While it is well known that “MPG will vary” based on how one drives, little independent research exists on the aggregate fuel savings potential of improving driver efficiency and on the best ways to motivate driver behavior changes. This paper finds that reasonable driving style changes could deliver significant national petroleum savings, but that current feedback approaches may be insufficient to convince many people to adopt efficient driving habits. To quantify the outer bound fuel savings for drive cycle modification, the project examines completely eliminating stop-and-go driving plus unnecessary idling, and adjusting acceleration rates and cruising speeds to ideal levels. Even without changing the vehicle powertrain, such extreme adjustments result in dramatic fuel savings of over 30%, but would in reality only be achievable through automated control of vehicles and traffic flow. Considering the effects of real-world driving conditions, efficient driving behaviors could reduce fuel use by 20% on aggressively driven cycles and by 5-10% on more moderately driven trips.
To evaluate potential receptiveness to changing driving habits, the project team conducted a literature survey of driver behavior influences and observed pertinent factors from on-road experiments with different driving styles. This effort highlighted important driver influences such as surrounding vehicle behavior, anxiety over trying to get somewhere quickly, and the power/torque available from the vehicle. Existing feedback approaches often effectively deliver efficiency information and instruction, but do not always do so in an easy way that avoids unintended consequences and helps trump other driving behavior influences. Based on these findings the report details three recommendations for maximizing fuel savings from potential drive cycle improvement: (1) Leverage applications with enhanced incentives, (2) Use an approach that makes it easy and is widely-deployable to motivated drivers, and (3) Utilize connected vehicle and automation technologies to achieve large and widespread efficiency improvements.