Journal of Cleaner Production 228 (2019) 1354e1367

Contents lists available at ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

A break-even analysis for battery electric trucks in Latin America

Martín Tanco*, Luis Cat, Santiago Garat
CINOI, Universidad de Montevideo, Luis P. Ponce, 11300, Montevideo, Uruguay

a r t i c l e i n f o a b s t r a c t

Article history: The transport sector stands in the core of Latin-American economies but is also responsible for 19% of
Received 6 November 2018 CO2 emissions in the continent. Battery electric trucks (BET) are a modern alternative to diesel trucks
Received in revised form with the potential to mitigate the freight transport emissions of CO2. This study makes an economic
25 February 2019
analysis, in a total cost of ownership (TCO) basis, for different weight classes and applications in five
Accepted 15 April 2019
Latin-American countries: Argentina, Brazil, Chile, Colombia and Uruguay. Based on data gathered by a
Available online 16 April 2019
state-of-the-art literature review, the aim of this paper is to calculate the break-even year for each
segment's BET by means of a quantitative linear model. This model considers multiple parameters that
Keywords:
Battery electric trucks (BET)
vary between the different countries, such as purchase price, fuel and electricity costs, insurance and
Latin America registration costs, maintenance costs, import tariffs and charging infrastructure costs. Results show that
Total cost of ownership (TCO) Chile and Uruguay are the first countries to achieve break-even while Argentina and Brazil the last. A
Public policies lower mileage leads the transition together with light-duty trucks, whereas heavy-duty trucks fight to
Break-even analysis reach parity with diesel trucks. Initial investments on BETs are highlighted as the main barrier for their
adoption since it is the main cause for the TCO gap between both alternatives, while the electricity and
fuel cost differential is also identified as a relevant factor to achieve parity. Public policies such as tax
exemptions and subsidies are analysed by studying their impacts in achieving an earlier break-even.
Finally, this research provides insight into a currently deserted venue as it is the insertion of BETs in
Latin America.
© 2019 Elsevier Ltd. All rights reserved.

1. Introduction order to attack the climate repercussions. Nevertheless, regulations

on the transport sector are uneven throughout the continent, with
The transport sector is a main driver of Latin-American econo- some countries as Chile adopting Euro VI measures (since 2014)
mies, but it is also accountable for the generation of negative ex- while others such as Uruguay have no regulations at all in this
ternalities like climate change. In Latin America, the transport aspect. Despite these disparities, Latin America presents the best
sector is responsible for 19% of CO2 emissions in the region (United conditions for battery electric vehicles (BEVs) due to the high
Nations Environment Programme, 2016). Additionally, the renewable electric energy generation potential (United Nations
International Council on Clean Transportation (2017) has projected Environment Programme, 2016). Therefore, the transition towards
a 75% increase in CO2 emissions in the region between 2010 and electric mobility may ultimately result in a major boost for coun-
2030 (with the exception of Brazil). Albeit these worrying figures, tries of the region in completing the goals set in December's 2015
Latin America remains with one of the lowest CO2 emitting electric Paris Agreement.
generation matrices due to the high presence of hydroelectric po- A global insight exhibits that Latin America faces several of the
wer and a growing development of non-conventional renewable typical entry barriers for the expansion and usage of BEVs, with
energies (United Nations Environment Programme, 2016). Conse- scarce incentives as the ones used worldwide to propel electric
quently, different transportation modes in the region, such as vehicles sales (Marcha
n and Viscidi, 2015). As with regulations,
battery electric trucks (BETs), merit profound study. uneven usage of incentives in Latin America is also common: tax
The reduction of the transport sector's emissions is critical in and import tariffs exemptions exist in some countries and are
inexistent in others. For example, Brazil and Colombia have VAT
exemptions for BEVs; Argentina, Chile and Uruguay do not. This
exposes the inconsistent approach for efficient and sustainable
* Corresponding author. transportation in the continent. Furthermore, not only timorous
E-mail address: [email protected] (M. Tanco).

