AI for Supply Chain and Logistics · Solutions

The dispatcher approves each day’s drafted routes.

AI Route Optimisation reads the day’s delivery list, the vans available and their load limits, and drafts an order of stops for each driver on a map. Stops can be moved. Approval belongs to the dispatcher. Nothing reaches a driver before that.

Share your requirements

What it is

Routes are built from memory. The planner knows the old market jams before ten and that one customer accepts goods only after lunch, yet none of it is written down. Then the day goes wrong.

Orders and time windows go in. A suggested sequence of stops for every van comes out as a draft, with the distance, the expected arrival and the load shown against each stop. Sending waits for the dispatcher.

Where it goes wrong today

The problems this is meant to ease.

Routes live in one planner’s head

Only the dispatcher knows which gate shuts for lunch, which lane floods after rain and which customer refuses afternoon drops, and none of it is written down anywhere. Their sick day becomes everybody’s bad day.

Vans leave overloaded or half empty

Loads are shared out by eye, so one van carries a full body of heavy pallets while another drives a long loop with a few boxes. Fuel and overtime pay for the mismatch.

Customers wait without an answer

The buyer rings the office to ask where the van is, and the office rings a driver who is carrying a box up two flights of stairs. Nobody gets a clear answer.

Disruptions break the plan

A closed road, a breakdown or a cancelled drop forces the planner to redo the round by phone, and the other drivers hear one call at a time. The change is never recorded.

Nobody can say what a delivery costs

Distance, fuel, driver hours and failed drops are scattered across books, so any single delivery’s cost is a guess made when the books close. Pricing rests on that guess.

Inside the module

What AI route optimisation does.

Routes from customer locations

Each order carries a pinned address. Every drop is pinned on the map and the gaps between them are measured, so the planner starts from real geography and not from a street name.

Stops grouped by vehicle

Orders are shared out across vans. Each van gets a load that fits its weight limit and body type, so chilled orders stay off open trucks and heavy ones stay off small vans.

Many stops, one order

Many stops, one van, one order. Different sequences are tried, with distance and expected arrival shown beside each drop. At a glance the dispatcher sees which van has the longest day.

Distance, travel time and limits

Limits are stored as rules. Delivery windows, vehicle weight limits, driver hours and gates that close for lunch are held with the customer or the vehicle and applied to every plan that includes them.

Priorities and schedules

Some orders cannot wait. A perishable load, a rush order or a customer on hold is marked, and the schedule places those drops first and fills the rest around them.

Mid-day changes

Roads close and customers cancel. A traffic jam, a breakdown, a cancelled drop or a late order brings a revised sequence that shows its effect on the other vans. Drivers see nothing until accepted.

Maps, GPS and dispatch links

Maps, trackers and dispatch tools plug in. Address search, live position and trip messages come from the mapping, GPS and dispatch accounts you already hold, and each link is set up and tested with you.

Planned against actual

Plans are compared with real trips. Driven distance, stop times and failed drops are set beside the draft, so the transport manager can see which rules need changing.

Where it gets used

What this looks like in a working week.

A bakery distributor with early vans

Dozens of shops want bread before opening, and the order of stops decides which shop gets crates first and which waits until the shutters are already up. The windows are kept as rules.

A furniture retailer with bulky drops

Sofas and wardrobes need two people and a van with a tail lift, so the planner groups them by vehicle type before choosing an order. Access notes sit with the customer.

A pharmacy wholesaler with urgent orders

One clinic needs insulin today. The shop can wait. Urgent drops are marked and placed first, and the rest of the round is arranged around them so the van still ends near the depot.

A repair workshop visiting customer sites

A repair workshop sends technicians to customer sites, each visit needing certain tools and a customer who is home only in the morning. Visits are grouped by area.

A school bus operator after a road closure

A bridge closes overnight. Each affected bus gets a proposed new pickup order with the moved stops marked. The transport manager approves it before any parent is told.

Who touches it

The people who open this every day.

Dispatcher

Reviews the suggested routes each morning, moves the stops that look wrong and approves the plan before any driver receives a message.

Driver

Gets the day’s ordered list. Marks each stop as done or failed from a phone, and adds a line when the plan did not fit the road.

Transport manager

Compares plans with real trips monthly. Decides which rules need changing, and whether a seventh van would really help or whether a better plan on the six existing vans would do.

Customer service

Reads the expected arrival off the screen. Answers the buyer from the screen, instead of ringing a driver who is carrying a box up two flights of stairs right now.

How it is set up

From first conversation to daily use.

