As the New York Times tells it, supply chains are changing (New Hubs Arise to Serve ‘Just in Case’ Distribution, Feb 12).
Major storms like Hurricane Sandy and other unexpected events have prompted some companies to modify the popular just-in-time style of doing business, in which only small amounts of inventory are kept on hand, to fashion what is known as just-in-case management. …
Just-in-case is a response to the vulnerability of just-in-time supply chains, said Rene Circ, CoStar’s director of industrial research. Since the 1990s, just-in-time has made sense for many companies looking to reduce the cost of keeping large inventories on hand. Technology enabled retailers and manufacturers to closely track and ship items to replace merchandise sold or components consumed in production.
This model also reduced transportation costs, because goods would be shipped only as necessary. By combining the just-in-case with just-in-time strategy, Mr. Circ said, companies are trying to strike a balance between “carrying the minimum inventory possible, yet never running out of things, because inventory equals cost.”
I’ve been trying to think what I should say about this article for several weeks. I have felt conflicted because, on the one hand, it hits on some interesting points. On the other hand, it also leads with one of my pet peeves of business reporting. Specifically, it links any change in inventory management to some failure of just-in-time management. However, I am not convinced that is actually a good description of what is going on here.
Consider the following example from the article.
For example, Ranger Steel, a company based in Houston that describes itself as the largest privately owned steel plate distributor in the nation, recently expanded its network of distribution centers. Until the late 1990s, Ranger Steel regularly trucked its heavy steel plates directly from a distribution center at the Port of Houston to customers throughout the United States.
“For a long time that concept worked like a charm,” said Jochen Seeba, the company’s vice president and chief operating officer. “Then you started to see the spike in fuel pricing, and new governmental rules and regulations on insurance coverage for truck drivers that made truck transportation very expensive.”
Like many retailers and wholesale suppliers, Ranger Steel in recent years has added distribution centers to its network, cutting delivery times by moving its inventory closer to customers.
Today, suppliers ship steel in bulk to Ranger Steel’s seven distribution centers around the country. From there, drivers can deliver orders to most of the company’s customers in 24 hours, leveling the playing field with local competitors in markets like St. Louis, where the firm opened a distribution center last year.
So where’s the failure of just-in-time inventory? Isn’t this simply a response to changing economic and industry conditions? If fuel and drivers become more dear, it makes sense to decentralize distribution so more product can be shipped by cheaper means (e.g., train) without delaying service to customers. If anything, having Ranger Steel hold more product locally allows more of its customers to operate in a just-in-time fashion since they can get product quicker.
The article also goes on to discuss retail supply chains who are under pressure to offer faster and faster service.
In an effort to gain an advantage, online retailers are increasingly offering rapid shipping from nearby distribution centers, Mr. Circ said. “It’s driving this reconfiguration of supply chains and the building of these large warehouses, not just in a few key markets as it used to be, but a lot more widely spread” across the country, he said.
Again, a failure of just-in-time management seems less the issue than shifting customer expectations. But that raises the question of what retail supply chains will look like going forward. That is the topic of a recent article on the McKinsey Operations Extranet (The future of retail supply chains, Feb 21). Here is there diagnosis of the issues facing retailers.
First, most retailers’ networks do not include enough distribution centers (DCs) to cover individual customer orders cost-effectively for a large geography. Exhibit 2 highlights this challenge; with two to three optimally located DCs, a retailer can reach almost all of its customers with two-day shipping. But moving to one-day shipping would require a huge increase in the number of DCs. Retailers without enough DCs will either have to bear the burden of priority shipping costs or redesign their networks to be closer to customers.
Second, today’s retail supply chains are optimized for stores, with online often treated as a bolt-on, separate business. This legacy leads to poor cross-channel coordination across channel-specific inventory pools and fulfillment processes, causing higher out-of-stocks and markdowns in any given channel, especially during peak seasons.
Third, the huge proliferation of SKUs to fulfill the “endless assortment” promise of an online channel is creating new capacity and cost challenges, including expenses for picking additional online orders, acquiring additional DC space, and processing peak-season demand.
That analysis leads to the following estimate of Amazon’s advantage over — well over pretty much anybody whose offerings overlap with Amazon.
What is the McKinsey recommendation that goes with this? Not surprisingly it starts with building out a broader distribution network. If not with full-blown DCs, then developing the capability to ship individual orders from stores. (As we have written about before, both Macy’s and Nordstrom do this.) That, of course, means that IT systems need to be upgraded so that there is a common view of inventory across the firm. Again, Nordstrom is leader on this measure. Note that assuring that there is complete and up-to-date information about what the firm has somewhere may, in fact, lower the inventory the firm needs to carry. If store inventory can back up distribution centers, one can start the season with less inventory without jeopardizing sales.
A final point that the article makes is an interesting research area. If a firm has a handful of DCs and hundreds of stores, from where should an item ship when an order comes in? There are likely to be dozens and dozens of options. Some are clearly brain-dead (shipping to New Jersey from the West Coast when there are is plenty of inventory in the Eastern time zone) but many are likely to be viable. A good decision will necessitate considering both order fulfillment costs as well as how much inventory is sitting where. Some work has been done on this (see here) but it is clearly an open area.