Working Paper

Avoiding a gap between certified and real-world CO2 emissions: Technical considerations for on-board fuel consumption measurements in trucks


The monitoring of on-board fuel consumption metering (OBFCM) data is aimed at preventing a growing gap between the certified and real-world CO2 emissions from trucks. This can only be achieved, however, if regulation mandates a certain accuracy for the OBFCM methods in real-world operation. The first step in achieving this is to determine a reference method against which the on-road accuracy of the OBFCM methods will be validated.

The portable emissions measurement system (PEMS) method is commonly used for the determination of pollutant emissions but is not necessarily the most viable option to evaluate fuel consumption, as it requires the careful alignment of two signals as well as knowledge of the fuel properties. An additional analysis of the test fuel during type approval or in-service conformity testing would be required to make PEMS a viable reference to evaluate the on-road accuracy of OBFCM data. Additionally, the viability of the PEMS method would increase if the monitoring of OBFCM data focused on CO2 rather than fuel consumption.

Direct fuel flow measurements, such as those obtained with a portable fuel flow meter, might be more appropriate as the reference method. Under laboratory conditions, the measurements obtained with the portable fuel flow meter were found to match those obtained with the reference constant volume sampling method within a ±1% accuracy envelope. Consequently, fuel flow meters, which are already required as part of the on-road verification testing procedure for CO2 emissions, could also be used to verify the accuracy of OBFCM data.

The most likely option for OBFCM to date is to monitor the estimates from the engine control unit (ECU) using the vehicle’s on-board diagnostics system. However, for all tested trucks, it presented an estimated 3.5% bias towards underestimating the fuel consumption compared to the measurements obtained with the fuel flow meter. As only the OEMs have access to the data processed by the ECU, there is currently no way to verify whether the ECU accuracy being implemented is the highest possible. Therefore, high OBFCM accuracy requirements of around 1.5% seem to be already achievable with some additional calibration effort by manufacturers. Overall, this would require more transparency from OEMs as to how the ECU signals are being processed.