Real-Time Fleet & Logistics Tracking Platform for 500+ Vehicles

The Challenge

The client was coordinating a 500-vehicle fleet using WhatsApp groups and spreadsheets. Dispatchers had no live visibility, customers had no tracking, and route planning was done manually each morning — taking 3 hours and frequently producing wrong routes.

• Zero real-time visibility into vehicle locations or delivery status
• Manual route planning consuming 3 hours every morning
• Customers calling support for ETAs — 800+ calls per day
• Fuel costs 30% above industry benchmark due to unoptimised routes
• Driver performance tracked only through end-of-day paper logs
• No SLA tracking or proof-of-delivery documentation

Our Strategic Approach

We spent the first two weeks riding along with dispatchers and drivers to understand the real workflow before designing anything. The key insight: dispatchers trusted their gut over any system, so the UI had to surface data as decision support — not replace their judgement.

The Solution We Delivered

A full-stack logistics platform with a real-time web dispatcher console, driver mobile app, customer tracking portal, and a route optimisation engine using Google OR-Tools. All vehicle positions stream over WebSocket with sub-5-second latency.

• Live map dashboard with all 500+ vehicles updated every 4 seconds
• AI route optimisation engine reducing total distance by up to 28%
• Driver mobile app for Android with turn-by-turn navigation and ePOD capture
• Customer self-serve tracking portal with live ETA
• Automated SLA breach alerts via SMS and email
• Digital proof of delivery with photo, signature, and geotag
• Fuel consumption analytics and driver behaviour scoring

Technologies Used

• React — Dispatcher web console with live map and fleet overview
• React Native — Driver mobile app for Android with offline support
• Node.js — API layer, WebSocket server, and route optimisation orchestration
• PostgreSQL + PostGIS — Geospatial data storage and route history
• Google OR-Tools — Vehicle route optimisation with time windows and capacity constraints
• Redis — Real-time vehicle position cache for WebSocket fan-out
• Google Maps Platform — Mapping, geocoding, and turn-by-turn navigation
• AWS — EC2, RDS, ElastiCache, and SNS for SMS alerts

Development Process

1. Field Research & Journey Mapping — Observed dispatchers and drivers for 2 weeks to understand real workflows and pain points
2. Data Architecture — Designed the PostGIS schema for geospatial queries and time-series position history
3. Real-Time Position Streaming — Built the WebSocket pipeline from driver app through Redis fan-out to dispatcher console
4. Route Optimisation Engine — Integrated Google OR-Tools with time window and vehicle capacity constraints
5. Driver App — Built React Native app with offline-first architecture for low-connectivity areas
6. Customer Tracking Portal — Built the self-serve customer portal with live map and push notification ETAs
7. Rollout & Training — Phased rollout across 3 cities with on-ground driver training before full fleet deployment

Results & Impact

The platform rolled out to all 12 cities in 14 weeks. Within 90 days, operational metrics improved dramatically and customer support volume dropped sharply.

• Delivery time per route reduced by 31% on average
• Fuel costs down 22% through optimised routing
• Customer support calls dropped from 800 per day to under 120
• Driver on-time rate improved from 61% to 89%

Project Gallery