Courier & Delivery Management

Part of the worlds-biggest-software-project initiative.

An open-source, AI-augmented platform for courier and last-mile delivery operations -- from order intake and route optimisation through dispatch, real-time tracking, proof of delivery, and driver settlement.

Courier and delivery businesses coordinate hundreds to thousands of consignments daily across networks of drivers, vehicles, and zones. Existing platforms are proprietary, expensive, and force operators to stitch together separate tools for billing, settlement, and warehouse operations. This project aims to deliver a self-hostable, open-source alternative that unifies the entire delivery lifecycle and uses machine learning to improve dispatch decisions, predict delivery failures, and automate exception handling.

Why Courier & Delivery Management?

No complete open-source option exists. Fleetbase is the only open-source platform in the space, but it lacks mature billing, driver settlement, and advanced route optimisation. Every other serious tool (Onfleet, CXT Software, Dispatch Science, Track-POD) is proprietary SaaS with opaque pricing.
Fragmented workflows increase cost and errors. Most platforms require third-party integrations for accounting, invoicing, and driver payroll. Dispatchers juggle multiple systems, and data falls through the cracks between them.
Pricing scales steeply with fleet size. Commercial tools charge per driver per month, with add-on fees for notifications, analytics, and API access. A 50-driver operation can easily spend thousands monthly before accounting integrations are factored in.
AI capabilities remain shallow. Incumbents offer basic rule-based auto-dispatch and distance-based ETAs. Predictive failure scoring, demand forecasting, and natural-language dispatch interfaces are absent or nascent across the market.
Vendor lock-in limits operational flexibility. Proprietary platforms control data, integrations, and feature roadmaps. Operators cannot self-host, customise routing constraints, or extend the system without vendor approval and additional cost.

Key Features

Order Intake & Management

Multi-channel order intake via web portal, REST API, and EDI
Automatic validation and rate calculation on submission
Customer self-service portal for order placement, tracking, and invoice viewing
Full chain-of-custody barcode scanning and parcel management

Dispatch & Route Optimisation

Manual, assisted, and fully automated dispatch modes
Multi-stop route optimisation using open-source VRP engines (VROOM, Google OR-Tools) with time-window, capacity, and driver-break constraints
Dynamic re-routing in response to traffic, cancellations, and additional pickups
Live dispatch board with all routes and drivers visible on a single map

Driver App & Proof of Delivery

Mobile app (iOS/Android) with offline mode and automatic sync on reconnect
Electronic proof of delivery: signature capture, photo evidence, barcode scan, recipient name
GPS position reporting for real-time tracking by dispatchers and customers
Exception codes and failed-delivery workflow with re-delivery scheduling

Customer Notifications & Tracking

Automated SMS and email alerts at key milestones (accepted, en route, ETA, delivered, failed)
Live customer tracking page with estimated arrival time updates
White-label notification support preserving the operator's brand

Settlement, Billing & Invoicing

Driver settlement with configurable rate structures (per-drop, commission, zone-based)
Consignment-level billing with zone, weight, and service-type rate tables
Invoice generation and accounts-receivable integration
Accounting system export (QuickBooks, Xero, and similar)

Analytics & Reporting

On-time delivery rates, driver performance, and route efficiency dashboards
Planned vs. actual time reporting at the task level
Customer-level volume and revenue analysis
Fuel consumption and cost tracking

AI-Native Advantage

Machine learning replaces the rule-based heuristics used by incumbent platforms across several critical operations. Dynamic dispatch models learn from historical driver performance, customer preferences, and real-time conditions to make smarter assignment decisions. Failure probability scoring uses address, time-slot, and customer history data to flag high-risk deliveries before they leave the depot, enabling proactive intervention. ETA prediction incorporates weather, driver pace, stop complexity, and live traffic rather than relying on simple distance calculations. Exception triage applies NLP to driver notes to automate re-delivery scheduling and customer communication, and anomaly detection flags billing irregularities before invoices are issued.

Tech Stack & Deployment

Route optimisation built on open-source VRP solvers (VROOM, Google OR-Tools) for transparent, extensible routing logic
Real-time communication via WebSocket or push-notification infrastructure for low-latency driver position updates and dispatch instructions
Self-hosted or cloud deployment with multi-tenant and white-label support (custom portal domains and branding)
API-first architecture with REST APIs and webhooks for all order and route lifecycle events
E-commerce connectors for Shopify, WooCommerce, and marketplace operators
EDI integration (X12/EDIFACT) for enterprise-tier connectivity
Horizontal scalability to handle peak-volume surges (e.g. Black Friday) without degradation

Market Context

The courier and last-mile delivery software market is mature and expanding, driven by e-commerce growth and consumer expectations for real-time tracking. Leading commercial platforms (Onfleet, CXT Software, Dispatch Science, Track-POD, Route4Me) operate on subscription models that scale with fleet size, often reaching thousands of dollars monthly for mid-size operations. Primary buyers are courier companies, 3PL providers, and e-commerce businesses running their own delivery fleets, ranging from 10-driver local operations to 1,000+ driver enterprises.

Project Status

This project is in the research and specification phase. Contributions, feedback, and domain expertise are welcome.

Contributing

We welcome contributions from developers, domain experts, and potential users. See CONTRIBUTING.md for guidelines.

Important: All contributions must be your own original work or clearly attributed open-source material with a compatible licence. Copyright infringement and licence violations will not be tolerated and will result in immediate removal of the offending contribution. If you are unsure whether a piece of code, text, or other material is safe to contribute, open an issue and ask before submitting.

Licence

Licence to be determined. See discussion for context.