Autonomous Vehicles: Uncertainties and Energy Implications For Fourth International Transport Energy Modeling (iTEM4) workshop October 30, 2018 | Laxenburg, Austria By John Maples, Team Lead Nicholas Chase, Lead Economist U.S. Energy Information Administration Independent Statistics & Analysis www.eia.gov
Overview • Background • AEO2018 Issues in Focus • Ongoing work – Levels 1-3 automated technology adoption – Multiyear effort to model key energy effects of automated vehicles – Geographic population density and travel patterns John Maples and Nicholas Chase, Laxenburg, Austria, 2 October 30, 2018
Background John Maples and Nicholas Chase, Laxenburg, Austria, 3 October 30, 2018
Definition of vehicle automation • Operational and safety-critical control functions occur without driver input • Connected and automated vehicles Source: U.S. Department of Transportation, Automated Driving Systems 2.0, A Vision for Safety John Maples and Nicholas Chase, Laxenburg, Austria, 4 October 30, 2018
Potential benefits underlie interest but there are also key uncertainties and obstacles Benefits Obstacles • Road safety • Consumer acceptance • Increased system efficiency • Technology cost and function – Route harmonization • Cybersecurity – Reduced congestion • Legal framework • Increased mobility for • Infrastructure underserved population • Policy • Less time driving John Maples and Nicholas Chase, Laxenburg, Austria, 5 October 30, 2018
Range of potential effects of autonomous vehicles on light-duty vehicle energy consumption 2017 U.S. delivered energy consumption quadrillion Btu 45.9 quadrillion Btu 50 (24.9 million b/d oil equivalent) 40 30 15.3 quadrillion Btu (8.3 million b/d oil equivalent) 20 6.1 quadrillion Btu +200% (3.3 million b/d oil equivalent) 10 -60% 0 high energy efficiency with less light-duty vehicles (2017) low energy efficiency with more vehicle miles traveled vehicle miles traveled Source: 2017: EIA, AEO2018 Reference case, extrapolation based on upper and lower limits from Estimated Bounds and Important Factors for Fuel Use and Consumer Costs of Connected and Automated Vehicles (Stephens et al) John Maples and Nicholas Chase, Laxenburg, Austria, 6 October 30, 2018
There is uncertainty about how highly automated vehicles could affect future transportation energy demand Changes in light-duty vehicle miles traveled Changes in light-duty vehicle fuel efficiency % range % range less energy more energy less energy more energy faster travel ease increased feature content cost of driving collision avoidance underserved population less congestion empty miles V2I eco-driving mode switching traffic flow + drive profile ridesharing de-emphasized performance parking platooning MaaS vehicle resizing -75% -50% -25% 0% 25% 50% 75% 100% 125% 150% 175% 200% -75% -50% -25% 0% 25% 50% 75% 100% 125% 150% 175% 200% Sources: Help or Hindrance? The Travel, Energy, and Carbon Impacts of Highly Automated Vehicles (Wadud et al); Estimated Bounds and Important Factors for Fuel Use and Consumer Costs of Connected and Automated Vehicles (Stephens et al) John Maples and Nicholas Chase, Laxenburg, Austria, 7 October 30, 2018
Additional ways vehicle automation technology could affect transportation energy consumption • Alternative fuels and energy efficient powertrains • Commercial trucks • Mass transit John Maples and Nicholas Chase, Laxenburg, Austria, 8 October 30, 2018
AEO2018 Issues in Focus — Autonomous Vehicles: Uncertainties and Energy Implications John Maples and Nicholas Chase, Laxenburg, Austria, 9 October 30, 2018
Description of scenarios • Reference case – Autonomous vehicles enter fleet light-duty vehicles • 1% of new sales by 2050 – Autonomous vehicles used more intensively • 65,000 miles/year and scrapped more quickly – Autonomous vehicle fuel type • 100% conventional gasoline internal combustion engine – Autonomous vehicles affect mass transit • Increases use of commuter rail • Decreases use of transit bus and transit rail John Maples and Nicholas Chase, Laxenburg, Austria, 10 October 30, 2018
