IBM says its new Dynamic Inventory Optimization Solution helps retailers achieve the delicate balance between demand and inventory levels by mining data for customer order patterns. Inventory levels have been slashed by 40 percent in some cases, according to IBM.
The developer says the solution uses data sources once thought to be too detailed to quickly and properly parse efficiently. In the retail industry for example, forecasts using point-of-sale (POS) data are traditionally prepared weekly. However, IBM says advanced algorithms help analyze daily POS and other vital data to project stock overages and shortages. The solution then evaluates retailer and vendor sourcing rules, and suggests orders and replenishments to help maintain the optimal balance of stock and service at the store level.
IBM says it recently deployed the Dynamic Inventory Optimization Solution for Retail at Max Bahr, the Germany-based do-it-yourself retailer. Previously, the company had relied on local planners from each of its 90 stores to manually forecast inventory needs for over 70,000 items.
Working with Max Bahr, IBM Global Business Services implemented a system which automatically generates the retailer's order proposals and provides improved forecasts for its network of retail outlets. Each evening the solution takes the roughly 15 to 20 million POS transactions from all 90 stores and analyzes them against a two-year history of each product Max Bahr has sold. Overnight the solution calculates approximately 340 million replenishment targets and automatically turns 90 percent of them into orders, allowing planners to focus on managing the exceptions.
"We've seen our service levels in all our stores and warehouses reach 99 percent, significantly reducing the probability items will be out of stock and not available for our customers," says Anja Schoning, project manager for Max Bahr.
Timely, incisive articles delivered directly to your inbox.