modeling the transition to advanced vehicles and fuels - mit...

1

1

Modeling the Transition to Advanced Vehicles and Fuels:Review & Hydrogen Scenarios

Paul N. Leiby, David L. Greene, Zhenhong Lin, David Bowman, Sujit Das

Oak Ridge National LaboratoryJune 9, 2009

Presented at the MIT-Ford-Shell Workshop, “Strategies for Market

Transitions to Alternative Energy and Transportation Systems

2

ORNL Research Path for Alternative Vehicle and Fuels Transitions

Alt Fuels Transition

Model (AFTM)1996-1998

Transitional Alternative

Fuels & Vehicles

Model (TAFV)2001-2004

Hydrogen Transition

Model (HyTrans)2005-08

Hydrogen/ PHEV

transition Model

(HyTrans II)2009

Long-run (2010) static. Multiple fuels & vehicles market model, international fuel trade. Cost min.

International fuel trade, refinery submodel

PHEVs, PFCVs, international scenarios

Dynamic market model, vehicle scale economies, fuel availability. Intertemporal optimization.

Other AFVs: Meth/Ethanol, CNG, elec.

H2 FVCs & HEVs, H2 spatial fuel infrastructure, Learning-by-doing

2

3

“Market Potential and Impacts of Alternative Fuel Use in Light Duty Vehicles: A 2000/2010 Analysis, DOE/PO-0042, Office of Policy

Analysis, U.S. Department of Energy, Washington, DC, January, 1996.

Transition Barriers Matter:Static Vs. Transition Analyses: Differing

Conclusions2010 Alternative Fuel Demand Shares, Base Case, No New Policies

GasolineCNG

Ethanol

Methanol

LPGElectricity

Gasoline CNG

Ethanol Methanol

LPG Electricity

No Transitional Barriers

Source: P.N. Leiby, D.L. Greene and

Harry Vidas. 1996.*

(AFTM analysis, 1997)

4

Source: P.N. Leiby, D.L. Greene and

Harry Vidas. 1996.*

“Market Potential and Impacts of Alternative Fuel Use in Light Duty Vehicles: A 2000/2010 Analysis, DOE/PO-0042, Office of Policy

Analysis, U.S. Department of Energy, Washington, DC, January, 1996.

Transition Barriers Matter:Static Vs. Transition Analyses: Differing

Conclusions2010 Alternative Fuel Demand Shares, Base Case, No New Policies

GasolineCNG

Ethanol

Methanol

LPGElectricity

Gasoline CNG

Ethanol Methanol

LPG Electricity

No Transitional Barriers

Gasoline

Gasoline CNG

Ethanol Methanol

LPG Electricity

Transitional Analysis With Barriers

(TAFV analysis, 2001)(AFTM analysis, 1997)

3

5

Key findings that were not well known or not measured

• Found that transition matters a lot• Can ID some of most important barriers• Forcing adoption by a few (e.g. fleets) doesn’t

work unless technology has private appeal • Without compelling private advantages,

sustained policy intervention is required– But Temporary policies can sometimes drive

technology to competitiveness• Transitional analysis still uncertain, since key

factors little-understood– How strong is lock-in/out, how many techs can

co-exist?

6

The report, published 2008, documents the

first integrated national hydrogen transition analysis.

Responded to the NAS’ call to better understand what a

transition to H2 powered vehicles

would require

cta.ornl.gov/cta/Publications/Reports/ORNL_TM_2008_30.pdf

4

7

Approach: HyTrans integrates established data and models with new components, to simultaneously represent key agents: 1) fuel supply, 2) vehicle manufacture, 3) consumer choice.

GREET

H2 FeedstockSupply Curves

HYTRANS Components

H2A H2Production

Models

H2A H2DeliverySystemsModels

HydrogenInfrastructureInvestment

HydrogenAvailability

Price ofHydrogen

PSAT ASCMVehicle

Attributes

VehicleChoiceModel

Make & ModelDiversity

Vehicle Price &Sales by

Technology

HydrogenUse

VehicleStock &

Fuel Use

LDVSales

Non-TransportHydrogen Use

VehicleSupply ModelTech. Change

LearningScale Econ.

AEO EnergyScenarios

GHG EmissionsCosts

Blue: models

from other Team

members

Yellow: HyTrans

unique components

White: essential HyTrans outputs

8

• Fuel infrastructure (density/distance/cost)• Limited fuel availability• Limited make and model availability• Scale (dis)economies• Learning-by-doing• All are represented in HyTrans

HyTrans models the excess “transition costs” incurred in overcoming the natural

market barriers to a new transportation fuel.

5

9

Economies of scale were the chief factor in reducing hydrogen supply costs.

