Autonomous AI Fuel Theft Detection & Dispenser Audit Platform
FuelIQ combines BLE 5.4 capacitive fuel sensors, multi-cloud telemetry AI, and dispenser flow matching to eliminate fuel siphoning and station pump fraud in real time.
LIVE ALERT
Truck #TR-108 | -38.5L Siphoning Detected | Location: Lagos-Ibadan Express | Time: 02:14 AM
The Cost of Unchecked Fleet Fuel Leakage
Fuel accounts for 35%–45% of total operating costs for logistics firms. Fraud and siphoning severely bleed operational margins.
12%
Annual Revenue Leakage
Average percentage of fuel budget lost to driver tank skimming, night siphoning, and illegal resale syndicates.
$120,000+
Annual Loss Per Fleet
Unrecoverable margin lost per 20 heavy-duty trucks operating long-haul routes.
Zero Proof
False Alarm Problem
Standard telematics fail to distinguish between fluid sloshing on inclines and actual siphoning, generating excessive false alarms.
Intelligent Fluid Telemetry Meets Edge AI
BLE Capacitive Probes
0.1mm volumetric fluid measurement with internal tilt-stabilization hardware.
Dynamic Slosh Filtering
Machine learning algorithms separate sloshing on steep inclines from true theft events.
Pump POS Reconciliation
Cross-checks fuel station payment receipts directly against actual liters added to the vehicle tank.
AI TECHNOLOGY
Multi-Model Telemetry Intelligence
FuelIQ's AI engine combines recurrent neural networks, dynamic time warping, and isolation forest anomaly detection to analyze volumetric fuel data in real time — distinguishing genuine siphoning events from motion noise with 98% accuracy.
Data Ingested
1Hz BLE capacitive probe readings (volumetric fluid level)
CAN-bus engine load and RPM signals
GPS speed, altitude, and route coordinates
Accelerometer tilt vectors (3-axis)
POS fuel station transaction receipts
Historical fuel consumption baselines per vehicle
What the AI Detects
Real-time siphoning events (gradual and rapid extraction)
Tank skimming during overnight parking
Dispenser calibration tampering at fuel stations
Volumetric anomalies vs. motion-induced sloshing
Route-deviation fuel theft (corridor siphoning)
Transaction fraud (payment for fuel not received)
BLE 5.4 Capacitive Probe Specs
0.1mm volumetric measurement resolution
Internal tilt-stabilization (3-axis accelerometer)
1Hz telemetry transmission frequency
IP67 waterproof rating for fuel tank submersion
Encrypted BLE 5.4 communication with 100m range
Over-the-air firmware updates
Models & Techniques
LSTM/GRU Recurrent Neural Networks — time-series volumetric forecasting for expected fuel level prediction per route segment
Dynamic Time Warping (DTW) — cancels vehicle motion noise by aligning accelerometer patterns with fuel level signals
Isolation Forests — unsupervised anomaly detection for transaction records and dispenser calibration drift
Gradient Boosted Trees — route-level fuel consumption prediction incorporating weather, load, and terrain data
Why GPU/Cloud Infrastructure Is Needed
Real-time inference: Sub-15-second alert dispatch requires low-latency model serving via NVIDIA Triton Inference Server
Training at scale: LSTM models trained on millions of time-series data points using NVIDIA RAPIDS and CUDA acceleration
Stream processing: Apache Kafka and Flink handle 50,000+ telemetry events/second from fleet sensors across multiple countries
Multi-cloud deployment: Azure IoT Hub for device management, AWS Lambda for serverless alert dispatch, ensuring 99.9% uptime
Scalability: Cloud infrastructure auto-scales to support fleet growth from 420 to 10,000+ vehicles without re-architecture
HOW IT WORKS
From Sensor to Alert in Under 15 Seconds
FuelIQ automates the entire fuel monitoring pipeline — from hardware installation to real-time AI-powered theft detection and instant alert dispatch.
Install BLE Probes
Our BLE 5.4 capacitive sensors are installed inside each vehicle fuel tank. Installation takes under 30 minutes per vehicle with zero downtime.
