use and GHG emissions.3 The optimistic and midrange estimates reflect the committee’s appraisal of the overall development challenges facing the general pathways, and the promise of the various technologies that might be employed to meet the challenges. These estimates do not consider issues of market acceptance, which are addressed in Chapters 4 and 5, and are not based on specific policies to encourage market acceptance. Both estimates assume that policies are adopted that are sufficiently effective to overcome consumer and infrastructure barriers to adoption.

The committee reviewed a wide range of studies on technology potential and cost but was not able to find a study based on up-to-date technology assumptions and a consistent methodology for all types of technologies through 2050. The 2017-2025 light duty fuel economy standards were based on analyses that included major improvements in data and estimation of technology benefits and costs, but assessed technology only through 2025 (EPA and NHTSA, 2011). The 2009 MultiPath study (ANL, 2009) used a consistent methodology through 2050, but it lacked this recent data. Thus, the committee performed its own assessment of technology effectiveness and costs, as described below and in Appendix F.

In order to compare technologies, all costs discussed in this chapter assume the economies of scale from high volume production even in the early years when production is low. The modeling in Chapter 5, which estimates the actual costs of following specific trajectories, modified these costs for early and low-volume production.

Great care was taken to apply consistent assumptions to all of the technologies considered. For example, the same amount of weight reduction was applied to all vehicle types, and vehicle costs were built up from one vehicle type to the next (e.g., hybrid costs were estimated based on changes from conventional vehicles, and PEV costs were based on changes from hybrid vehicles). This approach does not reduce the large uncertainty in forecasting future benefits and costs, but it does help ensure that the relative differences in costs between different technologies are appropriately assessed and are more accurate than the absolute cost estimates.

The committee made every attempt to ensure accurate technology assumptions. Fundamental limitations for all technologies were considered for all future assessments, such as the ones discussed below for lithium-ion (Li-ion) battery chemistry and for engine losses. As these limits were approached, the rate of technology improvement was slowed down to ensure that the estimates stayed well short of the limits.

On the other hand, learning occurs primarily because manufacturers are very good at coming up with better and more efficient incremental improvements. For example, 10 years ago technology that uses turbochargers to boost exhaust gas recirculation (EGR) was virtually unknown for gasoline engines. This new development, enabled by sophisticated computer simulations and design, has the potential to improve overall ICEV efficiency by about 5 percent. Certainly some of the currently known technologies will not pan out as planned, but it is equally certain that there will be incremental improvements beyond what we can predict now. The estimates in this chapter reflect an effort to strike a careful balance between these considerations.

Learning also applies to cost. Historically, technology costs have continuously declined due to incremental improvements. For example, 6-speed automatic transmissions, currently the most common type, are cheaper to manufacturer than 4-speed automatic transmissions, thanks to innovative power flow designs that allow additional gear combinations with fewer clutches and gearsets.

Although significant continuing R&D yielding sustained progress and cost reduction in all areas is essential, the technology estimates used for the committee’s analyses do not depend on any unanticipated and fundamental scientific breakthroughs in batteries, fuel cell systems, lightweight materials, or other technologies. Therefore the estimates for improvements may be more readily attained, especially for 2050, when technology breakthroughs are quite possible. For example:

  • Batteries beyond Li-ion were not considered for PEVs because the challenges facing their development make their availability highly speculative.
  • Fuel cell efficiency gains were much less than theoretically possible, based on the assumption that developers will consider reducing the cost of producing a given power level to be more important.
  • Reducing weight with carbon fiber materials was not included in the analyses, because the committee was uncertain if costs would be low enough by 2050 for mass market acceptance.
  • The annual rate of reduction for the various vehicle energy losses was assumed to diminish after 2030, usually to about half of the historical rate of reduction or the rate projected from 2010 to 2030. This reflects reaching the limits of currently known technology and implicitly assumes that the rate of technology improvements will slow in the future, despite the current trend of accelerating technology introduction.
  • Only turbocompounding was considered for waste heat recovery, even though other methods with much


3The committee did not assess GHG emissions from the production of vehicles or include such emissions in its analyses of emissions trends later in this report. Given that vehicles are expected to last about 15 years, any differences in production emissions will not make a large difference in lifetime emissions. In addition, data on emissions from the production of vehicles is poor, and estimates for advanced vehicles in several decades will be even more uncertain.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement