UPS is building a real-time digital twin of its entire logistics network, updated every 10 minutes

The company says AI-driven network planning now completes analyses in one afternoon that previously took engineering teams months.


UPS is building a real-time digital twin of its entire logistics network, updated every 10 minutes Image by: Jaggery

TL;DR

UPS detailed AI plans including a digital twin of its full network, agentic control towers, RFID tracking, and AI customs clearance hitting 97% first-day rates.

UPS detailed a slate of AI-powered logistics initiatives on Wednesday, including a real-time digital twin of its entire global network that updates every 10 minutes. The system creates a digital replica of facilities, air and ground networks, and end-to-end package flows. It continuously tracks performance so the network can adjust and, in UPS’s words, “self-heal in real time.

The company says its AI-driven network planning tools now complete analyses in one afternoon that previously took engineering teams several months. Forecast accuracy has improved by up to 40%, according to Supply Chain Dive, shrinking costly buffer capacity and enabling a 9.9% reduction in US labour hours during recent volume declines.

After 118 years of reinventing logistics, we have entered a defining moment, using AI to simplify how we work across the enterprise,” said Carol B. Tomé, UPS chief executive officer. “We are pairing the deep expertise of our people with the power of AI to drive faster decisions and a better experience for our customers around the globe.”

The 💜 of EU tech

The latest rumblings from the EU tech scene, a story from our wise ol' founder Boris, and some questionable AI art. It's free, every week, in your inbox. Sign up now!

UPS is also deploying what it calls agentic “control tower” capabilities on-site with customers. These combine data, predictive models, and connected services to go beyond shipment tracking. They flag, prioritise, and help resolve disruptions across complex, multi-carrier networks. Agentic AI is increasingly moving from theory to production across enterprises, and UPS’s deployment is one of the largest logistics implementations announced so far.

On the tracking side, UPS has embedded RFID sensors in all US package delivery vehicles, in delivery facilities across the country, and on every package shipped through over 5,500 The UPS Store locations. The company says misloads have dropped by nearly 70% since the rollout began three years ago. Combined with AI, the RFID data delivers near real-time, package-level visibility.

UPS’s AI customs brokerage is clearing 97% of shipments on the first day of entry, according to the company. The system uses AI and cross-border data to interpret customs requirements worldwide, and the company says it is outpacing competitors on clearance speed. A separate product, UPS Export Assure, uses AI for more accurate product classifications, reducing errors and accelerating processing.

Happy Returns, the reverse logistics company UPS acquired, is using AI to combat returns fraud. The system photographs return contents and compares them against expected product images using its “Return Vision” AI tool. Happy Returns piloted the system with Everlane, Revolve, and Under Armour. Returns fraud is a $76.5 billion problem for US retailers, according to the company.

UPS reported $88.7 billion in revenue for 2025 and has approximately 460,000 employees globally. The company’s “Network of the Future” initiative is backed by a planned $9 billion investment over five years, expected to generate $3 billion in recurring annual savings. Digital twins and AI-driven operational planning are spreading across industries from motorsport to shipping, but UPS’s implementation, covering every mode of transport across more than 200 countries, is among the most ambitious in scale.

Get the TNW newsletter

Get the most important tech news in your inbox each week.