AI agents for transport and logistics: practical uses

Transport and logistics depend on constant coordination: a dispatcher planning the rounds, drivers on the road, customers who want to know where their goods are, a workshop handling maintenance, a back office doing the invoicing. Between these people, information travels by phone, text, email, spreadsheets and sometimes paper in the cab.

AI agents do not drive the trucks and do not replace a transport management system (TMS) where one exists. They play a more modest and very useful role: moving information between tools and people, without re-keying and without anything being forgotten. This guide is for small and medium-sized businesses in the sector: carriers, local courier services, last-mile delivery, removal firms, logistics providers, and ride-hailing and taxi operators organised as small companies.

The most common frustrations#

Before talking about automation, here are the situations that come up in most businesses in the sector:

• a customer calls to ask whether their delivery is on its way, and nobody has the information to hand;
• operations know about a delay, but the customer is only told after the scheduled time;
• proof of delivery (photos, signatures, signed delivery notes) stays on drivers' phones;
• the date of a vehicle's next roadworthiness test or service is noted in a spreadsheet nobody looks at;
• transport requests arrive by email in every format and have to be re-keyed;
• invoicing waits until all the documents have come back.

Each of these frustrations corresponds to a linking task an agent can take over.

Use 1: preparing delivery rounds#

What the agent does#

The delivery rounds template starts from the list of the next day's deliveries (export from the order software, spreadsheet, customer emails) and proposes a grouping by area and a sensible delivery order, taking into account the time slots customers have requested. It sends the proposal to the dispatcher the evening before.

What it does not do#

It does not replace route optimisation software for large volumes, and it does not know the constraints you have not given it (a low bridge, no access for heavy goods vehicles, a customer who opens at 10 am). Add those constraints to its description or to a context file, and keep the dispatcher's approval.

Brief to copy#

Every day at 5 pm, read the deliveries planned for tomorrow in the "Deliveries" spreadsheet (the day's tab). Group them by area (list of areas in the attached file), propose a delivery order for each of our three vehicles respecting customer time slots, and email me the proposal. Flag any delivery without a complete address or without a time slot.

Use 2: warning customers of a delay before they call#

What the agent does#

With the customers notified of delays template, as soon as a delay is reported (by the driver, the dispatcher or a tracking tool), the agent sends the customer concerned a clear message: delivery delayed, new estimated time, a contact if needed. The message can go by email, text or WhatsApp depending on the customer's preferences.

Why it helps#

A customer warned in advance can plan around it. A customer waiting without news calls, gets annoyed, and ties up operations at the very moment they have the least time. Warning early reduces incoming calls and improves the relationship, even when the delay is unavoidable.

The rule to set#

Never announce a time that operations have not confirmed. If the time is uncertain, the agent says so: "new time being confirmed, we will get back to you before 2 pm".

Use 3: sending proof of delivery#

What the agent does#

The proof of delivery sent template collects the photos and signatures sent by drivers (on WhatsApp, by email or through an app), matches them to the right delivery, files them in a folder by customer and date, and sends the customer a confirmation with the proof attached.

The direct result: invoicing can go out sooner, since supporting documents are centralised as they come in. And in case of a dispute, the proof is found in seconds.

Use 4: tracking fleet maintenance#

What the agent does#

The fleet maintenance template keeps the key due dates up to date for each vehicle: services, roadworthiness tests, oil changes, tyres, insurance, and the other regulatory checks that apply to your business. Every week, it sends the list of upcoming due dates and can prepare appointment requests for the garage.

What to give it#

• the list of vehicles with their registration and mileage;
• the dates of the last services and inspections;
• the intervals that apply to each vehicle (according to the manufacturer and the regulations specific to your business);
• the contact details of your garage or garages.

The agent does not know your precise regulatory obligations on its own: it is up to you to tell it, or to check the intervals with your garage or trade association.

Use 5: tracking shipments and suppliers#

For businesses that ship rather than transport (e-commerce, wholesale, manufacturing), two templates round out the setup:

• shipment tracking: the agent tracks parcels handed to carriers, spots those that are stuck or late and alerts the customer or customer service;
• late suppliers: the agent tracks orders placed with suppliers, compares them with promised lead times and alerts you before a delay holds up production or deliveries.

Use 6: handling transport requests#

Quote requests arrive by email with scattered information: pickup address, destination, weight, volume, type of goods, requested date. An agent can read each request, extract the useful details into a table, flag missing information and prepare a draft reply or quote from your rate card. The dispatcher approves and sends. The request qualifier template can serve as a starting point.

Linking agents together#

These uses really come into their own when they hand work over to each other. Example chain for a regional delivery business:

On Future Agent, each agent writes to a shared table and the next agent starts when the previous one has finished, reading only the new rows. The workflow map shows you the chain in action.

An example day with agents#

Take a regional delivery business with eight vehicles, one dispatcher and an administrative assistant.

• 6:30 am: the dispatcher receives the round proposal prepared the night before and adjusts it for a vehicle that is off the road.
• 9:15 am: a driver reports a traffic jam on the ring road by voice message. The dispatcher confirms a delay of about forty minutes; the next four customers receive a message with the new time.
• Throughout the day: every photo of a signed delivery note sent by the drivers is filed and the customer gets their confirmation.
• 5 pm: the assistant finds in a table the list of the day's proven deliveries, ready to invoice, and those whose proof is missing.
• The following Monday: the fleet agent points out that a vehicle is due for a service in three weeks and suggests a slot at the garage.

Drivers at the heart of the setup#

None of this works if drivers have to learn a new tool. The right approach is to use what they already use: most often, WhatsApp. A driver sends a photo of the signed delivery note, a voice message to report a delay, a simple "delivered": the agent takes care of the rest.

A few simple rules help adoption:

• a single number or group for sending in information;
• short messages accepted as they are (no form to fill in);
• never any request while driving: the agent does not expect immediate replies from drivers.

Precautions specific to the sector#

• Personal data: delivery addresses, recipients' phone numbers and location data are personal data. Limit them to what is necessary and document the processing.
• Employee location tracking: if you use vehicle location data, the rules on tracking employees' location apply; find out about the requirements for informing staff and for proportionality.
• Liability: an agent does not commit the business to a lead time or a price without human approval.
• Emergencies: accident, breakdown, dispute over damaged goods: the agent alerts a person immediately, it does not handle it.

Where to start#

Templates for the sector are grouped in the library, transport and logistics sector, and the transport use cases page presents more examples. Each agent is described in a chat or on WhatsApp, with a price announced before launch; the features page details triggers, shared tables and workflows.