LogisticsIndustry ContextThursday, August 13, 20265 min read

AI agents can find the load. MapUp helps them determine its profitability

Freightwaves5h agogeneral
AI agents can find the load. MapUp helps them determine its profitability
Executive Summary

Autopilot dispatchers rank freight by posted rate because nobody gave them the arithmetic. MapUp’s new MCP server prices a lane’s fuel, tolls and time in seconds. The post AI agents can find the load. MapUp helps them determine its profitability appeared first on FreightWaves.

Source Lens

Industry Context

Useful background context, but lower-priority than direct platform, community, or operator intelligence.

Impact Level

medium

Use this briefing to decide whether your team needs an immediate workflow, policy, or reporting change.

Key Stat / Trigger

No single quantitative trigger surfaced in this report.

Focus on the operational implication, not just the headline.

Relevant For
Brand SellersAgencies

Full Coverage

AI dispatch agents have gotten good at finding freight. They’ll pull from load boards, filter by equipment and radius, and return a ranked list in seconds. What they still struggle to do is tell an operator whether that top-ranked load actually pays. MapUp aims to close that gap with the launch of FuelGuru MCP.

The company calls it the first production Model Context Protocol server built for fleet fuel purchasing, tolls and commercial truck routing. It lets any AI agent, load board, dispatch tool or TMS copilot ask MapUp’s engine what a trip will really cost for a specific truck on a specific day under a fleet’s own rules.

“While these AI agents are already helping to pick a good load, the next level of intelligence is defining what is a good load,” MapUp CEO Katie Mahlawat told FreightWaves. “Right now the good definition is time and cost and how that trade-off plays out.” The load that looks fine but really isn’t Mahlawat used a concrete example.

A broker posts $1,800 all-in for a five-axle dry van from Harvey, Illinois, to Philadelphia. On the 773-mile practical route, that works out to about $2. 33 a mile. The posted rate alone says nothing about what remains after route-specific fuel, tolls and time. In an August 2026 FuelGuru analysis, MapUp priced three commercial routes for the same trip.

The practical I-80 route covers 773 miles in 11 hours and 49 minutes of driving, more than a solo driver can legally run without a 10-hour break, with $192. 97 in tolls and $554. 58 in fuel, leaving $1,052. 45 of the rate before driver pay, equipment and fixed costs.

The fastest option, which combines I-80 with the Pennsylvania Turnpike, arrives 23 minutes sooner but leaves only $908. 91. That is $144 in extra fuel and tolls to buy back 23 minutes — a marginal tractor hour priced at $374. Very few fleets in North America treat a marginal hour of driver time at that rate.

Many simply price all-in or chase a target rate per mile. The truck then takes the faster route because nobody priced the alternative before it rolled. The cheapest line on the table punishes the opposite instinct. An I-70 routing carries the lowest tolls at $150. 36 but adds 39 miles and 62 minutes.

At a $54-an-hour loaded driver cost, that extra hour costs about $56 to save $22. 77 in combined fuel and tolls — a net loss before it threatens the delivery window. “This is why ‘avoid tolls’ is bad policy and ‘always take the toll road’ is worse,” wrote MapUp co-founder and Chief Technology Officer Maneesh Mahlawat in the company’s launch note.

“Neither is a decision. Both are habits.” Nick Brooks, vice president of technology and marketing at Western Express, said the timing of the calculation is what carries the operating value. “Being able to come back with the real fuel cost and the real toll cost at the time you are going to incur it is infinitely valuable,” Brooks said.

“As toll routing gets more complicated, it is going to make profitability happen quicker.” A universal power adapter for freight math FuelGuru first launched last year at FreightWaves’ F3: Future of Freight Festival, where FirstFleet Chief Information Officer Austin Henderson demonstrated it alongside NavGuru, MapUp’s commercial navigation product.

It originally shipped as an API, so every platform that wanted the engine had to build a custom integration. “MCP is like a universal power adapter,” Katie Mahlawat said. “You could say you want to connect your own phone to anyone else’s speaker. How do you like them to talk to each other? That’s it.” Now the integration is largely a credential.

“Get an API key from us, plug it in your Claude configuration, it gets added as a skill,” she said. “Now you have an agent which is a FuelGuru brain.” An agent hands over the load, truck position, equipment, appointment windows, tank level, fuel economy, card pricing and fleet rules.

FuelGuru returns practical, fastest, cheapest and alternate routes with drive time, vehicle-specific tolls, fleet-net fuel cost and prescribed stops. The pricing is fleet-specific, not a national average.

FuelGuru costs the negotiated rate the fleet’s card will actually capture at that station, against the state tax spread, the out-of-route miles to reach it, and the arrival time the hours-of-service plan predicts. Lane profitability used to take a month and five teams At a large carrier, pricing a lane often feels like navigating a committee.

A sales team quoting next quarter’s rates collects cost history from the fuel desk, the toll team and finance. Those groups rarely hold the same version of the truth because of different incentives and data silos.

“That process takes somewhere between two, three weeks to a month, as well as coordination between four or five different teams, for their sales team to have that at their fingertips,” Mahlawat said.

For fleets already running MapUp, including FirstFleet and Western Express, the MCP server can read the carrier’s own history instead of relying on averaged or lagged data. T

Original Source

This briefing is based on reporting from Freightwaves. Use the original post for full primary-source context.

View original
LinkedIn Post Generator

Style

Audience