Hydrogen Production and Delivery Costs, Los Angeles, Future #4 ($/kg)

$0.00

$0.50

$1.00

$1.50

$2.00

$2.50

$3.00

$3.50

$4.00

2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

SMR Dist.

Coal w/o Seq. by Pipeline

Biomass w/o Seq. Pipeline

10Case: 1000 kg/day station, 31 miles plant to city, 10% Summer Surge, 8% Friday Peak

Demand density and distances strongly affect delivery costs.

Reduced Form Cost Curves for Infrastructure Developed From Latest H2A/HDSAM Models

Delivery Cost: Pipeline Mode

4034

066

098

01,3

001,6

201,9

402,2

602,5

802,9

003,2

203,5

403,8

6030

570

1,110

1,650

2,190

2,730

0.00

1.00

2.00

3.00

4.00

5.00

6.00

$/kg

City H2 Use (Thousands kg/day)

City Area (square miles)

5.00-6.004.00-5.003.00-4.002.00-3.001.00-2.000.00-1.00

4034

066

098

01,3

001,6

201,9

402,2

602,5

802,9

003,2

203,5

403,8

6030

570

1,110

1,650

2,190

2,730

0.00

1.00

2.00

3.00

4.00

5.00

6.00

$/kg

City H2 Use (Thousands kg/day)

City Area (square miles)

5.00-6.004.00-5.003.00-4.002.00-3.001.00-2.000.00-1.00

Delivery Cost: Liquid (Cryo) Truck Mode

E.g. Smooth Delivery and Forecourt Costs as Function of Demand Volume and Area (Very different shapes by mode)

Demand Volume Demand Volume

City Area

City Area

6

11

Fuel Cell Vehicle Retail Price as a Function of Learning, Scale and R&D in Scenarios 2 & 3

$0

$50,000

$100,000

$150,000

$200,000

$250,000

$300,000

$350,000

2011 2013 2015 2017 2019 2021 2023 2025

RPE

of D

rivet

rain

+ G

lider

Assumed CostPredicted 2Predicted3

• FCV cost estimates discussed with OEMs – GM, Ford, DC, Toyota, Honda, Nissan, BMW.

• Proprietary cost estimates by year & volume provided by GM, Ford & DC.

• Average used to calibrate HyTranslearning and scale economy functions.

• Results reviewed by GM, Ford, DC and judged reasonable.

In all scenarios FCV costs decline dramatically with reasonable correspondence to the average of

the manufacturers’ estimates.

12

The vehicle cost model calibrated with data provided by three OEMs is a dynamic function of past and current FCV production volumes.• Three multiplicative factors:

– Independent tech-progress (autonomous), – Learning-by-doing and – Scale economies.

• Vehicle Price = Glider Cost+ Long-run Drivetrain Cost x Technology(time) x Learning-by-doing(stock) x Scale(volume)

• Independent Technology progress– calibrated to DOE 2015 goals– “in the lab” + available in vehicles in 5 years

• Learning & Scale – calibrated to central tendency of

manufacturers’ cost estimates.

7

13

Make and Model Diversity Cost (Passenger Cars)

$0$1,000$2,000$3,000$4,000$5,000$6,000$7,000$8,000

0% 20% 40% 60% 80% 100%

Fraction of Makes and Models Offering Alternative

Ret

ail P

rice

Equ

ival

ent

Cost of Limited Fuel Availability (Passenger Car)

-$16,000-$14,000-$12,000-$10,000

-$8,000-$6,000-$4,000-$2,000

$00% 10% 20% 30% 40% 50%

Fraction of Stations Offering Fuel

Pres

ent V

alue

Dol

lars

Own Region (Nicholas, etal., 2004)Three Regions

Greene, 2001

Impacts of limited make and model available and limited fuel availability on consumers’ choices were

explicitly included (although uncertain)

Still leading edge. Issue of flexibility/ customizability of FCV

Substantial progress and convergence of views. Issues:

-Home refueling

-Interregional Avail.

-Station redundancy?

14

Scenario 1: 100s per year by 2012, tens of thousands of vehicles per year by 2018. On-road fleet of 2.0 million FCVs by 2025.Scenario 2: 1,000s of FCVs by 2012, tens of thousands by 2015 and hundreds of thousands by 2018. On-road fleet of 5.0 million FCVs by 2025.Scenario 3 (NRC scenario): 1,000s of FCVs by 2012, and millions by 2021, 10 million on the road by 2025.

These scenarios do not represent a policy recommendation.

Stakeholder workshops led to Lighthouse strategy. Three Early Vehicle Scenarios intended to span a

range that would encompass an efficient transition.

0

200

400

600

800

1000

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

2024

2025

Year

Ann

ual V

ehic

le S

ales

(tho

usan

ds)

Scenario 1HEV +15 yearsScenario 2Scenario 3HEVs +12 years

Scenarios 1 and 2 are consistent with current and projected HEV penetration rates

8

15

The Lighthouse Concept of infrastructure build-out reflects a trade-off between need to concentrate infrastructure and need to maximize hydrogen availability.

16

Scenarios: Premises matter.• All DOE FreedomCar program goals met on schedule.

– Vehicle cost and performance estimates based on PSAT/ASCM analysis (Rousseau et al., 2000).

– Estimates in “DOE Goals” scenario based on meeting program goals in 2010 and 2015, with 5-year lag to the first production vehicles.

• H2A production and delivery models used for H2 supply costs.

• CO2 price impact was investigated in sensitivity cases• 2006 AEO oil price scenarios

– High Oil Price Case used as base case…$72/bbl in 2015– Also Reference Case….$43/bbl in 2015

• HyTrans constrained to follow scenarios to 2025, then simulate for endogenous market solution.

9

17

Vehicle Production Share

0%10%20%30%40%50%60%70%80%90%

100%

2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

Advanced GasolineGasoline HybridH2 Fuel Cell


0%10%20%30%40%50%60%70%80%90%

100%

2005 2010 2015 2020 2025 2030 2035 2040 2045 2050


Scenario 1

• Policies will almost certainly be required for early transition period (2012-2025). (

• Assumes High Oil price case and the Hydrogen and FreedomCar Programs achieve full success.

• Does not consider impact of uncertainty on willingness to invest.

Scenario 3

All three scenarios produced a sustainable transition to hydrogen powered light-duty vehicles without any additional policy measures after 2025.

18

Simulated Auto Industry Cash Flow From Sale of Hydrogen Fuel Cell Vehicles, No Policy Case

-$5

-$4

-$3

-$2

-$1

$0

$1

$2

$3

2010 2015 2020 2025

Bill

ions

of D

olla

rs

Scenario3Scenario2Scenario1

The need for transition policies is indicated by the excess costs of the

transition scenarios.• Without government

policies, the entire transition burden would have to be borne by industry.

• Automotive and energy industries faced with years of billion dollar+ losses without government policy.

• Investment unlikely until outer years (2045+)

Simulated Auto Industry Cash Flow From Sale of Hydrogen Fuel Cell Vehicles, Policy Case 2

-$1

$0

$1

$2

$3

$4

$5

2010 2015 2020 2025

Bill

ions

of D

olla

rs


10

19

Policy Case 2 provides a reasonable assessment of the costs the government might shoulder to

induce a transition to hydrogen.• Assumes “Fuel Cell Success”• FCV vehicle production costs (vs advanced HEV)

shared• 50% total vehicle cost through and including 2017• Tax credit covers 100% of incremental cost 2018 to 2025

• Station capital cost starts at $3.3 million, declining to $2.0 million• Cost share $1.3 million/station, 2012-2017• Cost share $0.7 million/station, 2018-2021• Cost share $0.3 or 0.2 million/station, 2022-2025

• H2 fuel subsidy• $0.50/kg through 2018• Declines to $0.30/kg by 2025

20

Cost Sharing and Subsidies, Scenario 3, Fuel Cell Success, Policy Case 2

0

1

2

3

4

5

2010 2015 2020 2025

Billi

ons

of 2

004

Dolla

rs/Y

r Scenario3 Station Infr.Scenario3 Fuel SubsidyScenario3 Vehicles

Cumulative Cost Sharing and Subsidies, Scenario 3, Fuel Cell Success, Case 2

0

5

10

15

20

25

30

2010 2015 2020 2025

Bill

ions

of 2

004

$ (U

ndis

coun

ted) Scenario3 Station Infr.

Scenario3 Fuel Subsidy

Scenario3 Vehicles

Simulated Auto Industry Cash Flow From Sale of Hydrogen Fuel Cell Vehicles, Policy Case 2

-$1

$0

$1

$2

$3

$4

$5

2010 2015 2020 2025

Bill

ions

of D

olla

rs


In Policy Case 2, scenario 3 annual costs peak near $5B and cumulative costs rise to $26B by 2025.

11

21


0%10%20%30%40%50%60%70%80%90%

100%

2005 2015 2025 2035 2045


AEO Ref., $50/bbl


0%10%20%30%40%50%60%70%80%90%

100%

2005 2015 2025 2035 2045


No Policy Scenario

In the absence of an early transition FCV deployment, advanced hybrid electric vehicles come to dominate light-duty vehicle propulsion.

At a lower AEO oil price of $50/bbl in 2030, a sustainable transition to FCVs occurs but there is stronger competition from hybrid electric vehicles.