https://doi.org/10.1016/j.jclepro.2019.04.168
0959-6526/© 2019 Elsevier Ltd. All rights reserved.
 M. Tanco et al. / Journal of Cleaner Production 228 (2019) 1354e1367 1355

approaches are being made towards more efficient mobility (this also given as to how the TCO may vary with fluctuations that may
includes and focuses on electric mobility), but also important fuel occur to some parameters.
subsidies currently exist in the region which ultimately affect the
adoption of alternative energy vehicles such as BEVs. Latin America 2. Literature review
has seen a small increase in BEVs purchases with just a few thou-
sand vehicles currently circulating within the whole region. Most of Research on BEVs has increased over the years and a significant
which are corporate or governmental vehicles, taxis and buses, amount of studies have been made in the field. Nonetheless,
with just hundreds belonging to private individuals (Marcha
n and research in BETs is still young, even more so in Latin America. This
Viscidi, 2015). research compiles studies of both streams and gathers data to
Finally, in Latin America, the literature review made exposes provide a basis upon which to study new horizons of BETs in Latin
that there are not many total cost of ownership (TCO) studies on America. Using Scopus database, over 200 papers were found
BETs. There are recent reports that present the general situation in through the use of several keywords such as “battery electric ve-
the region and some particular conditions given in some countries. hicles”, “electric mobility”, “battery electric trucks” and “total cost
The UN report (2016) presents a wide view of Latin America's of ownership”. This number of papers was narrowed to 47 when
relationship with electric mobility, the current market, the context selecting the most relevant and current articles for this investiga-
for their introduction and the existing incentives for electric vehi- tion. In addition to these papers, government sources and inter-
cles together with different future scenarios that might arise with national reports were consulted to obtain additional information
electric mobility's growth. Marcha
n and Viscidi (2015) study fo- regarding electric mobility.
cuses in the perspective for electric mobility in four countries The TCO is one of the main aspects studied when economically
(Brazil, Chile, Colombia and Mexico) and offers recommendations comparing BETs and internal combustion engine trucks (ICETs). It
on key public policies to promote this new technology. However, no basically looks into all the costs incurred during the life cycle of a
TCO study is done and there is no mention of BETs. Another grand vehicle, providing an additional tool for potential buyers’ decision-
view of the region's situation and future perspective on electric making. The number of TCO studies on electric trucks versus con-
mobility is done by the Inter-American Development Bank (2016) ventional trucks is not as extensive as the number of TCO studies for
with the singularity that this report includes a TCO analysis in the passenger vehicles. Nevertheless, a reasonable number of studies
Latin-American context for a specific list of countries (Argentina, have been made with different approaches and applications. Davis
Brazil, Chile, Colombia, Mexico and Peru). Nonetheless this TCO and Figliozzi (2013) developed a vehicle routing problem model to
analysis is based on the comparison of only passenger cars. Some evaluate the competitiveness of electric delivery vehicles in com-
recent studies on electric mobility in the region were also found, parison to their diesel counterparts. Zhou et al. (2017) compared a
such as da Silva et al. (2018), but no reference to BETs were found. class 6 medium-duty electric truck against a diesel counterpart in
Their aim is to “present an overview of the economic aspects of BEV Toronto, Canada and arrived to the conclusion that for the base
purchases from the consumers' perspective in all Brazilian state case, the lifetime TCO is greater for the BET. Taefi et al. (2017)
capitals” focusing in passenger cars. Despite the growing interest in answered the question of which daily mileage is the most cost-