Walk the current planning day
Watch the dispatcher plan one morning. Note every rule they apply from memory, from gate times to the vans that cannot enter the old market, and write each one down.
Prepare addresses and vehicles
Addresses come first, then vehicles. Clean the address book, pin the large customers on the map, then enter each van with its load limit and body type and each depot with its opening hours.
Trial on last month’s orders
Real records, no live trucks. Feed in last month’s orders and compare each draft with the route actually driven, marking where the draft missed a local rule.
Check drafts with dispatchers and drivers
Show the drafts to the dispatcher and a few experienced drivers, and record every correction they make, whether a delivery window, a no-entry lane or a late gate. Each correction becomes a stored rule.
Go live on one depot
Begin with one depot. No driver sees a plan until the dispatcher approves it. The old paper method keeps running beside the software until the drafts stop needing large edits.
Review after live use
Review comes after live use. Compare planned and driven distance, failed drops and the edits dispatchers kept making, then decide which rules to add and whether the mapping and GPS links return usable data.

Measuring it

How you would know it is working.

Record where you stand before it goes live, so the change is measured against your own figures.

Distance travelled is taken from tracker or odometer totals for each van on each round, and the baseline is last month’s totals before the software goes live.
On-time delivery counts drops that arrived inside the customer’s window against all drops, with signed delivery notes or driver marks from the month before go-live giving the starting figure.
Cost per delivery divides fuel, driver pay and vehicle running cost by drops completed, with last month’s books, taken before go-live, as the starting figure.
Failed and repeated drops are counted from driver marks in the same way, so the baseline is the count kept before the software is switched on.
Planning effort is noted as the hours the dispatcher spends on routes each morning, written down for a few days before the trial begins.

Used responsibly

Where a person stays in charge.

The dispatcher signs off every plan

Drafts never reach drivers. Approval of the day’s routes is stored with the dispatcher’s name and the hour. Any later dispute then has a record.

Drivers can depart from the plan

Drivers know things the software does not. The sequence they actually drive is recorded beside the draft, which shows over time where a rule is missing and never counts against the driver.

Data quality limits the draft

Weak input gives weak routes. Poor addresses, guessed weights and trip times that were never written honestly lead to drafts that look neat and fail on the road, so cleaning comes first.

Driver positions are personal data

Tracker data shows where a person is. Positions serve replanning and arrival estimates, are kept only as long as your own policy says, and reach customers only as an expected arrival. Rules on tracking staff vary by country. Your own lawyer agrees the rules with you, and the software offers no legal advice.

How it fits

It is one part of a system, not an island.

A module earns its place by what it passes to the next one. These are the connections we set up most often.

Order management
Vehicle and trip management
Shipping integration
Warehouse management
Customer address book
Driver mobile app or WhatsApp

Already built in

Products that include it.

Built for you

How we would put it in.

Common questions

Questions owners ask about AI route optimisation.

When is route planning worth it?

When a dispatcher spends a long morning on routes, or two people plan the same round differently, it starts to repay the setup effort. A fixed round with a driver who knows every gate needs only a paper list and a working phone.

What if our delivery addresses are poor?

The suggestions will be poor too. Many customers are saved only as a landmark and a phone number, so we clean the address book and pin the biggest customers on the map first.

Can a driver ignore the suggested order?

Yes, a driver can. Drivers know about a locked gate before any software does, and the sequence they actually drive is recorded, which over time shows what they never follow.

Does it use live traffic?

Only if your provider supplies it. Otherwise the software works from your own past trip times, which are often good enough for a fixed area, and the dispatcher can still correct them by hand.

Who changes a route mid-day?

That is the dispatcher’s call. A revised order is proposed with its effect on the other vans shown, and a change reaches the driver only after the dispatcher accepts it.

What systems does it need to connect to?

Orders and vehicles come first. Mapping, a GPS tracker feed and a dispatch tool or WhatsApp group for drivers are optional links, and each depends on an account you already hold or will buy.

How much effort and cost is involved?

Most of the effort goes into cleaning addresses and writing down local rules, and a smaller part into the software setup itself, so early work looks like data work. Cost follows vehicles, depots and links. We quote after seeing your data.

Who needs it most

Businesses that ask for this.

Next to it

Modules that work with it.

Talk to us

Tell us how you handle AI route optimisation today.

A spreadsheet, a register or another system: say which, and we will tell you plainly what is worth changing.

No obligation
A reply within one business day
Your details stay private

Start with your requirements

Seven short steps that tell us what you need. An engineer reads them before getting in touch.