Technician route optimisation with a custom algorithm: cost and gains 2026

The verdict in three sentences

For a team of 25 field technicians, a route optimisation algorithm targets 10 to 20% fewer kilometres and roughly one extra job per day per technician. An off-the-shelf field service tool is enough while your constraints stay simple; a custom module at 20,000 to 60,000 EUR excl. VAT becomes relevant once skills, customer time slots, van stock and SLAs intertwine. Payback is measured in months, provided you start from clean data on actual job durations.

The constraints that make the problem hard

Optimising a route is not drawing the shortest path. The engine must solve a vehicle routing problem with time windows, under several stacked constraints.

Open source solvers (OR-Tools, VROOM, jsprit) handle most of these constraints. The custom work lies in modelling your business rules, integrating with your scheduling tool or ERP, and the technicians' mobile app.

What an optimisation module costs in 2026

These amounts are 2026 ballpark figures. API cost stays marginal: 25 technicians, 6 jobs a day and a distance matrix recomputed several times a day mean a few tens of thousands of requests a month, usually under 200 EUR monthly. For reference, Kolonell's grid for a custom application runs from 2.5M to 4M FCFA (3,810 to 6,100 EUR) for a simple scope, 4M to 7M FCFA (6,100 to 10,670 EUR) for medium complexity, and 7M to 15M FCFA and more (10,670 to 22,870 EUR and more) for a complex project with an optimisation engine.

Mini case study

Illustrative scenario: Claire, after-sales manager of an HVAC maintenance company in Lyon, runs 25 technicians who drive an average of 160 km a day over 220 working days.

• Annual mileage: 25 x 160 x 220 = 880,000 km. At 0.45 EUR per km (fuel, servicing, wear), the line weighs 396,000 EUR.
• 12% gain assumption: 105,600 km avoided, i.e. 47,520 EUR saved per year.
• Time freed: one extra job a day for half the team, 12 x 220 = 2,640 extra jobs, at 90 EUR margin each = 237,600 EUR of potential margin.
• Investment: custom module at 45,000 EUR excl. VAT, plus 8,000 EUR annual maintenance.

On mileage savings alone, the module pays back in just over a year. The extra capacity shortens that to a few months, provided the order book keeps up.

FAQ

What data do you need before starting?

You need at least 6 months of job history with addresses, actual durations and skills used. Without reliable durations, the algorithm plans unrealistic days and adoption drops in the first week.

Does the algorithm handle same-day emergencies?

Yes, dynamic re-optimisation reassigns remaining jobs in a few seconds. Allow 8 to 15 extra development days for this real-time mode.

Do technicians need a mobile app?

It is strongly recommended: the app reports actual start and end times, which improves estimates by 15 to 25% after a few months. A PWA is often enough and costs less than a native app.

What if the dispatcher disagrees with the proposal?

The module should suggest, not impose: the dispatcher keeps control to lock an assignment. In practice, 80 to 90% of suggestions are accepted after 2 to 3 weeks of tuning.

Does GDPR apply to technician geolocation?

Yes, employee geolocation must be proportionate, switchable off outside working hours and disclosed to employee representatives. Position data is usually kept no longer than 2 months unless a specific purpose justifies it.

Let's scope your project. Tell us your technician count, skill constraints and current scheduling tool: we will estimate the budget and expected mileage gain. Detailed quote within 48 h. WhatsApp +221 77 596 93 33.