The transformation is not navigation assistance; it is moving routine sequencing authority from local manual planning into a network optimizer while preserving driver authority for physical-world exceptions.
AI value · Annual fuel consumed after completed U.S. ORION deployment
Company-reported
UPS reported 10 million gallons of fuel saved and 100,000 metric tons of emissions reduced each year
Operator-reported sustainability result; the report does not publish a randomized counterfactual or route-level distribution.
Before
Build and adjust a route sequence from service commitments and local knowledge. → Choose the driving sequence and react locally to changing conditions.
After
Calculates an optimized route that meets service commitments from package and network data. → Executes the proposed sequence and applies judgment when real-world conditions require deviation.
Human boundary
ORION proposes the stop sequence within service constraints; the driver remains responsible for safe vehicle operation and can deviate when conditions demand it.
Why it matters
That each driver or dispatcher must determine the daily stop sequence from experience and static route plans.
This case is company-reported. Use it for the operating-model shift; do not treat the numbers as independently measured.
How the work changed
Before
How the work ran before the change.
Step 1 of 2
Dispatcher and delivery driver
Build and adjust a route sequence from service commitments and local knowledge.
ControlHuman route judgment
Step 2 of 2
Delivery driver
Choose the driving sequence and react locally to changing conditions.
ControlService commitments and road rules
What changed
That each driver or dispatcher must determine the daily stop sequence from experience and static route plans.
Decision rightHuman authority remains at the consequential boundary
After
How the same work runs now.
Step 1 of 2
ORION
Calculates an optimized route that meets service commitments from package and network data.
ControlUPS service and route constraints
Step 2 of 2
Delivery driver
Executes the proposed sequence and applies judgment when real-world conditions require deviation.
ControlDriver retains safe-operation authority
Process model built from the published workflow evidence for UPS. Every step, actor, and control appears in full below.Every step, actor, and control
Exception path
Drivers handle road closures, unsafe maneuvers, access problems, and customer-specific conditions outside the route model.
Decision authority
ORION proposes the stop sequence within service constraints; the driver remains responsible for safe vehicle operation and can deviate when conditions demand it.
Before
#
Actor
Action
Control
01
Dispatcher and delivery driver
Build and adjust a route sequence from service commitments and local knowledge.
Human route judgment
02
Delivery driver
Choose the driving sequence and react locally to changing conditions.
Service commitments and road rules
After
#
Actor
Action
Control
01
ORION
Calculates an optimized route that meets service commitments from package and network data.
UPS service and route constraints
02
Delivery driver
Executes the proposed sequence and applies judgment when real-world conditions require deviation.
Driver retains safe-operation authority
Work that left the path
Repeated manual construction of routine stop sequences
Avoidable miles caused by locally optimized rather than network-optimized routes
Human role before
Dispatchers and drivers assembled and adapted route sequences manually.
Human role after
Drivers execute an algorithmically sequenced route and retain authority to handle road, safety, and service exceptions.
AI role
Advanced routing and optimization system that sequences stops and, in Dynamic ORION, adapts routes from changing operating data.
Outcomes
Annual fuel consumed after completed U.S. ORION deployment
Company-reported
Pre-ORION U.S. routing without the completed optimization deployment→UPS reported 10 million gallons of fuel saved and 100,000 metric tons of emissions reduced each year
Annualized result reported for 2016 · Completed U.S. deployment
Operator-reported sustainability result; the report does not publish a randomized counterfactual or route-level distribution.
What leaders can reuse
Anti-pattern
Treating the annual savings estimate as a controlled causal result or forcing drivers to follow an unsafe route recommendation.
Questions
01Which recurring movement decisions are still locally optimized?
02Where must operators retain immediate override authority?
Portability conditions
Dense repeatable routing data
Explicit service and safety constraints
A human override path for physical-world exceptions
Reputation risk
low
Evidence and authority
What the public record supports.
Current · updated
2 primary; publication outcomes are reported.
Bundle 1.0.0 · reviewed 2026-09-06 · stable ID 5ce05f8c60201a35