The term supply chain management has been around for more than 30 years, but according to Rob Barrett, managing director, supply chain & operations with consulting firm KPMG, its days are numbered. "The supply chain as we know it will be dead within the next five years," Barrett predicted at the recent IndustryWeek Best Plants 2013 conference, held in Greenville, S.C. Now, before you jump to conclusions and assume he was only saying that because he was hoping for a new job title on his business card, here's what he said after that provocative statement: "What we'll be talking about soon won't be supply chains but supply networks."
Actually, people have been talking about supply networks for years, since that's what the 21st Century definition of a supply chain is. "Supply chain" has caught on and is pretty much engraved in stone as the most precise and concise way to describe the connection (or network) between departments and organizations, focusing on the central tasks of plan, source, make, deliver and return. The distinction Barrett ultimately was making isn't really between "chains" and "networks," but rather between sequential forecasting and demand-driven forecasting.
Richard Dean, chief marketing officer with One Network Enterprises and Barrett's co-presenter at the Best Plants show, pointed to Del Monte Foods, a $3.7 billion processed food manufacturer, as an example of a company that's adopted demand-driven inventory planning and replenishment practices throughout its extended supply chain. Del Monte's retail-centric supply chain of 3,500 SKUs includes more than 1,300 growers, 41 manufacturing and co-manufacturing locations in the U.S., 10 strategically located distribution centers and more than 5,500 ship-to-locations.
Del Monte's goal was to improve its level of customer service. The way it chose to get there was to use downstream data and store-level demand to significantly reduce the amount of inventory its retail customers needed to carry to prevent stock-outs. That meant Del Monte had to shift from push to pull processes, which resulted in an improvement in forecast accuracy by more than 400 basis points. In fact, eight of its retail customers have reduced their inventory levels to the point they have achieved "negative working capital."
Del Monte's work illustrates the current capabilities of supply chain management, but Bill Ferrell offered Best Plants attendees a glimpse at the potential future. Ferrell, a professor of industrial engineering at Clemson University, began by stating the obvious: "Today's logistics are inefficient and unsustainable." An average trailer load, he reports, is only 42.6% full, and 25% of the time they're completely empty. Warehouses and DCs are similarly underutilized. He points to U.S. Government statistics that project a $66 billion opportunity if we can come up with a way to get to 100% full.
One of those ways, he suggests, is the Physical Internet, "an open global logistics system, leveraging interconnected supply networks through a standard set of modular containers, collaborative protocols and smart interfaces for increased efficiency and sustainability." He compares it to an eBay-like freight transportation auction that utilizes a set number of "black box" modular containers, with the goal of standardizing container sizes while optimizing logistics networks.
Most of the work on the Physical Internet to date has been on the research side, but Ferrell offers a case study example of a retailer moving 1,715 products. By going to modular containers, the retailer was able to ship an additional 8% of volume at the container level, with a 20% savings at the pallet level. A load planning case study revealed similarly impressive numbers. Using such strategies as co-loading, continuous moves, back hauls and relay networks, cost-per-load decreased by 29%, while average fullness increased 34%. You can learn more about the Physical Internet, an open organization, at www.physicalinternetinitiative.org.
Ferrell also offered early results of a project that is attempting nothing less than an accurate way to predict the future. Using neural networks (a type of artificial intelligence modeled on the way the human brain works) and logistic regression techniques, researchers are studying ways to predict exactly when shipments will arrive, using supply chain data, with the goal of getting closer to the nirvana of the Perfect Order. Admitting to some early glitches in the interpretation of the results, Ferrell nonetheless believes these techniques show promise of at least improving forecast accuracy, if not necessarily actually getting it right every time.
In any event, no matter how sophisticated the processes and how futuristic the technologies, I think I'm safe in predicting that the term "supply chain" will be with us for a long, long time to come.