Fuel: the invisible cost that slips away
Fuel represents between 25% and 40% of a commercial fleet's total operating costs, according to studies by the American Transportation Research Institute (ATRI). However, most transportation and logistics companies lack real-time visibility into how fuel is consumed, where it is lost, and how much is wasted. This lack of control creates a silent problem with significant economic impact.
Globally, losses from fuel theft in commercial fleets are estimated to represent between 3% and 10% of total consumption. In emerging Latin American markets, where controls are less stringent and routes are longer, this percentage can be considerably higher. Fuel theft is not just a direct loss problem: it also distorts performance indicators, hinders maintenance planning, and affects the profitability of each route.
Types of fuel sensors for fleets
The technological foundation of fuel control is the sensor installed in the vehicle's tank. There are two main types of sensors used in commercial fleets, each with distinct characteristics and applications:
- Capacitive sensors: work by measuring the dielectric constant of the liquid in the tank. A submerged cylindrical probe acts as a capacitor whose capacitance varies according to the fuel level. They are the most common in commercial fleets due to their good balance between precision and durability. They offer accuracy of ±1% to ±2% of total tank volume and are resistant to vibrations and severe operating conditions.
- Ultrasonic sensors: use sound waves to measure the distance between the sensor and the fuel surface. They have no moving parts or direct contact with the liquid, making them especially resistant to corrosion. They are ideal for aggressive fuels or irregularly shaped tanks, although their accuracy can be affected by foam or turbulence in the tank during vehicle movement.
Both sensor types are installed without modifying the tank structure and continuously transmit data to the vehicle's GPS device, which in turn sends the information to the fleet management platform. To learn more about GPS technology, see our articles on Ituran GPS and the GPS guide for fleets in Peru.
Real-time monitoring: how it works
A modern fuel control system integrates three components: the tank sensor, the vehicle's GPS/telematics device, and the centralized management platform. The sensor measures fuel level continuously (typically every 30 to 60 seconds) and sends data to the GPS device, which adds geospatial and temporal context.
On the management platform, this data is visualized as fuel level graphs over time. A normal tank shows a predictable pattern: gradual decreases during operation and abrupt upward jumps during refueling. Any deviation from this pattern — such as a sudden drop without operational justification — is a potential anomaly indicator.
Advanced systems automatically cross-reference fuel data with GPS location, vehicle speed, and engine status to generate a complete context for each event. If a level drop is detected with the engine off and the vehicle stopped at an unauthorized location, the probability of theft is very high.
Anomaly detection algorithms
Fuel theft detection is not limited to observing level drops. Modern algorithms apply multiple layers of analysis to minimize false positives and maximize real detection:
- Drain detection: identifies level decreases exceeding a configurable threshold (for example, more than 10 liters in less than 5 minutes) with the engine off. The system correlates with location to determine whether the vehicle is at an authorized fueling point.
- Route performance analysis: compares actual consumption against expected consumption for each route segment, considering cargo weight, topographic profile, and traffic conditions. Significant deviations trigger investigation alerts.
- Sensor tampering detection: identifies signal patterns suggesting the sensor has been disconnected, short-circuited, or physically altered. This includes prolonged fixed readings, impossible level jumps, and signal loss.
- Long-term trend analysis: compares each vehicle's average consumption week over week and month over month. A gradual increase in consumption without operational changes may indicate mechanical leaks, engine wear, or systematic low-volume theft.
Fuel monitoring return on investment
Fuel control is one of the highest-return investments in fleet management. Industry studies and operational experience consistently document the following benefits:
- Theft loss reduction: fleets implementing fuel monitoring report reductions of 15% to 30% in unjustified losses during the first year of operation.
- Consumption optimization: visibility into consumption patterns enables identifying drivers with excessive fuel use, vehicles with mechanical issues, and inefficient routes. Documented fuel efficiency improvements range from 5% to 15%.
- Administrative cost reduction: automated fuel reporting eliminates manual reconciliation of invoices, vouchers, and fueling tickets. This frees administrative staff for higher-value tasks.
- Improved maintenance planning: a progressive increase in fuel consumption is frequently an early indicator of mechanical problems (dirty injectors, clogged filters, low tire pressure). Monitoring enables intervention before the problem escalates.
On average, companies recover their investment in fuel monitoring systems within 3 to 6 months, depending on fleet size and the magnitude of prior losses.
Integration with GPS and fleet management
The greatest value from fuel control is achieved when integrated with a complete fleet management system. This integration enables viewing fuel levels on the map in real time alongside GPS position, receiving automatic alerts via email, SMS, or platform when an anomaly is detected, generating consumption reports by vehicle, driver, route, and period, comparing fuel efficiency across same-type vehicles to detect outliers, and establishing geofences at authorized fueling stations to validate each refueling event.
The combination of fuel data with vehicle telematics creates a comprehensive operational view that enables data-driven decision-making, not estimates or assumptions.