on a blockchain will entail overcoming several challenges. One is the need for a governance mechanism to determine the rules of the system, such as who can be invited to join the network, what data is shared, how it is encrypted, who has access, how disputes will be resolved, and what the scope is for the use of IoT and smart contracts. Another challenge is figuring out how to address the impact that blockchain could have on pricing and inventory-allocation decisions by making information about the quantity or age of products in the supply chain more transparent. It’s hard to predict where in the supply chain the costs and benefits of this transparency will fall.
For these reasons, the companies that we studied were focusing on narrow applications such as the traceability of drugs and food products and the management of accounts payable—applications that are supported by well-defined use cases or regulatory requirements. Firms limit the types of information recorded on the blockchain to reduce the risk to data privacy and make the system more readily acceptable to supply chain partners.
Simpler consensus protocols.
Blockchain requires a consensus protocol—some mechanism for maintaining a single version of the history of transactions that is agreed to by everyone. Since cryptocurrency networks are peer-to-peer without a central authority, they use a complex method called proof of work. It ensures that all transactions on the network are accepted by the majority of participants, but unfortunately, it also limits the speed at which new blocks can be added. Consequently, it is too slow to handle the speed and volume of transactions in supply chains.
Consider the pharmaceutical industry, where 4 billion salable units enter the drug supply chain every year in the United States. Each unit is handled three to five times, on average. That translates to roughly 33 million to 55 million transactions a day, on average. The Bitcoin network, in contrast, allows only about 360,000 transactions a day.
Fortunately, if a blockchain is permissioned and private, the proof-of-work method is not necessary to establish consensus. Simpler methods can be used to determine who has the right to add the next block to the blockchain. One such method is a round-robin protocol, where the right to add a block rotates among the participants in a fixed order. Since all participants are known, a malicious actor would be discovered if it used its turn to modify the chain in a harmful or illegitimate way. And disputes can be resolved easily by participants’ validating previous blocks.
Security of physical assets.
Even when a blockchain record is secure, there is still the danger that a contaminated or counterfeit product might be tagged and introduced into the supply chain, either in error or by a corrupt actor. Another danger is inaccurate inventory data resulting from mistakes in scanning, tagging, and data entry.
Companies are addressing these risks in three ways. First, they are stringently conducting physical audits when products first enter the supply chain to ensure that shipments match blockchain records. Second, they are building distributed applications, called dApps, that track products throughout the supply chain, check data integrity, and communicate with the blockchain to prevent errors and deception. If a counterfeit or an error is detected, it can be traced to its source using the blockchain trail of the transactions for that asset. Third, companies are making the blockchain more robust by using IoT devices and sensors to automatically scan products and add records to the blockchain without human intervention.
One area where tokenization is sufficient to provide trust and security is the trading of assets like digital books and music. If the ownership of these assets is tied to a blockchain platform, counterfeits can be completely eliminated. For instance, universities commonly use digital reading packets for many courses, working in partnership with publishers and copyright owners. Significant efficiency gains could be generated by knitting this digital supply chain into a blockchain platform with smart contracts that can help participants access products, verify ownership, and handle payment.
There is considerable room to improve supply chains in terms of end-to-end traceability, speed of product delivery, coordination, and financing. Blockchain can be a powerful tool for addressing the deficiencies, as the companies we studied have proved. It is now time for supply chain managers who are standing on the sidelines to assess the potential of blockchain for their businesses. They need to join the efforts to develop new rules, experiment with different technologies, conduct pilots with various blockchain platforms, and build an ecosystem with other firms. Yes, this will require a commitment of resources, but the investment promises to generate a handsome return.