Deciding which vehicle takes which drops in which order — and the constraints that make the obvious answer wrong.
Route planning determines how orders are assigned to vehicles and in what sequence they are delivered, within constraints: vehicle capacity, delivery windows, driver hours, access restrictions and service times. The objective is usually lowest cost or distance subject to meeting every window.
Routing by hand or by habit produces routes that look sensible and are usually ten to twenty per cent longer than they need to be, because a person cannot evaluate the combinatorics. The constraints are what make it hard: a plan that ignores delivery windows or access restrictions is optimal on paper and fails on the road.
Vehicle capacity and type, delivery windows, access restrictions, driver hours and shift patterns.
Time at each drop, which varies by customer and is the figure most often wrong in a plan that fails.
Orders assigned to vehicles and sequenced, respecting every hard constraint and minimising the chosen objective.
The plan goes to drivers; actual progress is tracked against it so exceptions surface during the day, not after.
Where actual consistently exceeds planned time, the underlying assumption is wrong and should be corrected.
A business implements route optimisation and sees planned route times fall by 18 percent. Actual times barely move, and drivers report that the plans do not work.
The model was missing constraints that drivers held in their heads. Three customers accept deliveries only before eleven. One industrial estate has a height restriction excluding the larger vehicles. Two city-centre stops require a particular unloading bay that is occupied at lunchtime. Service time at a hospital is forty minutes, not the ten assumed for everything.
Optimisation is only as good as the constraints it knows about, and the constraints that matter most are usually undocumented because experienced drivers simply work around them. Capturing them is unglamorous interview work rather than a software task. Once service times were measured per customer type and access constraints recorded, the planned savings became real — smaller than the original 18 percent, and actually achievable.
| Term | What it means |
|---|---|
| Stop | A single delivery or collection point on a route. |
| Time window | The period during which a customer will accept delivery. |
| Vehicle constraint | Capacity, weight, dimensions or access restrictions limiting what a vehicle can do. |
| Drop density | Stops per unit area, which is the main determinant of route efficiency. |
| Dynamic routing | Re-planning during the day as conditions change. |
A supermarket delivery and a city-centre shop take very different times. One average makes every route wrong.
Vehicle size limits, timed access and unloading constraints invalidate otherwise optimal routes.
Without feedback, the same wrong assumptions are planned against indefinitely.
Distance is a proxy. Cost includes time, driver hours and vehicle type, which can rank routes differently.
| Metric | What it tells you | What distorts it |
|---|---|---|
| Planned vs actual route time | Model accuracy. A persistent gap means constraints are missing. | |
| Drops per hour | The core productivity measure, driven mostly by drop density and service time. | |
| Time window compliance | Deliveries made within the agreed window, which is what customers judge. | |
| Vehicle fill rate | Capacity utilisation by weight and by volume, which can differ substantially. |
We build and support the systems this page describes. Start with a free working proof of concept — no cost, no commitment.