Taisei Corporation and FANUC have jointly developed an AI-powered automatic picking system that retrieves specified products from mixed-load pallets without requiring pre-registered product data. The system, already deployed at a manufacturing factory in January 2026, eliminates manual picking work and reduces shipping errors while enabling warehouses to stack multiple product types on single pallets, significantly improving storage efficiency and reducing the need for facility expansion.
Summaries like this, in your inbox every morning.
Sign up free →What happened
Taisei Corporation and FANUC jointly developed an AI-powered picking system that automatically retrieves specified products from mixed-load pallets. The system uses image processing and three-dimensional recognition combined with automatic product identification, eliminating the need for pre-registered product data. It was deployed at a manufacturing plant in January 2026.
Why it matters
Warehouses and factories currently rely on workers to manually pick items from instructions, a repetitive task prone to shipping errors and product damage. This system automates unloading (depalletization) entirely, reducing human labor during an era of workforce shortages. It also enables mixed-load pallet operations — stacking multiple product types on one pallet — which increases storage efficiency and throughput compared to traditional single-product pallets.
What to watch
The system uses dual verification — image recognition plus barcode identification — to prevent shipping mistakes, and the robot arm adjusts handling based on product shape and weight to reduce breakage risk. Beyond newly built warehouses, it can also help existing facilities avoid costly expansions by optimizing space utilization.
Warehouse and factory operations have long struggled with manual picking work, where employees check picking lists and retrieve items one by one. This repetitive process creates two persistent problems: shipping errors and product damage are common, and as labor populations decline, staffing these roles becomes increasingly difficult. Taisei and FANUC's solution addresses both by automating the depalletization step entirely through AI-driven vision and robotic handling.
The technical innovation lies in eliminating the need for pre-registered product master data. Traditional vision systems require detailed information about each product's dimensions, appearance, and ID to be entered beforehand—a time-consuming setup step. By combining three-dimensional image recognition with automatic cross-referencing of shipping instructions, this system can identify and handle new products without prior configuration, enabling rapid deployment. The dual-verification approach (image plus barcode scanning) and weight-adaptive robot control further reduce both human error and equipment damage.
Beyond labor savings, the system unlocks an efficiency gain that has been constrained by traditional warehouse design. Single-product pallets—the industry standard—become increasingly inefficient as items are removed and inventory levels drop. By enabling mixed-load operations, warehouses can pack pallets more densely and consistently, reducing the total storage footprint required. This means existing facilities can avoid costly expansions, and new facilities can be sized more optimally. For businesses managing logistics at scale, the combination of reduced headcount, fewer errors, and lower capital requirements represents a meaningful shift in warehouse economics.
No comments yet. Be the first to share your thoughts!
Log in to join the discussion
Get curated AI news from 200+ sources delivered daily to your inbox. Free to use.
Get Started FreeFree · takes 30 seconds · unsubscribe anytime
1 minute a day. The AI essentials.
200+ sources · Email / LINE / Slack