BETs and emergence of new prototypes and announcements of efficient for light and medium-duty electric vehicles. Sen et al.
production in Brazil as early as 2020, there are practically no studies (2017) used a life-cycle assessment method to analyse and
made in Latin America, a niche that this paper aims to tackle. compare the life cycle costs of several HDTs, that include two
Drivers and barriers intrinsically exist to any insurgent tech- computer modelled BETs. Lastly Morrison et al. (2018) estimated
nology, with the latter overwhelming the former. This paper in- the TCO of several battery electric and fuel cell light duty vehicles.
tends to compile all the existing barriers and drivers currently Another fundamental aspect when studying BETs are the bat-
identified and consequently calculate the TCO for light-duty (LDT), teries. They substitute the fuel tank and are the most expensive
medium-duty (MDT) and heavy-duty trucks (HDT) in diverse single element of electric vehicles. The most common batteries
application scenarios (urban, regional and long-haul cycles) and used nowadays are lithium-ion and differ between each other in
different Latin-American countries. In section 2 a literature review the cathode material (anode made of graphite) (Helbig et al., 2018;
on electric mobility worldwide, with focus on battery electric Pehlken et al., 2017; Zubi et al., 2018). The two most common types
trucks, is shared. The main barriers that the technology faces and of cathodes in BEVs battery packs are lithium-nickel-cobalt-
the trends in battery costs and development are highlighted. Once aluminium oxide (NCA) and lithium-nickel-manganese-cobalt ox-
the literature review is presented, this paper will continue to ide (NMC) (Rohr et al., 2017; The Boston Consulting Group, 2010).
analyse the potential break-even points for BETs in diverse sce- These two have the longest life span and the largest energy and
narios across Latin American countries, particularly in Argentina, power densities (The Boston Consulting Group, 2010; Zubi et al.,
Brazil, Chile, Colombia and Uruguay. The first four are naturally 2018). NCA cell cathode is comprised 80% of nickel, 15% cobalt
alluring cases as they compose the main economic force of the and 5% aluminium, while NMC cell cathode is 33% nickel, 33%
continent and have successfully introduced policies related to manganese and 33% cobalt (Zubi et al., 2018). Helbig et al. (2018)
electric mobility (except Argentina). Uruguay, due to its political found that the biggest supply risks involved in the production of
and geographical situation, showcases an extremely interesting lithium-ion batteries are lithium and cobalt elements due to the
case study where renewable energy accounted for 96.2% of total huge increase in demand expected in the coming years. Zubi et al.
generation in 2018.1 Section 3 explains the methodology applied to (2018) arrived at the same conclusion, with lithium being “near
calculate the TCO for the different countries and scenarios. Results critical” and cobalt being “critical”. The term ‘critical’, represents
are presented in Section 4, followed by a sensitivity analysis por- materials that have a supply disruption risk. The status of ‘critical’
trayed in Section 5. These results are finally discussed in Section 6, for cobalt and ‘near critical’ for lithium does not necessarily imply
complemented by possible political practices and measures that that they are scarce elements. Cobalt is a supply risk because half of
would increase adoption of this technology. A qualitative insight is its production is concentrated on the Democratic Republic of Congo
where serious ethical issues exist (U.S. Geological Survey, 2018;
Zubi et al., 2018). On the other hand, lithium's greatest challenge
1
Uruguayan energy mix: http://www.ute.com.uy/SgePublico/ is for production to keep the pace of global demand (Narins, 2017;
ConsComposicionEnergeticaXFuente.aspx. Speirs et al., 2014; U.S. Geological Survey, 2018; Zubi et al., 2018).
1356 M. Tanco et al. / Journal of Cleaner Production 228 (2019) 1354e1367

