Skip to content
Plymouth AI Lab

AI dispatch and routing

AI dispatch and route optimization for trucking and delivery teams

AI dispatch uses software to read incoming loads, suggest which driver and truck should take each one, and draft the messages that go with it. Route optimization plans which vehicle visits which stops, and in what order. We build both around the dispatch software and TMS you already run, and a dispatcher approves every assignment before it goes out.

Last updated 26 October 2026

On this page: Dispatch software or an AI assistant · What the assistant does · Route optimization · Buy or build · What you need in place · What it costs to start · Risks and when it is not worth it

Trucking dispatch software, a TMS, or an AI dispatch assistant?

The three get mixed up, so here is the difference in plain words.

What it isBuy or build
Dispatch softwareThe board where loads are assigned to drivers and trucks, often with a driver app for status and documentsBuy. Many products exist for trucking and courier fleets, and many TMS products include it
TMSThe system of record for loads, rates, carriers and billingBuy. Plymouth does not sell one
AI dispatch assistantA layer that reads load requests, suggests assignments and drafts messages, working inside or next to the two aboveBuild or add on. This is what we build

What an AI dispatch assistant does

The dispatcher stays in charge. The assistant prepares the work. A person decides and sends.

Hours of service come first

Any assignment suggestion has to respect the driver's legal hours. In the US, a property-carrying driver may drive a total of 11 hours, and may not drive once 14 consecutive hours have passed since coming on duty after 10 consecutive hours off duty (49 CFR 395.3, checked 26 October 2026). Under Canada's federal rules for extra-provincial carriers, south of latitude 60°N, a driver may not drive after 13 hours of driving time or 14 hours of on-duty time in a day (Commercial Vehicle Drivers Hours of Service Regulations, SOR/2005-313, section 12, checked 26 October 2026). Those are the headline limits; both rule sets have more conditions. An assistant reads the hours left from your ELD or TMS and leaves a driver out when the load would not fit. Compliance stays with the carrier and the driver.

Route optimization: what it is and what it needs

Route optimization means choosing which vehicle visits which stops and in what order, under your rules. Google's open-source OR-Tools describes the vehicle routing problem as finding optimal routes for multiple vehicles visiting a set of locations, with constraints such as vehicle capacity and time windows (Vehicle Routing Problem, OR-Tools documentation).

"Optimal" depends on what you ask for: the shortest total distance, the earliest finish, the fewest trucks. A route optimizer is only as good as what it is told:

Shippers and 3PLs rate routing highly. In the 31st Annual Third-Party Logistics Study, transportation planning and routing gave the strongest return on investment from AI, scoring 3.15 out of 4 for 3PLs and 2.71 for shippers (2026 3PL Study highlights AI adoption, freight fraud and supply chain disruption, Logistics Management, 6 October 2026). That is a respondent rating, not a measured saving.

Route optimization software: buy or build?

If you need a route planner for a delivery fleet, buy one. Products such as Routific, Onfleet and Upper offer route planning, and for a standard delivery fleet buying one is usually the simpler path. For a handful of stops a day, a free route planner app may be all you need. Where we help is around the planner: getting orders into it without retyping, sending the planned routes to drivers and customers, and feeding actual times back so the plan gets better.

For teams with unusual constraints, such as mixed equipment, tight time windows or multi-day trips, we can build routing on open-source solvers like OR-Tools, which Google publishes under the Apache License 2.0, so it is free to use (OR-Tools licence). Either way, the routes land in the tools your dispatchers already use.

Worked example: a change that waits for approval

In our Logistics OS demo, an operations copilot can find, create and edit shipments, but it never writes on its own. Its example of upgrading a delayed shipment from standard to express shows the change side by side: service Standard to Express, delivery Thursday 22 October to Wednesday 21 October, charge US$1,284.00 to US$1,432.00, with Approve and Reject buttons. Nothing changes in the shipment record until a person approves. The shipment and prices are demo data.

What you need in place

What it costs to start

A route planner is a subscription priced by each vendor, so check current prices with the ones you shortlist. OR-Tools has no licence fee, but a custom routing build still costs the work to model your rules and host it. The dispatch assistant itself is a fixed price agreed before we start. Starting with the messages, before any assignment logic, keeps the first build small.

Risks, and when it is not worth it

How we work

  1. Process before tools. We learn how the work really moves before we build anything.
  2. One bottleneck at a time. We pick the step that costs the most hours or the most mistakes and build for that first.
  3. On the systems you already run. We connect to your WMS, TMS, inbox and spreadsheets instead of asking you to switch.
  4. A person approves every action. Agents draft and propose. Nothing is sent to a customer, carrier or driver until someone on your team says yes.
  5. Fixed price, tested on us first. No billing by the hour, and we prove our method on our own work first.

Related services

Dispatch sits next to the rest of the office work: see logistics automation for load paperwork and billing, logistics dashboards for on-time and cost per load, or all logistics services.

Questions logistics teams ask

How can a small trucking company use AI to automate dispatch?

A common starting point is the messages rather than the decisions. An AI assistant can read load requests, fill in the load record and draft driver confirmations and customer updates, while the dispatcher still chooses who takes the load. Once that works, add assignment suggestions based on location, equipment and hours.

What is route optimization?

Route optimization is choosing which vehicle visits which stops, and in what order, so the plan meets your goal, such as the least driving or the earliest finish, within limits such as truck capacity, delivery time windows and driver hours.

What is the difference between dispatch software and a TMS?

Dispatch software is where loads are assigned to drivers and trucks, often with a driver app. A TMS is the wider system of record for loads, rates, carriers and billing, and many include dispatch. An AI dispatch assistant works alongside either one.

Does AI dispatch account for hours of service?

It should. A suggestion needs the hours each driver has left, read from the ELD or TMS, so it can leave out a driver who could not legally complete the load. The dispatcher still approves, and compliance stays with the carrier and driver.

Will AI dispatch replace our dispatchers?

No. It removes the typing, copying and checking that fill a dispatcher's day, so they can spend more time on decisions and exceptions. Every assignment and every message is approved by a person before it goes out.

What is the best route optimization software?

It depends on the fleet. Last-mile delivery teams often use planners such as Routific, Onfleet or Upper. Trucking fleets with long hauls and hours-of-service limits need routing that respects those rules. Pick the planner that fits your stops and constraints, then connect it to your orders so nobody retypes them.

Who builds it

Plymouth AI Lab is run by its three founders: Noam Turecki, Phil Swart and Alfredo Borras.

From data points to pathways.

Book a 1:1 intro call and see how a small AI system could work on the data your operation already has.

Book a call