Description of scenarios – two scenarios examine energy implications from more widespread use of autonomous vehicles • Identical assumptions – Autonomous vehicles enter household and fleet light-duty vehicles • 31% of new sales by 2050 – Autonomous vehicles used more intensively • 65,000 miles/year (fleet) ; +10% miles/year (household) on average – Autonomous vehicles affect mass transit modes • Increases use of commuter rail • Decreases use of transit rail • Decreases use of transit bus until mid-2030s, thereafter, increases transit bus use from automation technology – Automation technology included on long-haul fleet commercial trucks enables platooning John Maples and Nicholas Chase, Laxenburg, Austria, 11 October 30, 2018
Description of scenarios – two scenarios examine energy implications from more widespread use of autonomous vehicles • Autonomous Battery Electric Vehicle case – Increasing share of autonomous vehicles are battery electric through 2050 • 96% of fleet and 82% of household autonomous vehicles by 2050 • Autonomous Hybrid Electric Vehicle case – Increasing share of autonomous vehicles are hybrid electric through 2050 • 96% of fleet and 71% of household autonomous vehicles by 2050 John Maples and Nicholas Chase, Laxenburg, Austria, 12 October 30, 2018
Light-duty vehicle sales by fuel type across scenarios U.S. light-duty vehicle sales million Autonomous Battery Autonomous Hybrid 20 Reference case Electric Vehicle case Electric Vehicle case 18 16 14 12 10 8 6 4 2 0 2020 2030 2040 2050 2020 2030 2040 2050 2020 2030 2040 2050 Conventional gasoline Battery electric Hybrid electric Other Source: EIA, AEO2018 Reference case, Autonomous Battery Electric Vehicle case, Autonomous Hybrid Electric Vehicle case John Maples and Nicholas Chase, Laxenburg, Austria, 13 October 30, 2018
Light-duty vehicle miles traveled 14% above Reference case in 2050 and 35% higher in 2050 than in 2017 U.S. light-duty vehicle miles traveled billion 2017 history projection 4,000 Autonomous Battery Electric Vehicle case Autonomous Hybrid Electric Vehicle case 3,800 3,600 3,400 3,200 Reference case 3,000 2,800 2,600 2,400 2,200 ≈ 0 2,000 2010 2015 2020 2025 2030 2035 2040 2045 2050 Source: EIA, AEO2018 Reference case, Autonomous Battery Electric Vehicle case, Autonomous Hybrid Electric Vehicle case John Maples and Nicholas Chase, Laxenburg, Austria, 14 October 30, 2018
Transportation energy consumption higher in both cases compared to Reference case but still lower than 2017 U.S. transportation energy consumption quadrillion Btu 2017 history projection 28 Autonomous Hybrid Electric Vehicle case Autonomous Battery Electric Vehicle case 26 Reference case 24 ≈ 0 22 2010 2015 2020 2025 2030 2035 2040 2045 2050 Source: EIA, AEO2018 Reference case, Autonomous Battery Electric Vehicle case, Autonomous Hybrid Electric Vehicle case John Maples and Nicholas Chase, Laxenburg, Austria, 15 October 30, 2018
Transportation fuel consumption differs between cases because of changes in light-duty vehicle fuel type Transportation energy consumption by fuel quadrillion Btu motor gasoline diesel electricity 18 18 18 16 16 16 Reference case 14 14 14 Autonomous Hybrid 12 12 Electric Vehicle case 12 Autonomous Battery 10 10 10 Electric Vehicle case 8 8 8 6 6 6 4 4 4 2 2 2 0 0 0 2010 2020 2030 2040 2050 2010 2020 2030 2040 2050 2010 2020 2030 2040 2050 Source: EIA, AEO2018 Reference case, Autonomous Battery Electric Vehicle case, Autonomous Hybrid Electric Vehicle case John Maples and Nicholas Chase, Laxenburg, Austria, 16 October 30, 2018
Ongoing work John Maples and Nicholas Chase, Laxenburg, Austria, 17 October 30, 2018
Recent modeling Tech cost Fuel Scrappage curve economy rates Cost & utility of Cost ride Cost transit vehicle ROI hailing ownership Availability of ride hailing Deliveries The Urban v. Availability PMT VMT Empty miles suburban of transit BRAIN v. rural Traditional Population density Underserved population John Maples and Nicholas Chase, Laxenburg, Austria, 18 October 30, 2018
Recent modeling focus: adding levels of highly automated vehicles — • Levels of vehicle automation (introduction year, cost, weight, fuel economy, etc.): automation level description driver assistance technology Level 1 partial automation technology Level 2 conditional automation technology Level 3 low speed (<35 mpg) operation in limited geofenced areas such as urban centers Level 4a full speed operation in limited geofenced areas such as limited access highways Level 4b fully autonomous vehicle that can operate on all roads and all speeds Level 5 John Maples and Nicholas Chase, Laxenburg, Austria, 19 October 30, 2018
Recommend
More recommend
