XPO Logistics deploys machine learning to optimize supply chain (with video)

Harik said XPO spends $500 million annually on its proprietary technology systems, with a focus on four categories.

The company is using artificial intelligence and machine-learning capabilities to analyze consumer demand and predict inventory for its retail customers. Other applications include providing end-to-end shipment visibility and optimizing sequencing “so drivers have the best route possible,” Harik said.

XPO Connect, the company’s digital freight marketplace, features more than 60,000 carriers, and the company is using machine learning to become smarter at helping shippers and carriers buy and sell capacity. A new COVID-19 dashboard allows customers to see the impact of the pandemic on the supply chain.

Warehouse optimization is another area of interest. As shippers transition from moving pallets for retailers to individually purchased products for consumers, they need much more efficient warehouses, Harik said. Autonomous mobile robots are among the solutions that can make item picking more efficient and safer.

Many of the consumer purchasing trends that took shape during COVID-19 are likely to persist in a post-pandemic world, he believes. XPO is seeing millions of units per year of inventory returns, and that category will likely grow as more shoppers buy — and return — items online.

With reverse logistics, technology deployment only becomes more important, Harik said.