Nevertheless, Speirs et al. (2014) analysed seven authors that engine and transmission, on BEVs energy is lost when charging the
explored lithium availability and found that only one of them battery, in the power converter, electric motor and finally in the
predicts a lithium supply constraint. transmission that is usually comprised of 1 gear (Gustafsson and
Another important parameter for TCO predictions is the battery Johansson, 2015; Vepsa€la
€inen, 2017). Overall efficiency of BEVs is
end of life (EOL). Battery cost is a considerable percentage of total usually north of 80% (European Federation for Transport and
vehicle cost, and thus, having to change them during a vehicle's life Environment (T&E), 2018; International Energy Agency, 2017b;
has a negative effect over the TCO. According to the United States Sripad and Viswanathan, 2017) and depends on driving condi-
Advanced Battery Consortium (1996) vehicle battery pack EOL is tions because of regenerative braking (vehicle's motors can invert
defined via two metrics, when the battery has lost 20% of its power their purpose, charging the batteries). On the other hand, ICEVs'
performance or 20% of its rated capacity, which means State of efficiency is usually south of 35% (European Federation for
Health (SOH) reaches 80%. Battery ageing consists of two aspects, Transport and Environment (T&E), 2018; Gustafsson and
calendar ageing and cycle ageing. Calendar ageing is defined by the Johansson, 2015).
battery's degradation over time, depends on factors such as tem- To sum up, the problem and gap that this paper intends to
perature and state of charge (remaining battery capacity that is address can be identified as there are practically no investigations
available for discharge). On the other hand, cycle ageing is defined on the BET topic in Latin America. Therefore, this paper aims to
by the capacity lost each time the battery is charged and dis- make an economic analysis of the TCO for BETs against ICETs in
charged, depends on factors such as depth of discharge (DOD), Latin America.
charge/discharge rate and voltage (Ahmadian et al., 2018; Pehlken
et al., 2017; Pelletier et al., 2017; Wang et al., 2016; Xu et al., 2018).
3. Methods and data
After batteries in EVs achieve their EOL, they can either be recycled
(Rahman et al., 2017; Yun et al., 2018) or reused in second life ap-
The purchase of a vehicle does not only involve an initial buying
plications with lower performance requirements (Jiao and Evans,
price. As Lebeau et al. (2015) state, at the moment of deciding the
2016; Rohr et al., 2017), after which they should be recycled or
purchase of a vehicle, “a rational fleet manager should consider
disposed correctly. Lithium-ion battery recycling is not a normal
every cost related to the vehicle choice, and not only the purchase
practice, and only small quantities have been recycled over the
cost”. This means that the TCO ought to be analysed and not just the
years (Speirs et al., 2014; Vikstro € m et al., 2013). The United Nations
initial purchase price. When selecting this approach to compare
Environment Programme (2011) estimated that lithium recycling
between a conventional and electric truck, other variables besides
rates where at less than 1%. This small recycling rate can be
the buying price such as maintenance costs and fuel economy are
explained by the fact that recycled lithium costs nowadays as much
considered. Hence, the electric vehicle may turn more attractive for
as 5 times more than the one obtained from the least costly brine-
a potential buyer.
based processes (Rahman et al., 2017). However, U.S. Geological
In order to compare all the costs associated with the purchase
Survey (2018) states that recycling is small but it is increasing
and usage of a vehicle throughout its life cycle, the present dis-
steadily due to the growth in consumption of lithium-ion batteries.
counted value financial formula is applied. The present value of
Lastly, BEV battery packs that have reached their EOL for vehicle
every cost can be added to get the full cost of an alternative.
purposes, could be used in second life applications, postponing
recycling phase which is still underdeveloped (Jiao and Evans, I
2016; Rohr et al., 2017). The most known and promising second PV ¼ (1)
ð1 þ rÞn
life application is the use of batteries as stationary energy storage
systems. With PV being the present value, I being any particular cost
As for batteries prices, most authors forecast they will decrease incurred in year n and r the discount rate.
(Berckmans et al., 2017; BNEF, 2017; International Energy Agency, To calculate the TCO, the purchase price and electric vehicle
2017a; Nykvist and Nilsson, 2015; United Nations Environment supply equipment (EVSE) are considered together with the yearly
Programme, 2016). The highlighted factors are several, such as costs (subject to the financial formula of the present value) as
intense competition and improved processes to reduce production expressed in equation (1). The purchase price and EVSE costs are
costs. BNEF (2017) survey coupled with Berckmans et al. (2017), considered a one-time cost made in the year of acquisition of the
exhibit that prices in the future will tend to decrease to values vehicle and therefore the present value formula is not applied to
between 50 USD/kWh and 100 USD/kWh or even less for new them. All other TCO costs are repeated over the considered life of
battery technologies. BNEF even states that by 2030 batteries will the vehicle (10 years).
compose 18% of the vehicle cost compared to 48% in 2016. Elon
Musk, Tesla's CEO, a pioneer in BEV production, stated that they will X
nþ10
CO
achieve a battery pack cost of 100 USD/kWh by 2021, years earlier TCO ¼ Purchase priceðnÞ þ EVSEcosts ðnÞ þ
than most existing forecasts (Holland, 2018). i¼n ð1 þ rÞin
Other important aspects of electric vehicles are their lower þ ResaleValue ðn þ 10Þ (2)
maintenance costs and their better overall efficiency. Maintenance
costs are a major advantage for BETs over ICETs, since the battery, With CO expressing the costs of ownership.
motor and electronics require little maintenance. The main reason To calculate and compare the life cycle costs, a model in MAT-
for the smaller costs, are the number of moving parts on the ve- LAB® was developed. For each of the countries studied, the costs
hicles. While electric vehicles have few moving parts, conventional that trucks incur during their life cycle were obtained as well as
ones have hundreds of them (Douris, 2017; Feng and Figliozzi, local taxes and import tariffs. All costs are expressed in US dollars.
2012; Lebeau et al., 2015; Quak et al., 2016; Weldon et al., 2018; Ultimately, with all the data gathered and defined parameters,
Zhao et al., 2016a). An example are the electric motors, which the model finds the annual TCO for both alternatives in a given
have one moving part compared to the dozens of the internal segment and consequently compares the values obtained. Through
combustion engines (Douris, 2017). this iteration, the model returns the year in which the TCO of the
A final aspect covered in the literature review is the efficiency BET results lower than the TCO of the ICET, i.e. when the result of
inherent to electric vehicles. While on ICEVs energy is lost in the equation (3) is negative.
 M. Tanco et al. / Journal of Cleaner Production 228 (2019) 1354e1367 1357

