/
Use case 2: Fuel management for fleet efficiency

Get Involved

◎ COVESA Community Calendar
◎ COVESA Getting Started
◎ Understanding and Engaging in COVESA
◎ Project Launch Process

Groups and Projects

Groups and Projects continued...

◎ Vehicle Signal Specification (VSS)
◎ AOSP App Framework
◎ Commercial Vehicle and Fleet Group
◎ Generative AI Group
◎ Central Data Service Playground

◎ Vehicle Data Model
◎ Cyberecurity Group
◎ In-Vehicle Payments Group
◎ Connected Safety
◎ All Groups and Projects

Resources

◎ COVESA Events
◎ Industry Events 2025
◎ Antitrust and Open Statement
◎ Member Visibility

Join/Sign Up

◎ Join COVESA
◎ Sign Up for Member Newsletter
◎ Request Wiki Account
◎ Request Access to Slack

Use case 2: Fuel management for fleet efficiency

Jun 11, 2025

Optimize fuel consumption by analyzing idle times and driving behaviour in order to lower operating costs and reduce the CO2 emissions of the fleet.

Simulation flow

  1. Simulated vehicles report idle times, speeds, fuel consumption

  2. Cloud aggregates and identifies unnecessary idling based on:

    • Context (e.g., traffic vs. parked)

    • Environment (cold vs. warm start)

  3. Driver receives notification:
    “Your idle time is 23% above average, costing you ~€15/week in fuel and 11kg CO₂. Would you like to enable EcoStart mode?”

  4. Fleet manager sees heatmaps of idling across cities, identifies hotspot areas for rerouting or coaching.

Aspects overview

  • Deployment Aspects

Component

Description

Component

Description

In-Vehicle ECUs

Last state telemetry, idling duration, GPS location, speed, gear status. Basic reasoning

Customer Devices (Mobile App)

Visualizes personal fuel efficiency and receives feedback/coaching

Cloud/Backend Infrastructure

Data persistence (time series, driver profile, vehicle data,..), advanced reasoning 

Cross-Domain Connections

V2C (Vehicle to Cloud), Device2V (Driver App gets live trip feedback), Device2C (Cloud alerts on trend detection)

  • Input Data Layer Aspects

Source

Data

Source

Data

Vehicle Sensors

Engine status, RPM, GPS, idle time, speed, fuel flow

Driver Data

Unique driver ID, preferences (e.g., eco-mode), driving style

External Sources

Traffic congestion zones (e.g., idling at red lights), weather (cold starts)

  • Information Layer Aspects

Component

Role

Component

Role

VSS (Vehicle Signal Specification)

Standardizes all signals: Vehicle.Powertrain.CombustionEngine.IdleDuration, Vehicle.CurrentLocation, etc.

User Profile Abstraction

Abstracts driver IDs with linked behavior history

Bidirectional Data Sync

 

Unified Access API (VISS/Info API)

VISS on-board vehicles, cloud middleware can be VSS compliant.

Time-Series Storage

Fuel and idling logs stored in time-series DB

Schema Generation

VSS-based schema used to define cloud DB schema

  • Knowledge Layer Aspects

Component

Function

Component

Function

Semantic Rules

 

ML Models

 

Symbolic AI

 

Real-time Knowledge Conversion

 

AI Agents

 

  • Wisdom Layer

Component

Role

Component

Role

Driver App

Shows fuel-efficiency score, idling history, behavior improvement suggestions

Fleet Dashboard

Aggregates vehicle-specific and driver-specific fuel reports

Decision Support

System recommends:

  • Route with fewer stop-and-go zones

  • Vehicle-level maintenance if fuel inefficiency persists

  • Driver coaching sessions based on repeated patterns |

 

  • Other Essential Aspects

Area

Application

Area

Application

Vendors

Combines hardware (OEM ECUs), mobile apps, cloud DB, AI toolkits

Security/Privacy

Role-based data access, driver-anonymous behavioral tracking

Scalability

 

Diagnostics

 

Extensibility

Modular: can integrate new sensors or driving behavior types

Interoperability

Unified VSS-based APIs

Multi-Cloud / Edge Support

Pre-processing at edge for live feedback; cloud for batch learning

Efficient Pipelines

 

Industry Alignment