Blockchain In Supply Chain — How To Use The Distributed Ledger To Trace Products
Solving Real-World Problems
I recently attended a blockchain conference hosted by the McCombs School of Business at the University of Texas at Austin. Among the speakers were Goldman Sachs, Walmart, USAA, Merck and others. Each of the speakers spoke about the potential implication of blockchains on their businesses.
For instance, Frank Yiannas, the vice president of food safety at Walmart, addressed the difficulty of tracing a product to its source. Remember the E. coli outbreak in 2006 spread by spinach? According to the CDC, during the outbreak, 199 people were infected, 102 were hospitalized and three died. For about two weeks, retailers stopped selling spinach all over the United States until the farm that caused the outbreak was identified. For two weeks there was a danger of people getting sick. There was also a loss of revenue for all retailers selling spinach and for all the farmers that grew it.
The reason it took so long to find the source stemmed from the complicated and often paper-based supply chain process leveraged by a network of growers, wholesalers, distributors and retailers. Many of these entities still use physical documents when providing information to their customers or vendors. Even when this is digitized, the data often only makes it to the next downstream customer and may not be passed further. Having to trace the product often translates to asking vendors for the info, who will, in turn, ask their vendors and so on. Multiply that across the 50,000 products Walmart carries, 6,000 stores and thousands of suppliers and suddenly tracing the origin of a product becomes a nightmare. The result is that even companies like Walmart who have one of the most sophisticated supply chain management systems in the world take around six days to track the origin of a product. Imagine if a product is recalled. That’s six days of revenue loss and six days for potentially bad reputations to take hold for farmers.
This is where the proponents of blockchain say the technology could shine. According to Yiannas, when Walmart tested IBM’s implementation of the blockchain to trace mangos, it took them 2.2 seconds to find the source. That’s down from six days. Critics could say that if the entire solution was centralized, 2.2 seconds is actually slow. Those critics are right.
The problem is that a centralized approach simply would not work in the real world. Imagine Walmart asking every farm and every supplier to integrate with its system. Data would be centralized in Walmart’s database and provenance lookup would be simple for them. Now, imagine every other retailer asking their supply chain partners to do the same thing. All of a sudden, every farm in the world would have to integrate with every supplier and every retailer that touches their product, which is impossible.
Alternatively, supply chain members could simply agree on a standard protocol to pass data to each other. Growers could use the GS1 standard to send data to their wholesalers, wholesalers would do the same for their distributors and so on. That could work. In fact, the pharmaceutical and other industries have done this for years using EDI and GS1. However, the resulting system is often a patchwork of different communication mechanisms, permissions and technologies. Further, when integration between partners breaks, so does the ability of other partners to receive that data. The entire supply chain is dependent on one system operating properly. It’s a mess. Instead, as it was in Walmart’s case, partners could leverage the distributed ledger technology (the blockchain) together with the GS1 standard to create a permissioned, fully distributed network of peers to share data in a standard, secure, transparent and robust way.
Here Be Dragons
While companies can often make a case for improving product traceability using the blockchain, this technology will not magically solve all problems. For instance, leveraging it does not mean that everyone will enter the right data. Depending on the level of automation, there may still be humans involved in data entry. On the other hand, the data can be corrected and there is always a record of it in the immutable ledger. Because of that, it’s also harder to cheat.
Depending on how a blockchain network is set up, there is also a danger of accidental collusion. If competitors are leveraging the same blockchain in a consortium environment, they need to be careful that certain data is not shared. Any plans for the future, production plans or price shared among competitors would be considered collusion by the Department of Justice. Additionally, most blockchain implementations are relatively new. Partners will still have to make changes to their systems to adopt it and there may be some anxiety over sharing data. What companies implementing blockchains should do is confront the brutal facts and leverage the technology only when it makes business sense. Companies should start with the biggest opportunity. They should also realize that partners will not want to change their business processes, so any new system would have to work with their current systems. Lastly, implementers should make sure everyone on the blockchain network benefits. For example, growers in Walmart’s case were happy to join to ensure that, if food safety issues occur, their name could be easily cleared.