Stream Telemetry
Sensors transmit 1Hz volumetric data, tilt vectors, and CAN-bus signals via EMQX MQTT brokers to our cloud ingestion pipeline.
AI Analyzes Patterns
LSTM/GRU networks forecast expected fuel levels. Dynamic Time Warping cancels motion noise. Isolation Forests flag anomalies in real time.
Alert & Audit
Within 15 seconds of detection, alerts are dispatched via SMS and WhatsApp. Full audit logs are generated for law enforcement and insurance.
Live Fleet Dashboard
Real-time monitoring of 42 vehicles across 6 transit corridors
Fuel Today
4,120 L
Theft Prevented
$1,450
Alerts
1
Fuel Level — TR-108
82%
TECHNOLOGY & INFRASTRUCTURE
Enterprise Multi-Cloud Architecture
Our platform uses cloud infrastructure for secure data storage, API hosting, AI model inference, real-time alerts, analytics, and user management. We leverage Azure for IoT device management, AWS for serverless compute and AI services, and NVIDIA for GPU-accelerated model training and inference.
Microsoft Azure
Primary Cloud
Azure IoT Hub — Device provisioning, telemetry ingestion, and firmware OTA updates for 420+ BLE sensors
Azure Event Hubs — High-throughput ingestion handling 50,000+ telemetry events/second
Azure Digital Twins — Real-time fluid tank graph modeling and fleet state representation
Azure Entra ID — RBAC security, SSO, and enterprise identity management
Azure Key Vault — Secrets management for API keys, certificates, and connection strings
Azure Cosmos DB — Globally distributed telemetry storage with millisecond read latency
Amazon Web Services
AI & Serverless
Amazon S3 & RDS — Cold storage for historical telemetry and relational fleet configuration data
AWS Lambda/ECS — Serverless alert dispatch, webhook processing, and scheduled report generation
Amazon Bedrock — FuelIQ Copilot natural language query agent for fleet managers
Amazon CloudWatch — Infrastructure monitoring, custom metrics, and operational dashboards
AWS CloudFront — Global CDN for dashboard static assets and API caching
NVIDIA Acceleration
AI Compute
NVIDIA RAPIDS & CUDA — GPU-accelerated time-series model training on fleet telemetry datasets
NVIDIA Triton Inference Server — Real-time telemetry scoring with sub-100ms inference latency
NVIDIA TensorRT — Model optimization for reduced inference cost and faster predictions
GPU-accelerated anomaly detection — Isolation Forest training on millions of transaction records
Future: NVIDIA Jetson — Edge AI inference for on-device theft detection at remote mining sites
IoT & Streaming Layer
Edge & Stream
BLE 5.4 capacitive probes — 0.1mm volumetric resolution with tilt-stabilized hardware
EMQX Enterprise MQTT Brokers — Device-to-cloud messaging with TLS 1.3 and mutual authentication
Apache Kafka — Event streaming backbone for real-time telemetry pipeline
Apache Flink — Stream processing for windowed aggregations and complex event detection
Apache Avro — Schema registry for telemetry data serialization and versioning
WHY NOW
Fuel Fraud Is a $30B Global Problem That's Only Getting Worse
Africa's logistics sector is projected to grow 8% annually through 2030, yet fuel theft costs operators 12–18% of their total fuel budget. As fleets scale across transit corridors from Lagos to Nairobi, traditional telematics — designed for mileage tracking, not fluid telemetry — fail to detect sophisticated siphoning operations. The convergence of cheap BLE sensors, edge AI inference, and multi-cloud telemetry now makes real-time fuel monitoring economically viable for the first time.
8%
Annual logistics sector growth across Africa
12–18%
Fuel budget lost to fraud and siphoning
0
Existing telematics platforms that monitor fluid volumetrics
Built for High-Stakes Fuel Operations
Long-Haul Logistics & Freight Fleets
Prevent fuel theft on transit corridors. Monitor every liter from depot to destination with real-time volumetric tracking.
Fuel Station Operators & Networks
Catch dispenser calibration tampering and pump fraud. Verify every transaction against physical fuel dispensed.
