POS DATA TO POWER EFFICIENT DISTRIBUTION

ADAM BLAIR

While ensuring that actual physical product is handled, sorted and transported as efficiently as possible will remain a crucial part of future distribution logistics, industry experts believe the biggest gains will come as item movement and customer information begin flowing back from the retailer's point-of-sale, toward distribution centers, wholesalers and trading partners.

As the speed of data delivery approaches real time, for example, replenishment decisions can hit new levels of accuracy. And as retailers and manufacturers more proactively share their purchasing, production and promotion plans, they can help flatten the supply chain "spikes" that add cost and confusion to these processes.

In addition, if initiatives such as scan-based trading and collaborative planning, forecasting and replenishment provide retailers and manufacturers with the sales increases and lower distribution costs they have promised in pilot programs, some believe that forward-buying practices, blamed for distorting the supply chain and adding unnecessary days of supply to inventory levels, will decrease.

Speaking about CPFR, Ralph Drayer, vice president of Efficient Consumer Response at Procter & Gamble, Cincinnati, said its benefits could outweigh current product deals and allowances.

"This new way of operating will deliver significantly improved business results," said Drayer. "It's a dramatic opportunity to take inventory out of the supply chain, lower costs, improve in-stock levels and improve sales. It will also clearly reduce the 'bulge' in inventory moving through the supply chain."

Experts agree that achieving such results industrywide from CPFR and SBT depends heavily on retailers' ability to gather clean, accurate POS data. In addition, the industry will need to use quick, cost-effective vehicles to transmit the data to the appropriate trading partners, either through electronic data interchange or the Internet.

Another option could be UCCnet, an industrywide extranet operated by the Uniform Code Council, Dayton, Ohio, which will be tested later this year.

The first step, however, is bringing the quality of scan data up to an acceptable level, especially for SBT, according to Doug Adams, vice president of Prime Consulting, Bannockburn, Ill.

"Absolutely top-quality scan data is crucial," said Adams, adding that retailers who have already begun using scan data to reorder product for their stores "have probably already dealt with scan hygiene."

"Daily POS data gives everyone in the supply chain visibility of demand," said Drayer. The goal is "matching the new, improved flow of information with the physical flow" of products, he added.

Existing supply-chain initiatives, such as continuous replenishment programs and vendor-managed inventory, have already improved the supply-chain's efficiency. Drayer credits CRP with significantly lowering inventory levels.

But the manufacturer, currently participating in CPFR pilots with five retailers, believes the inventory reduction with the new initiative will be "equal to or greater than what we've seen with CRP," he said. "It could cut another 50% of inventory across the entire supply chain."

The difference is the use of POS data, Drayer added. CRP and VMI programs can only go so far, according to supply-chain experts.

"What we really need is a speed-to-shelf measurement," said Don Vehlhaber, president of the consulting firm Strategy Partners Group, Atlanta. He esplained that such a measurement could be useful with automated ordering systems that are based on POS movement of specific items. "It takes an accumulation of buyers taking green beans off the shelf to trigger the order," said Vehlhaber. "How long it takes to put that can back on the shelf is the 'speed to shelf.' "

The speed of data flow is another issue limiting the usefulness of CRP programs. "The CRP model suffers from 'information float,' " said Bob Drury, vice president of management information systems at Schnuck Markets, St. Louis.

If a product promotion increases consumer demand, for example, "Currently, the promotion is almost over before the manufacturer knows what the [consumer] pull has been."

Quicker access to data can improve results, according to Vehlhaber. "POS tracking systems are now available to look at a promotion Monday morning that broke on the previous Thursday, and to provide store-level analysis," he said.

P&G's Drayer believes the best results will occur when daily POS data is combined with historical item movement data. This combination allows partners to "much more accurately predict demand," he said.

Schnuck's POS-replenishment program, currently being used for frozen products in six supermarkets, incorporates both current and historical data, as well as a number of other measures.

Schnuck sets a store-by-store shelf-stocking target at the stockkeeping unit level, said Drury, adding that the retailer makes a weekly replenishment forecast, based on 13 weeks of historical scan data, on a by-store, by-SKU basis.

Individual product amounts are allocated by the day of the week, dependent on each individual store's sales pattern, he added, and the retailer's time to shelf is adjusted for each store's shipment and stocking schedules.

Finally, each store's shipment forecast is adjusted based on the store's scan data approximately one hour before the picking cycle begins in Schnuck's distribution center, said Drury.

"This tells us if the product is pulling through at the rate we predicted," he said. "It's just about in real time. We believe we have to get the retail side working at this level in order to involve manufacturers."