Forecast accuracy, overstocks, and expediting comprise widespread inventory management pain points

Given changing market forces, manufacturers and distributors apparently need a new inventory-management education, according to a new survey conducted by Cummaquid, Mass.-based Industry Directions. “They're stuck in the 1970s in managing inventories,” says Julie Fraser, principal analyst.

By Staff September 1, 2007

Given changing market forces, manufacturers and distributors apparently need a new inventory-management education, according to a new survey conducted by Cummaquid, Mass.-based Industry Directions .

“They’re stuck in the 1970s in managing inventories,” says Julie Fraser, principal analyst. “They’re not using timely tools. They continue to struggle with forecast accuracy, overstocks, and expediting, which represent the more costly areas where a mismatch between plan and reality can come back to bite you. As market cycles accelerate, supply chains lengthen and go global, and products proliferate, companies must employ practices and systems that can keep pace.”

Fraser says conclusions were drawn from 190 respondents—48 percent manufacturers, 38 percent distributors, and 14 percent retailers. Less than 40 percent have forecast accuracy above 80 percent, even for near-term, 90-day horizons. And an overwhelming majority—83 percent—say overstocks are common; with another large majority—73 percent—indicating that expediting is common.

A key problem, Fraser explains, is companies aren’t forecasting with enough granularity, typically continuing to use classes or families of products.

“Much of it comes down to the need to forecast at the SKU level,” she says. “From a business perspective, it’s more critical to forecast not only at the SKU level, but for individual distribution centers as well.”

The trick is correctly balancing service levels with stocking levels. “It’s always a trade-off—that is, whether you have the right amount of inventory to hit the service levels you need to maintain, and having the right level for each product at each location across a multi-staged supply chain,” says Fraser.

Demand variability caused by forecast error and inherent market variability impacts service, as do lengthier supply chains when there is disruption in the flow.

“Manufacturers and distributors must push the envelope further than they’ve ever had to before,” says Jeff Bodenstab, VP of marketing for ToolsGroup , a supplier of optimization software for demand-driven environments, and one of the sponsors of the Industry Directions research. “Managing demand-driven supply chains [should result in] higher customer service levels. That’s the challenge: How do I manage demand variability and still deliver service levels at 98 percent?”

Bodenstab says ToolsGroup focuses on both issues. “We offer tools that enhance the forecast and drive to a more granular level of detail, using advanced consumption logic applied to individual SKUs,” he says. “We also have inventory optimization software that helps define the optimal safety stock across the supply chain.”

Fraser contends that while companies are actively pursuing innovation in their product lines, the business processes in the supply chain haven’t caught up. “It’s the tail end of the innovation wave,” she says, “where people are only just realizing they need to innovate there as well.”

The saving grace is that there are supply chain solutions that can bring businesses processes back up to the level of the product innovation curve.

“Demand variability becomes extreme when you’re talking intermediate demand of individual parts in distribution operations,” says Charles Smart, president of Smart Software , a vendor of forecasting and inventory optimization solutions. “An example would be the automotive aftermarket, where there are tens of thousands of SKUs, and demand may be zero for certain parts, but then spikes and drops again.

“We help customers hit the ‘inventory sweet spot’—the minimum amount of inventory at the SKU level to meet demand over a given lead time at a specific service level,” Smart continues. “Most people start with inventory levels that are woefully unbalanced: too much of one item and too little of another. We have a patent on intermittent demand forecasting. With intermittent demand, we can usually guarantee a 10-percent reduction in dollar value overall in the first year alone.”