UNFI's Fuel-Cost Response Needs a Driver-to-Route Exception Queue

When diesel prices surge, a fleet dashboard can explain the damage without changing the outcome. The better operating model is an exception queue that identifies avoidable fuel consumption, assigns each event to the person who can fix it, and records the gallons and dollars saved.

That distinction matters now. Supply Chain Dive reported that U.S. on-highway diesel reached $6.18 per gallon on October 12, 2026, up $2.51 year over year. United Natural Foods Inc. is combining fuel hedging with connected-fleet technology to improve deliveries and reduce unnecessary miles across roughly 2,000 delivery vehicles and nearly 50 distribution centers.

Hedging can stabilize the price paid for fuel. Operations must control how much fuel the network consumes. That requires separating market inflation from operational waste and converting telematics signals into owned, time-bound work.

Separate the Fuel Price From Fuel Waste

A useful fuel-cost model starts with two variables:

Fuel cost = gallons consumed × effective price per gallon.

Procurement and finance influence the effective price through contracts, hedges, fuel cards, and surcharge programs. Transportation teams influence gallons through routing, driving behavior, equipment condition, detention management, and network design. Combining both variables in one variance hides ownership.

Create a baseline for each route using miles, payload, equipment type, terrain, temperature-controlled load requirements, number of stops, and expected dwell. Then compare actual performance against that operating baseline. A route should not trigger an alert merely because diesel became more expensive; it should trigger when its consumption or activity deviates from what planners could reasonably control.

The distinction also keeps teams focused during price shocks. Reuters reported earlier in 2026 that diesel is the second-largest operating expense for truckers and that the national average retail price had risen $1.89, or 50%, after disruption to shipping through the Strait of Hormuz. When the market moves that sharply, eliminating one hour of idle time or one unnecessary trip becomes far more valuable—but the price increase itself is not a dispatcher failure.

Build Four Exception Lanes

Not every inefficient trip belongs in the same queue. Route each exception according to its probable owner.

1. Driver exceptions

Assign prolonged idling, speeding, harsh acceleration, excessive engine revving, and unauthorized route deviations to driver coaching. The record should include the event, location, duration, estimated gallons, operating context, and a link to the relevant trip. Configure thresholds carefully so refrigeration needs, traffic controls, safety stops, and severe weather do not create false accusations.

Coaching should focus on repeatable behavior, not isolated alerts. One hard acceleration event is noise; a persistent pattern across comparable routes is actionable.

2. Dispatch exceptions

Assign avoidable empty miles, poor stop sequencing, late re-routing, missed backhauls, and equipment mismatches to dispatch. These events often look like driver inefficiency when viewed only through vehicle data. Connecting the trip to orders, appointments, capacity, and route plans reveals whether the driver followed a bad plan perfectly.

Dispatchers need recommended actions with the alert: resequence stops, offer an available backhaul, switch equipment, consolidate a low-volume run, or move a delivery to a better departure window.

3. Customer exceptions

Assign recurring detention, refused early arrivals, unavailable docks, and appointment noncompliance to customer-service or account teams. A truck idling outside a grocery facility may be consuming fuel because the consignee cannot receive it, not because the driver chose to wait with the engine running.

Aggregate these events by customer and location. The commercial team can then renegotiate appointment windows, add detention terms, change delivery frequency, or document accessorial charges with trip-level evidence.

4. Network-planning exceptions

Assign structurally long routes, chronic empty return legs, overlapping service territories, and recurring capacity imbalances to network planning. These are not same-day dispatch problems. They may require adjusting distribution-center assignment, delivery cadence, pool points, fleet mix, or customer sourcing rules.

UNFI's scale makes this category especially important. A small percentage of unnecessary mileage repeated across 2,000 vehicles can outweigh dozens of successful driver-coaching events.

Make the Queue Operational

Each exception needs a severity score, owner, due date, recommended response, and financial estimate. Severity should combine excess gallons, current fuel price, recurrence, customer impact, and confidence in the diagnosis. This prevents a flood of low-value alerts from crowding out a repeated route-design problem.

Use workflow states such as new, acknowledged, investigating, actioned, dismissed, and verified. Require a reason when an exception is dismissed. Over time, those reasons expose bad thresholds, missing operating context, and data-quality gaps.

The queue also needs escalation rules. A single 20-minute detention event may remain with dispatch. Five events at the same receiving location in two weeks should escalate to the account owner. A repeated empty return across several routes should move to network planning.

Measure Avoided Gallons, Not Dashboard Activity

Logins, alerts opened, and coaching sessions completed are adoption measures. They do not prove a fuel program works. The primary scorecard should quantify operational results:

• excess gallons before and after intervention;
• fuel cost avoided at the actual or hedged price;
• unnecessary miles eliminated;
• idle hours reduced;
• empty-mile percentage by lane;
• detention recurrence by customer location; and
• savings sustained for 30, 60, and 90 days.

Use a comparable baseline and document external factors. A lighter payload, different weather, or a route reassignment can improve miles per gallon without any intervention. Conversely, a route can consume more total fuel while becoming more efficient per delivered case.

The strongest metric is cost avoided per verified intervention. It connects a specific action—driver coaching, route resequencing, a revised appointment, or a network change—to the change in gallons consumed. It also tells managers where another hour of effort will produce the largest return.

Turn Volatility Into Daily Control

Fuel hedging and connected-fleet visibility solve different problems. One manages exposure to price volatility; the other can reduce operational consumption. Neither delivers its full value if insights remain trapped in finance reports or telemetry dashboards.

A driver-to-route exception queue closes that gap. It directs controllable waste to the right owner, preserves evidence, escalates recurring patterns, and verifies whether an intervention reduced gallons. In a high-price environment, that is the difference between observing fuel inflation and operating against it.

Ready to connect route plans, shipment execution, and fuel exceptions in one workflow? Request a CXTMS demo to see how transportation teams can turn fleet signals into measurable action.