Working Paper

Measuring in-use fuel economy: Summary of pilot studies

Vehicle fuel economy is determined via a type-approval or certification process, which involves testing vehicles under laboratory conditions. But the information obtained via chassis testing under laboratory conditions cannot fully reflect real-world driving conditions over a vehicle’s lifetime. The consequent discrepancies between real fuel economy and laboratory test results can have a significant impact in estimating real-world fuel consumption and greenhouse gas emissions.

Comprehensive real-world information is needed to assess fuel economy label adjustments, properly determine off-cycle credits for fuel economy standards, evaluate the actual effectiveness of different technologies, and estimate real total fuel consumption. And that need has increased over the last decade, which has seen significant developments in power train technologies (e.g. gasoline direct injection, hybrids, plug-in hybrids) and changing fuel quality to accommodate renewable fuel sources (primarily ethanol, in E10 and E15 blends).

Gathering in-use fuel-economy data on a large set of vehicles in operation over a significant period of time requires a method for capturing that information. The most cost-effective alternative is to use the on-board diagnostics (OBD) data stream to gather information either directly reporting fuel consumption rate or from which fuel rate can be calculated or estimated.

ICCT recognized that before a major OBD-based study of fuel economy characteristics could be undertaken, logistical and technical issues had to be addressed. ICCT therefore contracted with Eastern Research Group (ERG) in the US and TÜV NORD Mobilität (TNM) in Germany to conduct two pilot studies, identifying areas of concern and potential solutions in four areas: vehicle sample structure and size, vehicle recruitment methodology, data logger evaluations, and estimated project cost. This paper summarizes the results of those studies.