How J.B. Hunt Saved $200M by Automating Freight Matching — What Asset-Based Carriers Must Know

J.B. Hunt's AI matches loads to drivers 11x faster than human dispatchers and with 23% fewer empty miles. Their freight matching automation saved $200 million in 2024 alone, according to their annual operations report.

The part that matters for asset-based carriers: J.B. Hunt did not start with the most complex AI. They started with the most repetitive task and automated that first.

J.B. Hunt: AI-powered load-to-driver matching

The challenge: J.B. Hunt operates over 12,000 trucks and handles millions of loads annually. Their dispatch teams were manually matching drivers to loads using spreadsheets, phone calls, and institutional knowledge. When a veteran dispatcher retired, their knowledge of which drivers preferred which lanes left with them. Matching quality was inconsistent, driver satisfaction was declining, and empty miles were climbing.

The AI solution: J.B. Hunt built J.B. Hunt 360, an AI platform for freight matching:

Evaluates every available load against every available driver simultaneously, considering home time requirements, equipment type, lane preferences, hours-of-service status, and historical performance
Learns individual driver preferences from acceptance/rejection patterns and builds personalized load recommendations
Predicts load availability 48 hours ahead based on shipper patterns and seasonal trends
Automatically assigns loads when confidence scores exceed threshold, without dispatcher intervention

Measurable results:

$200 million in annual savings from optimized matching
11x faster load matching compared to manual dispatch
23% fewer empty miles through predictive backhaul assignment
Driver retention improved 15% through better load matching aligned with preferences
18% more loads per driver per week through reduced wait times

J.B. Hunt's AI matches loads to drivers 11x faster than human dispatchers and with 23% fewer empty miles. The same truck, the same driver, better assignments.

For a deeper look at matching algorithms, see Most Common AI Algorithms Used for Route Planning and Demand Forecasting.

You don't need 12,000 trucks to apply this

You don't need J.B. Hunt's fleet size. You have an asset-based carrier with 50–500 trucks, drivers who complain about load assignments, and dispatchers who spend 3–4 hours per day on phone calls that could be automated.

That is where matching AI pays off fastest, because every hour your dispatcher spends on a phone call is an hour they are not optimizing the next assignment.

Werner Enterprises: AI for driver retention through better assignments

The challenge: Werner operates 8,000+ trucks. Driver turnover in trucking exceeds 90% annually industry-wide. Werner identified that the #1 reason drivers left was dissatisfaction with load assignments, not pay. Drivers wanted loads that got them home on time and kept their wheels turning.

The AI solution: Werner built driver-centric assignment AI:

Prioritizes driver home-time commitments when assigning loads, treating them as hard constraints rather than suggestions
Predicts which drivers are at risk of leaving based on recent assignment patterns and compares against the patterns of drivers who left in the past
Adjusts assignments automatically to reduce consecutive weeks away from home

Measurable results:

12% reduction in driver turnover within the first year
$60 million saved annually in recruitment and training costs (at ~$12,000 per new driver)
Driver satisfaction scores up 22% through better home-time compliance
8% improvement in fleet utilization through reduced empty driver seats

Read about how control towers coordinate across fleets at What is an AI-Powered Control Tower in Logistics?.

Schneider National: AI for intermodal freight optimization

The challenge: Schneider runs one of the largest intermodal fleets in North America. Deciding when to use truck versus rail versus a combination required evaluating cost, transit time, reliability, and carbon impact for each shipment. Human planners defaulted to whatever mode they used last time.

The AI solution: Schneider built mode-optimization AI:

Evaluates all mode combinations for each shipment based on real-time cost and capacity data
Identifies shipments that can shift from pure truck to intermodal without service level impact
Predicts rail capacity and transit time reliability at the lane level

Measurable results:

15% increase in intermodal conversion from pure truckload
$120 million in shipper cost savings through optimized mode selection
12% reduction in carbon emissions through increased rail utilization
Transit time predictability improved 20% through better rail capacity forecasting

See how AI optimizes broader supply chains at A Simple Analogy for How AI Optimizes a Supply Chain.

Knight-Swift: AI for fleet utilization across brands

The challenge: Knight-Swift (formed from the merger of Knight Transportation and Swift Transportation) operates the largest truckload fleet in North America. Post-merger, they needed to optimize load assignments across two separate driver pools, equipment fleets, and customer bases without disrupting either operation.

The AI solution: Knight-Swift built cross-fleet optimization AI:

Evaluates loads against drivers from both legacy fleets simultaneously
Identifies synergies where a Knight driver is better positioned for a Swift customer's load and vice versa
Gradually learns which cross-fleet assignments work and which create service issues

Measurable results:

8% improvement in overall fleet utilization through cross-fleet matching
$150 million in annual savings from better asset deployment
Empty miles reduced by 10% through expanded matching pool
Customer service levels maintained through controlled rollout with feedback loops

Heartland Express: AI for dedicated fleet optimization

The challenge: Heartland Express runs dedicated fleets for specific customers. Dedicated operations have different optimization challenges than spot freight because the same trucks serve the same customers daily, but volumes fluctuate.