The need
for precise and efficient inventory allocation is a no-brainer.
Retailers and manufacturers alike have an obvious need to move the
right amount of goods to the right place at the right time.
The costs of poor inventory allocation are impossible to ignore
but difficult to quantify. Retailers that are out of stock risk
losing customers, but if retailers are overstocked, they might have
to discount goods, and their employees will need to manage inventory
when they should be serving customers.
Supporting this simple business need is tremendously complex. The
amount of historical sales data quickly ranges into terabytes at any
large company. On top of the huge data warehouses for demand
forecasting, analysts say, companies also need to implement and
eventually integrate systems — usually packaged ones — to manage
their product and promotion planning, distribution center operations
and transportation networks.
Big Rewards Loom
The prize for allocating inventory optimally is huge, according
to analyst Greg Girard at AMR Research Inc. in Boston. "That's been
one of the underpinnings of Wal-Mart's success," he says.
There are many examples of companies using information technology
to master their inventory. The Coca-Cola Co. in Atlanta offers to
help retailers analyze sales data to promote customer traffic and
plan space usage in stores (Business, June 21). Minneapolis-based
Cargill Inc. wants to use Web-based systems that facilitate
inventory data sharing to get better demand forecasts from customers
for its oils and syrups (News, Oct. 25). Longs Drug Stores Corp. in
Walnut Creek, Calif., uses a forecasting and replenishment system to
allocate high-margin inventory at its prescription drug warehouse
(Business, Oct. 4).
The Limited Inc., in Columbus, Ohio, a $9.3 billion retailer that
operates such store chains as Victoria's Secret, Express and Lane
Bryant, has implemented data warehouses in six of its 11 business
units and is upgrading its warehouse management systems to bring out
the flexibility inherent in its geographically decentralized
distribution centers, says Tom McFadden, group CIO for supply-chain
management. The company is also upgrading systems to improve
planning and transportation management, he says.
Guesswork Still Plays Role
But for all the technology that companies are installing,
guesswork and instinct are still parts of the process.
"It's a science and an art both," McFadden says. "The bottom-line
IT challenge, in my opinion, is how to replenish (merchandise) at
the (stock-keeping unit (SKU)) level. That's an extremely large
amount of data." The Limited manages supply for its nine retail
operations as well as for two it recently spun off. Each unit has
hundreds of stores, and each store has thousands of SKUs. The sales
each year in just one chain of stores can build up to a billion rows
in a bulging data warehouse, and there's no adequate forecasting
tool that can work with that level of detail, McFadden says.
Forecasting and replenishment systems must be detailed and be
able to quickly present the right information to the right people —
including suppliers. Shortening the lead time for replenishing
inventory can put hot items on the racks faster and lessen the
amount of "safety stock" a company needs to carry to avoid running
out, says Stephen A. Smith, a professor at Santa Clara University's
Retail Management Institute in California. But faster transportation
is usually more expensive, and it can be risky for retailers to ask
suppliers to directly supply merchandise to stores because suppliers
may be unwilling to break shipments into the exact quantities
needed. Detailed systems can help manage that, too.
"There's a real theme here about increasing granularity," Girard
says. Increasing granularity, or being able to show more fine
detail, also applies to assessing inventory costs. Companies also
need systems that can tell them the real "landed" cost of each item
in their inventory — the total cost including import fees, storage,
transportation and other costs, Girard says. Only when the costs are
fully understood can a business realize which items contribute most
to their margins and should, therefore, be a supply-chain priority.
Finally, retailers and other businesses need to keep their
various planning arms in synchronicity. Financial planners must talk
with store planners and product planners to ensure that inventory
moves in a way that allows each plan to be realized rather than
inadvertently thwarted, Girard says.
Although it's technically difficult, optimizing inventory
allocation is also fraught with new opportunities for businesses,
Smith says. Retailers, for instance, can figure the Web into
inventory allocation plans by using e-commerce sites to sell
uncommon sizes that might take up precious space in stores, he says.
Keeping the Stores Stocked
Replenishment is a small part of the process of allocating and
managing inventory, but it's a crucial one. It ultimately ensures
that stores are stocked with the proper quantities of products and
in the right time frame.
How products are replenished depends on many factors, such as
whether the supplier is willing to deliver exact quantities directly
to stores or just in bulk to distribution centers, according to
Stephen A. Smith, a professor at Santa Clara University's Retail
Management Institute in California. But he says the process in a
common retail setting such as a department store goes something like
the following:
1.Every night, a store computer calculates — based on the day's
sales — how much of each product is in stock. The computer then
compares the newly computed stock levels with a model of what the
inventory should be.
2. If the stock on hand is lower than it should be, the computer
sends an order, possibly in electronic data interchange format, to
the supplier or distribution center.
Replenishment can become an inventory allocation issue while
maintaining the model inventory. Retailers will refresh what
inventory is allocated in a model at least every season, but more
likely every time there's a sale or other promotion.
Sites 
