EN
Back to the archive

The encyclopedia · Engineering & Operations · Technical decision · 2008–2016

UPS's ORION software cut 100 million miles a year by recomputing every driver's route

ORION crunches telematics and 250M+ address points to route 55,000 drivers, saving an expected 100M miles, 10M gallons of fuel and $300-400M a year.

UPS

The solution

UPS drivers knew their routes by heart, and years of habit quietly added miles. UPS built ORION, an operations-research system that computes the best delivery order for every route instead of trusting routine.

The project was prototyped from 2008 to 2011, launched in 2013, and was expected to complete US deployment by the end of 2016, reaching 55,000 drivers across 1,000 buildings.

At full deployment UPS expected annual savings of 100 million miles driven, 10 million gallons of fuel, 100,000 metric tons of CO2, and $300-400 million in savings and cost avoidance.

UPS won the 2016 INFORMS Franz Edelman Award for the work, which involved more than 700 employees building and rolling out the system.

Why it worked

  • A machine can hold 250 million address data points; a driver holds one route.
  • Small per-driver mile cuts compound across 55,000 drivers.
  • Telematics made driver behavior measurable for the first time.
  • A system-wide optimum beats thousands of local habits.
What it achievedLet software own each day's routeclever

What can be applied

When thousands of expert habits each hide a few wasted miles, an optimizer that owns the daily decision compounds tiny savings into hundreds of millions — deployed with the drivers, not against them.

Aftermath

ORION became the flagship of UPS's analytics story and the platform its CIO said would fuel new products and services; the Edelman win made the project the public face of operations research in logistics.

Sources

spotted an error? The archive wants to know.

Related cases