Transportation Model

The 2021 Power Plan is the first to include a detailed analysis of the regional transportation sector. The primary focus of the work was on roadway modes of travel which includes passenger cars, vans and light duty trucks, as well as heavy duty trucks, and transit buses. The study covers a time horizon of 25 years, beginning in 2020 and ending in 2045.

To facilitate the study, a model of transportation was developed and implemented into Energy2020, our long-term load forecasting tool. Energy2020 is an all-fuels, end-use demand forecasting model with a base in consumer choice theory and system dynamics.[1]

Our work in the transportation sector revolved around three goals:

How much electricity demand growth could the region experience over the next 20 years as a result of electric vehicles?
How much does electrification of transportation lower regional emissions?
How could other alternative fuels – such as hydrogen – affect transportation?

Three study cases were explored for the plan. A Reference Case, which sets a base level of fuel use and electrification which is used in the planning reference forecast for electricity, a High Electric Case which models an aggressive electrification future, and a Hydrogen-Electrification Case (H₂E) which introduces fuel cell technology in heavy duty vehicle categories like freight trucking.

It’s important to note that this is just one representation of the regional transportation system; which in reality is a large and complex system. But in order to build a manageable modeling tool, assumptions and estimates were made to get what we hope are useful results. As with other systems, our transportation modeling will continue to evolve and be refined over time.

The key findings and a summary of the modeling results can be found here.

Definitions

The transportation model is structured into 5 roadway categories.

Roadway Transportation Mode Categories

Category Name	Description	Weight
LDV	Light duty vehicles such as passenger cars, mini vans, sport utility vans, and small pickup trucks	< 8,500 lbs.
HDV Light	Commercial light trucks, Class 2	8,500 to 10,000 lbs.
HDV Med	Larger trucks, Class 3 through 6	10,000 to 26,000 lbs.
HDV Heavy	Freight trucks, Class 7 and higher	> 26,000 lbs.
Bus	Transit and School

The model incorporates five basic vehicle fuel options to meet the forecasted transportation requirements. These options are listed in the table below, an X indicates that the fuel option is available in that category, depending on the case.

Transportation Vehicle Fuel Options

Fuel	LDV	HDV Light	HDV Med	HDV Heavy	Bus
Gasoline	X	X	X	X	X
Diesel	X	X	X	X	X
Natural Gas	X				X
Direct Electricity	X	X[2]			X
Hydrogen			X[3]	X³

For this study, direct electric vehicles include passenger cars and light duty trucks and vans with a drivetrain powered by an on-board battery and electric motor. These include plug-in battery electric vehicles (BEV) and plug-in hybrid electric vehicles (PHEV), which are electric vehicles with a gasoline powered range extender.

Hydrogen fuel cell vehicles were included in the H₂E Case for the HDV Med and HDV Heavy categories only. There are currently three auto manufacturers – Toyota, Hyundai, and Honda - selling fuel cell vehicles in the LDV category, however for this case we only modeled the trucking options, assuming a high direct electric market share for LDV.

The three transportation cases are outlined in the table below. The growth rates for transportation requirements, like vehicle miles traveled (VMT) remain the same across each case. The allocation of demand can change across vehicle fuel type.

Transportation Cases

Economic Growth	Case	Description
Power Plan reference case – economic growth drivers and travel and freight requirements remain consistent across cases	Reference	Forecasts a gradual shift to plug-in electric vehicles in the LDV category over the planning horizon. HDV remains primarily in diesel fuel. Direct electric vehicle options are only in the LDV and Bus categories
	High Electric	Builds on the Reference Case but with an aggressive move to plug-in electric vehicles in the LDV category, additional electrification of transit buses, and the addition of electric trucks in the HDV Light category.
	H₂E	Expansion of the High Electric Case with a transition to hydrogen fuel cell vehicles in the HDV Med and Heavy categories – this is considered indirect electrification

Methodology and Inputs

The model methodology begins by using historic transportation data, such as vehicle sales and stock, vehicle capital costs, fuel prices, vehicle efficiencies, population growth, and energy demand by fuel type to perform a calibration. The calibration also sets price effect variables, and non-price effect variables to be used to help estimate demands going forward by vehicle fuel type – like gasoline, diesel electric.

