Detect unnoticed out-of-stock inventory in stores
Target Audience
Store operations teams, inventory managers
Challenge
Physical audits revealed that half of Target's out-of-stocks were unknown to their inventory systems — meaning systems showed items as available when shelves were actually empty. Traditional tracking couldn't detect these discrepancies caused by shipping mistakes, theft, and misplaced items. With over 100,000 SKUs across nearly 2,000 stores, manual detection was impossible.
Solution Approach
Target created the Inventory Ledger to track every inventory change in real-time. They developed an ensemble of AI models, each specialized for different product categories, trained on millions of labeled examples of unknown out-of-stocks. The system automatically detects patterns indicating inventory inaccuracies, then triggers replenishment.
Value Add
Target Target reports 'substantial sales lift' for products that would otherwise have been unavailable. The AI eliminated the need for costly shelf-edge cameras and sensors that had proven difficult to maintain at scale.
Image credentials: Sandie Clarke/ Unsplash
More Interesting Use Cases
Supply Chain Optimization
Walmart leverages AI and machine learning across its entire supply chain to predict demand, optimize inventory placement, and improve logistics efficiency across 4,700+ stores and fulfillment centers.
AI-Enabled Antibody Discovery and Optimization for Faster Drug Development
AI accelerates antibody discovery, reducing optimization cycles and bringing better drugs to patients faster.
Accelerating Sales and Customer Service Efficiency
AI assistants cut customer research time from hours to minutes, recovering millions in lost sales revenue.