Case Study · Transportation
Fleet intelligence that flags vehicle problems before they cause downtime
A European transportation company sells a fleet management platform for urban and intercity logistics. AMT built its fleet intelligence system: route optimisation, asset tracking and IoT-driven predictive maintenance that warns fleet managers before small issues turn into breakdowns.
Schedule Technical DiscoveryYour first conversation will be with a senior technical partner, not a salesperson.
Published
At a glance
- Client
- European transportation company specialising in fleet management for urban and intercity logistics
- What AMT built
- Fleet intelligence system with route optimisation, asset tracking, predictive analytics and automated maintenance scheduling
- Data
- IoT sensors, real-time updates through a publish-subscribe pattern
- Apps
- Web front end plus native iOS (Swift) and Android (Kotlin)
- Result
- Less idle time, optimised routes, lower resource use, stakeholders kept up to date on fleet status
The short version
Fleet operators have to balance cost, regulation and day-to-day demands across large fleets. AMT built an intelligent system that optimises routes, tracks assets and predicts vehicle health from IoT sensor data, then schedules maintenance automatically. Real-time updates move through a publish-subscribe pattern on Kafka, so the platform scales with the fleet and everyone sees the same status.
The problem
- Precise vehicle tracking across urban and intercity routes
- Optimal route management to cut idle time and emissions
- Predictive insight into vehicle health, not just breakdown reports
- A design that scales to large fleets and still gives actionable insight
- Support for regulatory compliance and sustainable operations
What AMT built
Route optimisation and asset tracking
An algorithm built to manage large fleets supports real-time decisions on routing and vehicle use.
Predictive maintenance from IoT data
IoT sensors feed data analytics that spot potential vehicle issues early and alert fleet managers before they escalate. Maintenance schedules are generated automatically, which keeps downtime low.
Real-time updates at scale
A publish-subscribe pattern, built on Kafka messaging and Kafka Streams, pushes updates across the platform. It keeps every stakeholder current on fleet status and lets the network scale.
Web and native mobile apps
A web front end for operators, plus native iOS and Android apps.
Results
- Cost-effective operations
- Less idle time and optimised routes, which reduce resource use and emissions
- Better data flow, so every stakeholder sees current fleet status
- Improved operational efficiency and reliable service delivery
Technology stack
| Layer | Technology |
|---|---|
| Front end | AngularJS |
| Back end | JDK 1.8, Spring Boot, Spring Security with JWT, Spring Cloud Netflix, Spring Data JPA with Hibernate, Liquibase, Jasper Reports |
| Messaging | Kafka, Kafka Streams |
| Testing | JUnit, Mockito |
| Database | PostgreSQL |
| iOS | Swift |
| Android | Kotlin |
Before you build something similar
- What do the sensors actually send? Predictive maintenance depends on clean, regular data. Audit what your vehicles report before you design models.
- Who acts on an alert? An alert nobody owns is noise. Route warnings to the person who books the workshop.
- Real time or near real time? Publish-subscribe suits live fleet status. Reports and billing can run on a slower path.
- How long will the stack live? Platforms like this run for years. Plan upgrades of frameworks such as AngularJS and older Java versions as part of the roadmap.