# How Route Optimization Works: A Worked Example for a 40-Stop Day Route optimization takes a list of stops, the vehicles available and the rules each stop imposes, and finds the sequence and assignment that costs the least. Here's how the software actually does it — the problem it is solving, the constraints that make it hard, the heuristics that make it fast — with an example of forty stops across three vans. Source: https://www.highpointgps.com/resources/how-route-optimization-works By Israel Margulies, CEO & Founder — published 2026-09-18, updated 2026-09-18 ## The Problem Under the Software Every route optimizer is solving a version of the vehicle routing problem: given a depot, a set of stops, a set of vehicles and a cost — distance, time or money — find the routes that visit every stop once at the lowest total cost. With one vehicle and no rules it is the travelling salesman problem, which has no efficient exact solution; with forty stops there are more possible orderings than atoms in the observable universe. Real fleets add rules on top: time windows at each stop, vehicle capacity, driver shifts, a stop that needs a liftgate, a customer who will not take deliveries before ten. Each rule removes orderings but makes checking each one harder. The software does not try every ordering. It builds a good solution quickly and then improves it, which is why an optimizer returns an answer in seconds rather than years. ## How It Gets to an Answer Step one is the travel matrix: the drive time between every pair of stops, from a road network with the time of day's traffic, not straight-line distance. Step two is construction — a fast heuristic such as building routes by inserting each stop where it adds the least time, or sweeping stops by angle from the depot. Step three is improvement: the optimizer repeatedly swaps stops between routes and reverses segments within a route, keeping any change that lowers cost without breaking a rule, using methods with names like local search, tabu search or simulated annealing to avoid getting stuck. It stops when improvement stalls or a time limit hits. The result is not provably optimal; it is close enough that no dispatcher would beat it by hand, and it is repeatable. ## A 40-Stop, 3-Van Example Take a Montreal distributor with forty deliveries, three vans, two stops with a 7-to-9 a.m. window, one on the South Shore that needs a liftgate on the one van that has it, and drivers on eight-hour shifts. The dispatcher's hand-built plan, following the same neighbourhoods as last week, runs 412 km and finishes at 5:20 p.m., with one van doing the South Shore at the end of the day after crossing the bridge twice. The optimizer's plan assigns the South Shore stop to the liftgate van first thing, pairs the two early-window stops on the same route, and balances the remaining stops by drive time rather than by neighbourhood. It runs 348 km and finishes at 4:10 p.m., with no window missed. That is a 15% distance saving and an hour of paid time, on one ordinary day. ## Where the Live Data Comes In The plan is only as good as its inputs, and the inputs come from telematics. Actual service time per stop, from the geofence data, replaces the guessed fifteen minutes; actual travel times on the fleet's roads at the fleet's hours replace the map's averages; and during the day, the live position of each van lets the optimizer re-sequence when a customer cancels or a truck is stuck on the Champlain Bridge. Fleets that connect the optimizer to their tracking platform get plans that reflect their own operation, which is the difference between an optimizer that saves 5% and one that saves 15%. ## FAQ ### How does route optimization software work? It builds a travel-time matrix between stops, constructs routes with a fast heuristic, then repeatedly improves them by swapping and reordering stops while respecting time windows, capacity and shift rules. ### Does route optimization find the perfect route? No. The problem is computationally hard, so optimizers find a near-optimal solution in seconds rather than a provably optimal one in years. Near-optimal beats any hand-built plan. ### What constraints can route optimization handle? Time windows, vehicle capacity, driver shifts and breaks, vehicle-specific requirements like a liftgate, priority stops, and depot returns. ### How much does route optimization save? Typically 10 to 20% of distance and drive time versus hand-built routes, more where routes were built by habit rather than by data. ### Why connect route optimization to GPS tracking? Tracking supplies real service times and travel times, and the live positions that let the optimizer re-sequence during the day. ## Sources - [Geotab: Route Optimization](https://www.geotab.com/fleet-management-solutions/route-optimization/) - [Geotab Marketplace: Routing and Dispatch](https://marketplace.geotab.com/)