Skip to main content

Fleet Management System

Vehicle, trip, driver, fuel and maintenance management with GPS and telematics integration, a mobile app for field workflows, and analytics built for fleet cost and reliability decisions.

Client
Confidential enterprise client
Sector
Enterprise
Capabilities
digital platformsintegrationdata ai

The problem

Running the fleet was a manual exercise. Vehicles were booked and allocated by hand, and nobody could see availability without asking. Maintenance tracking was weak, fuel records were missing or incomplete, and trip approvals moved at the speed of whoever happened to be reachable. Which driver had which vehicle, and when, was difficult to establish after the fact.

Reporting was the sharper gap. Operational reporting was limited, and the analytics needed to make decisions did not exist at all. The questions the organisation could not answer were specific ones:

  • Which vehicle models break down most often?
  • Which models cost the most to maintain?
  • How do newer and older vehicles compare on maintenance cost?
  • Which maintenance and service vendors are most reliable?
  • What cost and performance trends are emerging across the fleet?

Those are the questions that decide what to buy, what to retire and who to send work to. Without them, fleet decisions were being made on impression.

Constraints

Integration with providers we did not control. The system had to work with GPS providers, telematics providers and other external systems, each with its own availability, its own data shape and its own rate of change -- and with limited access on our side to work against.

Client-controlled infrastructure. The application runs inside the client's own infrastructure under strict security restrictions, which shaped how we could reach the production environment, deploy into it and work within it.

Stakeholder availability. Decision-making was slow and key users had limited time, which lengthened every round of clarification, validation and approval.

There was no significant data-migration or data-quality problem on this engagement, which is worth stating plainly rather than implying a difficulty that was not there.

Our role

Cosmic Limit was responsible for requirements analysis, solution architecture, backend development, web frontend development, mobile application development, database design, systems integration, testing, application deployment, and ongoing system maintenance.

As with our other work of this shape, the underlying infrastructure is provided and managed by the client's own infrastructure team. We build and maintain the application; we do not run the environment beneath it.

The solution

The core system is a modular monolith supported by a small number of focused microservices, including a notification service. The functional areas are vehicle inventory, trip requests and approvals, driver management, vehicle assignments, fuel tracking, maintenance management, vehicle inspections, GPS and telematics integration, vendor management, notifications, and fleet analytics and reporting.

Analytics and data-driven decision-making are a core design objective, not a reporting screen added at the end. Operational data -- what each vehicle costs, how often it fails, which vendor serviced it, how much fuel it consumed -- is modelled so that it connects to the decisions that use it: maintenance cost, vehicle reliability, fleet age, fuel usage, vendor performance, and the broader cost and operating trends across the fleet. The design question throughout was not "what should this screen show" but "which decision does this data have to support".

The mobile app is where field operations actually happen. The React Native application covers trip requests and approvals, driver workflows, fuel entry, vehicle inspections, maintenance updates and notifications -- the work done beside a vehicle rather than at a desk. Capturing fuel and inspection data at the point of the event, rather than transcribing it later, is a large part of why the records are usable for analysis at all.

GPS and telematics integration is part of how the picture is assembled, feeding vehicle position and operating data into the same model as everything else.

Implementation is .NET for backend services and business logic, Next.js for the web frontend, React Native for mobile, and PostgreSQL as the primary relational database. RabbitMQ carries asynchronous communication and background processing, Redis provides caching, and the whole application is containerised with Docker and runs on Linux.

Outcome

No numerical results are published here. Fleet sizes, savings and uptime figures have not been verified for publication, and inventing plausible ones would undermine the parts of this account that are exact.

What changed: vehicle availability is visible rather than reconstructed, and trip approvals move through the system instead of through whoever is reachable. Fuel accountability improved because fuel is recorded where it is used, and maintenance planning improved because maintenance is tracked against the vehicle. Drivers and vehicle assignments can be followed. Vendor performance became comparable rather than anecdotal. And the fleet analytics answer the cost and reliability questions that prompted the engagement -- which vehicles are expensive, which are unreliable, and which vendors are worth keeping -- from operational data rather than from impression. Manual work fell and record keeping improved as a consequence of all of it.

Constraints

What shaped the solution

  • integration

    The system integrates with GPS providers, telematics providers and other external systems that we had limited access to -- each with its own availability, data shape and rate of change.

  • infrastructure

    The application runs inside client-controlled infrastructure with strict security restrictions, which shaped how we could reach the production environment and work within it.

  • stakeholder

    Decision-making was slow and key users had limited availability, which lengthened every round of clarification, validation and approval.

Technology

What was used, and for what

framework

  • .NET

    Backend services and business logic.

  • Next.js

    Web frontend.

  • React Native

    Mobile application for field and operational workflows -- trip requests, driver tasks, fuel entry, inspections and maintenance updates.

datastore

  • PostgreSQL

    Primary relational database.

  • Redis

    Caching.

integration

  • RabbitMQ

    Asynchronous communication between services and background processing.

infrastructure

  • Docker

    Containerisation and application deployment.

platform

  • Linux

    Production runtime environment.

Outcome

What changed

Visibility of the fleet

Vehicle availability, driver assignments and vehicle status can be seen rather than reconstructed, which is what allocation decisions were previously waiting on.

Faster trip approvals

Requests and approvals run through the system, including from the mobile app, instead of depending on whoever is reachable at the time.

Fuel accountability and maintenance planning

Fuel is recorded at the point it is used and maintenance is tracked against the vehicle, so planning works from the record rather than from recollection.

Vendor performance in view

Service and maintenance work is attributed to the vendor that performed it, which makes reliability and cost comparable across vendors.

Analytics that answer fleet questions

Which models break down most often, which cost most to maintain, how newer and older vehicles compare, and which cost and reliability trends are emerging -- answerable from operational data rather than estimated.

Less manual work, better records

Booking, allocation and record keeping moved out of manual process into the system, which is where the reduction in effort comes from.

Our role

What we were responsible for

Cosmic Limit's responsibilities on this engagement
  1. Requirements analysis and solution architecture

  2. Backend development and database design

  3. Web frontend and React Native mobile application development

  4. Integration with GPS, telematics and other external systems

  5. Testing and application deployment

  6. Ongoing system maintenance

Anonymised by owner decision recorded in the launch content handoff of 2 September 2026. No screenshots or diagrams are published: any future image needs sanitising and a confidentiality review first (spec 22.2, 26).

Facing something similar?

Tell us the constraint you are working against and we will tell you how we would approach it.