For historic transportation demand, data for each of the Northwest states is compiled from the US Energy Information Administration’s State Energy Data System (SEDS)[4]. The demand data is then allocated to the model vehicle categories based on more detailed transportation information from the Annual Energy Outlook (AEO)[5].

New energy demand requirements are determined each year in the forecast. The new demand may come from the need to fill in stock retirements, and/or to fill demand growth resulting from the forecast driver, in this case population growth for LDV, and gross domestic product for HDV. The model allocates the new energy demand among the competing vehicle technologies based on the individual technology costs (capital cost, fuel price, efficiency) in conjunction with the price-effect and non-price effect model variables. This allocation sets the market shares by technology (gasoline, diesel, electric, natural gas).

Vehicle unit sales and stock quantities are estimated in the model based on the forecast of energy by technology, and historic relationships between energy use and vehicle counts. It’s important to note that Energy 2020 is an energy model, not a units-based model. Sometimes the estimates from conversions between the two are not always an exact representative of what stock counts should be.

For electric vehicles, load scale factors were calculated exogenously and input to the model. These scale factors quantify the contribution of electric vehicle demand to the overall electric system peak. Derivation of the load profiles and factors are covered here.

A data set for historic vehicle sales and registrations (stock) was cobbled together from various sources. For the state of Washington, data from the Department of Licensing[6], and the Department of Transportation[7] and the electric vehicle dashboard[8] was used. Other sources that were tapped for the region include the Oregon Department of Transportation[9], the Oregon Department of Energy (Electric Vehicle Dashboard)[10],the US Department of Transportation Bureau of Statistics[11], and IHS-Global Insights.

Cost data from the EIA Annual Energy Outlook 2019[12] (AEO 2019) was used to derive the initial vehicle capital cost inputs. This is the upfront cost that consumers would see when considering which vehicle to purchase. For the electric vehicle category, this cost data set is dominated by Tesla model sales, which sit on the high end of vehicle prices. For instance, in the state of Washington, the combined sales of Tesla Model 3,S,X, and Y is over 50% of total electric vehicle sales. These high-end electric vehicle sales drove the cost input from the AEO report to over $55,000 for a new electric vehicle, while the typical cost for a new traditional gasoline power vehicle was around $37,500.

For the power plan, we are assuming that electric vehicles follow a declining cost curve. There are two reasons for this. First, there are many more electric vehicle model options coming – the auto manufacturers have embraced electrification and are re-tooling manufacturing lines to electric lines. For example, GM plans to release 30 new electric vehicle offerings by 2025. Many will not be as expensive as the Tesla models. Also, the cost for lithium-ion battery packs continue to decline, which will help to bring all electric vehicle model prices down. We are modeling a future where electric vehicles reach capital cost parity with traditional gasoline powered vehicles (in the personal vehicle space) by 2024.

For emission factors, the EPA publishes transportation emission factors based on fuel type for greenhouse gases like CO₂, CH₄ and N₂O.[13]

[1] https://www.energy2020.com

[2] High Electric & H₂E Case only

[3] H₂E Case only

[4] https://www.eia.gov/state/seds/seds-data-complete.php?sid=US#Consumption

[5] https://www.eia.gov/outlooks/aeo/tables_ref.php

[6] https://www.dol.wa.gov

[7] https://wsdot.wa.gov

[8] https://data.wa.gov/Transportation/Electric-Vehicle-Population-Data/f6w7-q2d2

[9] https://www.oregon.gov/odot/Pages

[10] https://www.oregon.gov/energy/Data-and-Reports/Pages

[11] https://www.bts.gov/browse-statistical-products-and-data