Guide
Route optimization: a practical guide for small fleets
Route optimization is the process of deciding which vehicle visits which stops, and in what order, to minimize distance or time while respecting real-world constraints. This guide explains the core ideas without the jargon.
What is route optimization?
Given a set of stops and a set of vehicles, route optimization finds an assignment and sequence that scores best against an objective, such as total distance, total duration or number of vehicles used.
Computer scientists call this the Vehicle Routing Problem (VRP). It is a generalization of the classic traveling salesman problem, and it grows very quickly in difficulty as stops are added, which is why spreadsheets and intuition stop scaling around a dozen stops.
The constraints that matter most
Vehicle capacity
Each vehicle can carry a limited load, measured in units, weight or volume. A plan that ignores capacity is not a plan.
Time windows
Customers expect service within a given period. Time windows turn a geometry problem into a scheduling problem and are often what makes manual planning fail.
Service time
Time spent at each stop counts against the driver's day and changes which stops fit.
Depot and shift limits
Vehicles start and end somewhere, and drivers work finite hours.
Manual planning vs. optimization software
Experienced dispatchers know the territory and build sensible routes. Software is better at evaluating thousands of alternatives consistently and at re-planning quickly when something changes.
The most practical setup keeps the dispatcher in charge: software proposes, the dispatcher reviews and adjusts.
How to start from a spreadsheet
AIRO is designed around exactly this workflow.
- Put one stop per row with a complete address, demand and any time window.
- List your vehicles with capacity and start location.
- Validate the data before optimizing: most bad routes start as bad data.
- Compare the optimized plan against what you would have done manually.
How AI assistants fit in
AI assistants are good at conversation and poor at guaranteed optimization. The safe pattern is to let the assistant call a dedicated optimization engine through a standard such as MCP, so the maths is done by deterministic code. Read more in our guide to MCP for logistics.
Frequently asked questions
Can I optimize routes from an Excel file?
Yes. AIRO is built around the spreadsheet you already use. Upload an Excel (.xlsx) or CSV file with your orders and vehicles, map your columns, fix any validation issues, and run the optimization. You do not need to reformat your file into a rigid template.
Can AIRO handle time windows?
Yes. Time windows are a core constraint in the optimization engine. You provide an earliest and latest arrival per stop, and the engine plans routes that respect them. Stops that cannot be served within their window are reported as unassigned, with the reason, instead of being silently dropped.
Can AIRO optimize multiple vehicles?
Yes. The engine plans multiple vehicles from a single depot, respecting each vehicle's capacity, and balances the work across the fleet. Multi-depot routing is on the roadmap and is not part of the first release.
What is MCP?
MCP (Model Context Protocol) is an open standard that lets AI assistants call external tools securely. In AIRO, MCP exposes actions like "validate import" and "optimize routes" to Claude or ChatGPT. The assistant is the interface; the AIRO backend does the work.
Join the waitlist
Get early access and launch updates. Double opt-in, unsubscribe anytime.