With no early transition scenario, FCVs do not begin to penetrate the market until after 2045.

22

If DOE 2015 High Tech Goals Met Except Storage Cost, Transition May Falter if Storage Costs Very High ($27/kWh)

• Scenario 3, High price, technologies meet DOE 2015 High Goals for all except on board storage costs (which is $27/kWh rather than $2/kWh)

Scenario: 3 USA Case: SVS000S25V265LZPNOBT00HHAHC30O3


0%10%20%30%40%50%60%70%80%90%

100%

2005 2010 2015 2020 2025 2030 2035 2040 2045 2050


Preliminary: Results based on Prior (2008 HyTrans Version

Vehicle Shares

12

23

CO2 Emissions From LDVs

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

2010 2020 2030 2040 2050

Gig

aton

s (tr

ill k

g) C

O2

Scenario0, CO2 Tax = $25/MT



• Scenario 0 $25/MT -> no transition policies with a carbon tax.

• Scenario 3 $00 -> no carbon policy, hydrogen may be produced from carbon-intensive sources such as coal without sequestration..

• In Scenario 3 $25/MT -> 2/3 reduction by 2050.

• Carbon emission policy = $10/ton of CO2 in 2010 and $25/ton in 2025.

Meaningful carbon mitigation policy is necessary to achieve dramatic reductions in GHG emissions.

24

Assessment of future market for H2-FCVs involves many of the Central Challenges of modeling intro-duction of new technologies and their infrastructure.

• 1. Representation of technological change and its components– including cost reductions through technological progress, scale economies, and

Learning-by-Doing.• 2. Accounting for special for geographic/spatial modeling issues

– and spatially-related costs and markets, and– the evolution of geographically differentiated infrastructure.

• 3. Describing plausible evolution of demand & infrastructure– Semi-rigid capital, slow adjustment, new investment, stranded assets– Smooth succession of compatible Technologies or sudden (“disruptive”) change?

• 4. Specifying the nature and structure of consumer choice among new vehicle types;

– Capturing the state of knowledge regarding how consumers value fuel availability and diversity of make and model choice;

• 5. Representation of interactions with other energy markets, especially feedstock supply costs, competition with other fuels/vehicles.

• 6. Interactions of a national program with global vehicle marketsduring the transitions stage; and

• 7. Accounting for impacts of risk and expectations.

Source: Paul N. Leiby, David L. Greene and David Bowman, “"Issues in Integrated Market Modeling of the Hydrogen Transition,“ORNL, Presented at the International Energy Workshop (IEW), Stanford University, Stanford, CA, June 25-27, 2007.

13

25

THANK YOU.

26

Publications and Presentations• Greene, D.L. and K.G. Duleep, 2008. “Bootstrapping a Sustainable North American Fuel Cell Industry:

Could a Federal Acquisition Program Make a Difference?”, Energy and Transportation Science Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee, April 18.

• Greene, D. L. and P. N. Leiby, B. James, J. Perez, M. Melendez, A. Milbrandt, S. Unnasch, M. Hooks, S. McQueen and S. Gronich, 2008. “Analysis of the Transition to Hydrogen Fuel Cell Vehicles & the Potential Hydrogen Energy Infrastructure Requirements”, ORNL/TM-2008/30, Oak Ridge National Laboratory, Oak Ridge, Tennessee.

• Greene, David L., Zhenhong Lin, Paul N. Leiby, David Bowman “Integrated Market Analysis of the Impacts of Hydrogen Storage Costs on the Transition to Fuel Cell Vehicles,” Oak Ridge National Laboratory, Presented at H2 Storage Tech Team Meeting, July 17, 2008, Detroit, Michigan.

• Bunzeck, Ingo (ECN) and Paul Leiby, (ORNL) "HyWays-IPHE Project: Dissemination Task – Summary of the Roadmap Workshop in Changsha, China, Aug. 29 2008," on behalf of the HyWays-IPHE consortium, Workshop on WP4 wrap-up and Further exchange, Washington D.C., September 29, 2008

• Greene, David L., Paul N. Leiby, Zhenhong Lin, David Bowman,“ HyTrans Model: Analyzing the Transition to Hydrogen-Powered Transportation, Supplementary Slides On Impact Of Variations In Vehicle Technology Costs,” Oak Ridge National Laboratory, September 26, 2008.

• Leiby, P.N, D.L.Greene, D. Bowman and E. Tworek, “Systems Analysis of Hydrogen Transition with HyTrans”, Transportation Research Record No. 1983, Transportation Research Board, National Research Council, 2006.

• Leiby, Paul N. and Jonathan Rubin, July 2003. “Transition Modeling of AFVs and Hybrids:LessonsLearned.”

modeling the transition to advanced vehicles and fuels - mit...

Documents