Mining & Construction Fleet Directors
Track off-grid heavy machinery fuel tanks. Maintain visibility over remote operations where traditional telematics fail.
MARKET OPPORTUNITY
Targeting Africa's Fast-Growing Fleet Economy
Africa manages over 2.5 million commercial fleet vehicles. Fuel represents 35–45% of operating costs, yet none of the dominant telematics providers (Geotab, Samsara, Cartrack) offer volumetric fluid monitoring with AI-powered theft detection.
Target Users
Long-haul logistics & freight fleets
Fuel station operators & networks
Mining & construction fleet directors
Oil & gas distribution companies
Government fleet management agencies
Launch Markets
Nigeria (Lagos, Abuja, Kano) — Live
Ghana (Accra) — Q4 2026
Kenya (Nairobi) — Q1 2027
South Africa (Johannesburg) — Q2 2027
Rwanda (Kigali) — Q3 2027
Revenue Model
SaaS subscriptions ($15–$29/vehicle/month)
Hardware probe sales (one-time per vehicle)
Enterprise custom plans (dedicated cloud tenants)
API access for fleet management integrations
Insurance verification data partnerships
Measurable Impact
98%
Theft Detection Accuracy
Eliminates false alarms through tilt-stabilized AI filtering.
15s
Instant Dispatch
Immediate WhatsApp/SMS alerts directly to dispatch security.
8%
Total Fuel Cost Savings
Immediate recovery of operating margins.
100%
Automated Audit Logs
Instant proof generated for law enforcement and insurance claims.
Traction & Early Results
420+
Long-haul trucks monitored across major African transit corridors.
$34,000+
In fuel loss prevented across 90-day active fleet pilots in Lagos and Abuja.
1,200+
Commercial vehicles queued on waitlist for Q4 hardware probe installation.
Product Roadmap
BLE Capacitive Sensor Integration
BLE capacitive sensor integration & EMQX telemetry ingestion pipeline.
Time-Series AI Model & Dashboard
Time-series slosh-filtering AI model launch & Web Dashboard.
Real-Time Event Streaming
Kafka real-time event streaming & Azure IoT Hub enterprise sync.
NVIDIA Inference & Bedrock NLQ
NVIDIA Triton inference optimization & Amazon Bedrock natural language query launch.
Direct POS API Integration
Direct POS API integration with Wayne and Gilbarco fuel dispensers globally.
Transparent Pricing
Scale your fuel protection as your fleet grows.
Starter
ProMost Popular
Enterprise
ABOUT US
Leadership & Engineering Team
FuelIQ was founded by a team of fleet operations, AI, and embedded systems experts with decades of combined experience across logistics, telematics, and hardware engineering.
Ehirim David Chukwuebuka
CEO, Co-Founder
Ex-Logistics Operations Director, 11+ yrs in commercial fleet management overseeing 200+ vehicle fleets across West Africa.
Dr. Aris Thorne
Lead AI & Data Architect
Ph.D. in Signal Processing & Time-Series Analytics. Former researcher at NVIDIA AI Labs with expertise in LSTM networks and anomaly detection.
Request a Live Demo
See FuelIQ in action. Our team will walk you through the platform with your fleet data.
Thank You!
Your demo request has been submitted successfully. Our team will reach out within 24 hours to schedule your personalized walkthrough.
What to Expect
30-min Discovery Call
We learn about your fleet operations and fuel loss pain points.
Live Platform Walkthrough
See real-time telemetry, AI alerts, and dashboard analytics.
Free Pilot Program
Install probes on 3–5 trucks and measure fuel savings over 30 days.
Custom Pricing Proposal
Tailored plan based on fleet size and integration requirements.
CONTACT
Get in Touch
Have questions about FuelIQ? Our team is ready to help.
Technical Support
Existing customer? Reach our support team for platform assistance.
support@fueliq.techHeadquarters
FuelIQ Technologies Ltd. 17 Ramoni Jimoh Street, NNPC Lagos, Nigeria
+234 913 429 2789