Route optimisation is the process of deciding which vehicle delivers which orders, and in what sequence, so the whole day costs as little as possible while still meeting every promise you've made to customers. It sounds simple. With ten drops it is. With two hundred drops, fifteen vehicles and a handful of awkward customers, the number of possible plans is larger than any planner could ever check by hand.
Why it isn't just "shortest distance"
A route that's short on a map can still be a bad route. Real delivery planning has to balance several things at once:
- Capacity. Every vehicle has a payload in kilograms and a number of pallet spaces. Heavy, dense goods hit the weight limit first; bulky goods run out of space first.
- Time windows. A restaurant that needs its delivery before 10:00 has to be visited earlier than a warehouse that's open all day.
- Site restrictions. Some customers can't take an artic, some need a tail lift and some only accept deliveries before 7am.
- Driver rules. The driver must hold the right licence category for the vehicle, and the shift must finish within working-time limits.
- Loading time. Time at each stop matters as much as time on the road.
Mathematicians call this the vehicle routing problem with time windows and capacity constraints. You don't need to know the name, but you should know that a route planner that ignores any of these will hand you routes that look fine and fail on the day.
How optimisers actually work
Most practical optimisers work in two stages. First they build a reasonable plan quickly, often by placing the hardest-to-serve orders first (the most constrained) and inserting each order wherever it adds the least cost. Then they improve that plan with local moves: shifting a drop to another route, swapping two drops, or reversing part of a route to remove crossovers (the "2-opt" move). Each move is only kept if the plan is still valid and cheaper.
The result isn't guaranteed to be mathematically perfect, but it is consistently far better than a hand-built plan, and it's produced in seconds.
What good route optimisation software should give you
- Constraints you can trust. If the planner says a route fits, it should fit: weight, pallets, windows and licences.
- Manual control. Planners know things software doesn't. You should be able to move drops and re-sequence, with the software warning you when you break a rule.
- Live ETAs. A plan is only a forecast. Once the driver is out, ETAs should update after every drop.
- A connection to your orders. Re-keying orders from your ERP is where errors creep in.
- Outputs the warehouse can use. Load sheets in reverse drop order and pick lists save time before the first vehicle leaves.
Where to start
If you're planning on spreadsheets or a whiteboard today, start by recording the constraints you already carry in your head: which customers can't take which vehicles, which drivers are restricted, which deliveries have windows. That list is the difference between a generic route planner and one that works for your fleet.
Fleetnyx's route planner is built around exactly these constraints, with UK licence categories and HGV limits built in. You can see how it fits your operation in a short demo.