The encyclopedia · Software & IT · Operational decision · 2023–2025
Meituan cut grocery spoilage 20% by rebalancing and repricing near-expiry stock
Meituan coordinated nighttime transshipment and daytime clearance pricing across micro-fulfilment centres so near-expiry food became profit, not waste.
Meituan
the move
Meituan's Little Elephant Supermarket runs a nationwide network of micro-fulfilment centres, and food near its expiry date was routinely wasted when demand and remaining shelf life did not line up at the right location.
The team built an integrated learning-and-optimization system that moves near-expiry stock between centres overnight and sets clearance prices by day, using causal models of price elasticity, a mixed-integer plan for rebalancing and a Markov decision process for pricing.
The rollout cut spoilage rates 20 percent, raised revenue on red-line inventory 14 percent, added about 50 million CNY (roughly 7 million USD) of annual profit and saved about 18 million CNY (roughly 2.5 million USD) of food waste across the network.
why it works
- Shelf-life rarely matches demand at any single micro-warehouse
- Rebalancing at night moves food where it will sell before expiry
- Clearance pricing finds the price that clears near-expiry stock
- Learning from each sale keeps the pricing and rebalance model current
what transfers
Perishable inventory is lost not from spoilage alone but from a mismatch of where and when it sits; rebalance and price it together so the network sells the shelf-life it has.
what came after
The work won the 2025 INFORMS Daniel H. Wagner Prize for Excellence in the Practice of Advanced Analytics and Operations Research, celebrated as a way to turn perishable inventory from a loss driver into a source of profit and sustainability.
references
- Rebalancing and Clearance Pricing of Near-Expiry Inventory in Online Grocery Retail
- Researchers from George Washington University, Meituan Group clinch INFORMS 2025 Daniel H. Wagner Prize
spotted an error? The archive wants to know.