Predictive dispatch for couriers, 3PLs and ecommerce fleets
Smarter routes.
More reliable deliveries.
RoutePilot is a proposed logistics optimisation platform that helps delivery companies plan routes, predict delivery times and manage fleet resources more efficiently — with dispatchers reviewing every recommendation before routes are assigned.
- Distance
- 125 kmvs 187 km by hand
- Late deliveries
- 0vs 2 by hand
- Fuel
- 15 Lvs 22 L
Intelligent route optimisation
Same stops. Same vans. A very different day.
Recommended by RoutePilot
125 km · 0 late · 15.0 LCore product features
Everything a dispatch desk needs.
Scroll sideways →
Route optimisation
Routes built around capacity and delivery windows, then tightened stop by stop.
Arrival-time prediction
Every stop gets a predicted arrival from your own delivery history — with its typical error.
Fleet utilisation
Load and hours per van, so no driver gets the whole city while another is home by ten.
Demand forecasting
Next week’s deliveries per weekday from recent volumes and trend.
Fuel consumption analysis
Kilometres and litres per route, before and after optimisation.
Exception management
Absent customers, refusals and access problems recorded and re-planned.
Dispatch dashboards
Routes, ETAs, utilisation and exceptions on one live board.
The core differentiator
Predictive dispatch, not just shortest paths.
RoutePilot combines expected delivery duration, vehicle capacity, delivery windows and historical demand. A stop at an apartment block with six parcels at rush hour is not the same as a house drop at noon — and the plan knows it.
- 1Expected delivery duration
- 2Vehicle capacity
- 3Delivery windows
- 4Historical demand
What the arrival-time model learned
- Base handover3.17 min
- Per km (off-peak)2.29 min
- Extra per km at rush hour1.11 min
- Per parcel0.85 min
- Apartment access6.12 min
- Office reception3.00 min
Fitted to 250 fictional past deliveries; typical error on 50 it never saw: ±1.6 min.
Next week’s demand
Target customers
Built for fleets that deliver every day.
Courier companies
Hundreds of drops, tight windows and drivers who know the city — supported, not overruled.
Third-party logistics providers
Many clients’ parcels on shared vans, with service levels to prove.
Ecommerce distributors
Volumes that spike on Fridays and promotions, planned ahead.
Decision-makers: fleet managers · logistics executives · operations directors
Business model
Priced by fleet size and delivery volume.
A recurring subscription per vehicle, with volume tiers for larger fleets. Move the slider to estimate.
Illustrative pricing for a proposed product.
Planned AWS infrastructure
How we use Amazon Web Services.
RoutePilot is a proposed AI logistics platform for courier companies, distributors and fleet operators. Its work combines prediction with heavy computation: every morning thousands of addresses must be turned into routes, and every route depends on how long deliveries really take. Designing on AWS lets prediction, mapping, optimisation and delivery records run as managed services that scale with fleet size and delivery volume.
- 01
Import
Amazon S3AWS Lambda
Delivery addresses, vehicle details and historical delivery records land in per-customer storage; Lambda validates them.
- 02
Map & geocode
Amazon Location Service
Addresses are geocoded and road distances and travel times are calculated between stops.
- 03
Predict
Amazon SageMaker AI
Models trained on delivery history predict stop durations, arrival times and next week’s demand.
- 04
Optimise
Amazon ECSAWS Fargate
Containerised optimisation builds routes that respect capacity and delivery windows.
- 05
Dispatch
Amazon DynamoDBAWS Lambda
Approved routes, delivery status and exceptions are stored and updated as the day runs.
Across every stage
- AWS Lambda runs operational workflows
- Amazon CloudWatch monitors imports, optimisation runs and APIs
- AWS KMS encrypts customer delivery data
- Amazon Cognito separates dispatcher and driver roles
Services and their role in RoutePilot
- Amazon SageMaker AI
- Trains and hosts delivery-time, arrival-time and demand-forecasting models.
- Amazon Location Service
- Geocodes addresses, draws maps and calculates road distances and travel times.
- Amazon ECS
- Runs the route optimiser as a containerised service.
- AWS Fargate
- Supplies compute for each optimisation run on demand.
- Amazon S3
- Stores historical delivery records, imports and exported route sheets.
- Amazon DynamoDB
- Holds deliveries, vehicles, routes, status updates and exceptions.
- AWS Lambda
- Runs validation, status updates and exception workflows.
- Amazon Cognito
- Signs users in and separates dispatcher, manager and driver roles.
- Amazon CloudWatch
- Collects logs and metrics for imports, optimisation runs and APIs.
- AWS KMS
- Encrypts delivery data at rest with customer-specific keys.
Planned, not yet deployed. This website and the dispatch planner run in your browser on fictional data.
Founder
John Kenneth
John founded RoutePilot after seeing dispatch teams plan hundreds of drops on a whiteboard each morning — and spend the afternoon apologising for missed windows. John leads product, customers and every pilot.
Progress & demos
See tomorrow’s routes before they happen.
RoutePilot is a proposed product in its building stage. A demo walks through route recommendations, arrival-time predictions and the dispatch board; early pilots start with address imports and vehicle setup for one depot.
- 2026Company founded
By John Kenneth, to make dispatch planning predictive instead of reactive.
- DonePlanning method defined
Delivery-time predictions, capacity- and window-aware routing, dispatcher approval.
- DoneDispatch prototype
Route recommendations, ETAs, utilisation, fuel, exceptions and demand forecast on a fictional day.
- DoneAWS reference architecture
Prediction, mapping and routing, optimisation, storage and workflow layers designed.
- NextFirst demos and pilots
Address imports and vehicle setup for one depot, with dispatchers approving every route.
FAQ
Questions from the dispatch desk.
Do dispatchers lose control?
No. RoutePilot recommends; dispatchers review, change and approve routes before vans are assigned.
What do we need to start?
A list of delivery addresses with windows and parcel counts, your vehicles and capacities, and some delivery history for predictions.
Which maps does it use?
The planned version uses Amazon Location Service for geocoding, maps and road travel times.
Is RoutePilot live?
Not yet. It is a proposed product; the dispatch planner is a prototype running in your browser on fictional data.
Contact
Talk to RoutePilot.
Questions about the product, data imports, maps or partnerships. Messages go directly to the founder.