Real-time dispatch
Your 7am route plan is out of date by 9am. Real-time routing adjusts it all day, as orders land and vans get stuck.
- Free 30-minute call
- No commitment
- You own the code
In short: A real-time route optimisation, also called dynamic vehicle routing, updates the day’s routes while the vehicles are already on the road. New orders, cancellations and traffic delays go straight into the plan, and each driver gets an updated list of next stops.
The problem
At seven in the morning, the vans leave on the plan you made last night. By nine, one driver is stuck behind an accident. A priority order lands with a tight deadline, and three customers have moved their delivery windows. The plan you trusted at breakfast is already wrong. You’re on the phone, and will be for the rest of the morning.
Why a static plan is out of date by nine
Most routing tools plan once, overnight. You feed in the known stops, get a plan back, and everyone follows it until it no longer fits the day. That plan rests on yesterday’s orders, yesterday’s traffic and windows customers keep changing. By mid-morning it’s out of date, and every fix is done by phone. When one van is stuck in traffic, its next three stops are late. You find out when the customers call.
What changes
Today, you patch the plan by phone and squeeze new orders in wherever there’s room. Our solution replans the affected routes as soon as something changes. It slots the new order in where it does the least damage and keeps every time window. You check the proposal and send it out.
Do you need real-time routing at all?
Not every operation does, and vendors rarely say so. Last month’s data will tell you. Count the orders, cancellations and window changes that arrived after the vans had left. If that’s three in two hundred on a normal day, a good overnight plan and a phone will do. If it’s sixty, or your worst days look like that, the morning plan is a guess. That’s when real-time route optimisation earns its setup cost.
Real-time route optimisation: what you get
With real-time route optimisation, you work from a plan that stays current all day. We start with a model of your daily routing, with the same rules for load, time windows and fleet. Around that model we build software that replans again and again during the day. You judge it on three numbers. How many deliveries still arrive on time when something goes wrong? How many extra kilometres does each added order cost? And how fast does the new plan arrive? Before anything goes live, we test it on your own orders, worst days included.
- A model written around your own real-time dispatch rules.
- Results measured in the numbers you already track.
- It slots into how you already work, and you can check every answer it gives.
How it works
We write your depots, your customers, your fleet and your rules into one model. You get the routes that cost the least and keep every rule. You still have the last word: you check the result and send it out.
Static versus dynamic vehicle routing
In planning, the overnight plan is called static VRP. VRP is short for vehicle routing problem. It asks which vehicle visits which stops, and in what order. In static VRP you get one plan, and it stays fixed. Dynamic vehicle routing keeps the model running while new orders arrive and vans fall behind. It slots new jobs into the live routes without throwing the plan away.
The overnight plan doesn’t disappear, though. Practitioners run both: a careful plan before the day starts, then fast fixes during it. During the day, the software only has to adjust what has changed since the morning. That’s why its answers come quickly. A good engine also starts from the routes your drivers know and changes only what it must.
How rolling horizon routing works
The method behind this is called rolling-horizon planning. The horizon is the part of the day the plan looks at, for example the next two hours. It moves forward with the clock, and the system plans that part of the day again and again.
Decisions already in motion stay locked, like the stop a van is driving to right now. Everything still ahead gets optimised again. When a disruption hits, the system re-routes the affected vehicles straight away. The stop a driver is heading to never changes. Only the stops after it can move to a better place in the plan.
New orders go in through an insertion heuristic. A heuristic is a rule of thumb that finds a good answer fast, without searching every possible plan. It tries the new order in every vehicle’s route and at every position in that route. Then it picks the option that changes the plan least. The new plan still keeps every time window and capacity limit.
A real-time delivery dispatch API
To be fast enough during the working day, this needs solid software as well as maths. So we build the optimisation as an API from the start. An API is a connection through which one program asks another for an answer. Here the optimisation engine is a service, and your other systems do the asking.
Your dispatch system, transport management system (TMS) or driver app calls it in real time. A new order triggers a request, and the reply is the adjusted route. The reply arrives within the time limit you set. Live dispatch plugs into the systems you already run, so you don’t get a second screen to watch.
Common questions.
Something else? Ask us directly.
With a static tool, you rebuild the affected routes by hand, often one order at a time. A dynamic system keeps what is already in motion and replans only what lies ahead. The new order lands with the vehicle that can take it at the lowest cost, without missing a time window. You see the proposal and confirm it.
Yes. The plan you calculate overnight does most of the work, because the model has hours to compare many plans. During the day, the live system only adjusts that plan. It looks only at what has changed. That split is what lets it answer fast. Drop the overnight plan, and the live engine starts every day from nothing.
Only if you let it. A good system starts from the routes your drivers already run. In the model, every stop that moves to another vehicle costs a small penalty. So the system changes a route only when the change saves more than that penalty. Stops a van is already heading to stay locked. Drivers get stable days, and the plan still reacts when a van is held up or a new order comes in.
Sometimes it can, and if that’s the case, we’ll tell you. Standard software is built for an average operation, though. Yours runs on its own rules: your contracts, your exceptions, the things you know by heart. A custom model writes those rules in, and that’s where the better plan comes from.
You don’t need either to start. We build the model from the data you’ve already got, even if it lives in three spreadsheets and one person’s head. You get a system you run yourself. We train your team until they no longer need us.
You see a first result with the prototype, before you commit to anything bigger. We start with a free 30-minute call about your operation and the model that would fit it. Then we build a working prototype and run it on your numbers.
Similar pages.
Terms on this page
Every abbreviation explained in plain words, in our article “Metrics and methods”.
- TMSTransport management system
- VRPVehicle routing problem
- Heuristic and metaheuristic
- APIApplication programming interface
Tell us what you’re planning.
In a free 30-minute call with a specialist, you find out whether optimisation can help.
- Free, no commitment
- You talk to a specialist directly
- A clear go or no-go