Table 1 are considered, except for road taxes. Therefore, the TCO is
Battery sizes in kWh. expressed as a function of the vehicle price, battery costs, charging
Weight class Urban (100 km) Regional (200 km) Long-haul (500 km) infrastructure, fuel and electricity costs, maintenance costs, insur-
LDT 82.2 108.72 373.8
ance and motor taxes. In the following subsections these parame-
MDT 124.4 149.76 519.6 ters are described together with the information used to determine
HDT 210.8 312 960.6 their values.

3.3.1. Initial investment

DTCO ¼ TCOBET  TCOICET (3) 3.3.1.1. Truck price. For each of the truck weight classes considered,
their local market values can be seen in Table A1 - 1. To calculate the
Based on Mckinsey's (2017) approach to study break-even BET's price, some TCO studies (California Air Resources Board
points for different truck segments in different countries, this (CARB), 2015; Lee and Thomas, 2017; Mareev et al., 2018; Sen
study focuses on five countries (Argentina, Brazil, Chile, Colombia et al., 2017) split the price into the truck's components, such as
and Uruguay), three applications (urban, regional and long-haul motors, power electronics, battery and body manufacturing. In this
cycles) and three weight classes2 (light, medium and heavy-duty way, BET and ICET costs can be compared without taking into ac-
trucks). Therefore, a total of 45 segments are to be analysed, sum- count the battery cost, that can vary depending on the range
marized in Table 1. The table showcases the 9 segments present in needed. This comparison can be seen in Table A1-2, where the BET's
each country together with their determined battery sizes. The price (without battery) is estimated as a percentage of the ICET's CIF
cases studied are subject to a tailor-made battery package and price. The percentage that will be used in this study, 98.8%, is the
charging infrastructure. Battery size is a sole function of distance mean value obtained from Table A1-2.
travelled and weight class. Moreover, the trucks' purchase price will change in time ac-
After the results for the break-even analysis are shown, the cording to the California Air Resources Board (2015) report on MDTs
Monte Carlo probabilistic method is integrated to the study in order and HDTs. The report established that ICET's prices will increase 3%
to find the effect that certain variables have over the results. The between 2020 and 2030 due to forced improvements in efficiency
Monte Carlo method simulates point values with variable distri- technologies, while in the same period, BET's price will decrease 3%
butions, allowing a range of results to be shown, instead of average because of mass production capabilities.
values (Sen et al., 2017). The parameters that will be varied are the
energy cost inflation, diesel cost inflation and BET initial cost. These 3.3.1.2. Battery sizing and pricing. In order to calculate the battery
parameters have the greatest impact over the TCO of diesel and pack cost for each truck application, these steps were followed:
battery electric trucks. firstly, the price to pay for a kWh of capacity at the battery pack
The different variables considered in the model are described level is defined; secondly, the energy consumed by the trucks in
below (adapting Lebeau's (2015) division of the financial formula), each application is studied in order to find the size of the battery
as well as data collection. Slight changes in their values according to pack; and finally, with battery size and price per kWh, the price for
the country, vehicle weight and/or distance to be travelled are also the battery pack can be obtained.
presented. In order to calculate the battery prices per kWh, several of the
most recent battery cost forecasts are used. These forecasts show
3.1. Period of ownership the battery pack costs of production but not the price that cus-
tomers will see, so in order to obtain the final selling price, a 55%
A 10 year life-cycle is considered for all the trucks after which profit margin is added composed of a 35% profit margin for the
they are left with a 20% resale value, following the same assump- manufacturer and an additional 15% profit margin from the retailer
tions used in the TCO study for delivery trucks by Davis and (Berckmans et al., 2017). The results can be seen in Fig. 1, where an
Figliozzi (2013). The number of working days for trucks is taken equation for battery price as a function of the year is deduced. The
as 250 per year and the number of kilometres per day depend on sources for the forecasts used include: Holland (2018), BNEF (2018),
the segment considered: for urban distribution 100 km, for regional the U.S. DOE (2017), the International Energy Agency (2017b),
200 km and for long-haul 500 km. These values are taken in Mareev et al. (2018) and Berckmans et al. (2017).
accordance to Mckinsey's (2017). Once the operational days are With the battery price defined (in USD/kWh), it is now neces-
determined, a simple multiplication gives the annual kilometres sary to obtain the battery size required for each given segment and
covered by a given segment. application. Consumption of the BETs is estimated using the con-
sumption of conventional trucks and what is known as the effi-
3.2. Discount rate ciency ratio, that will be later explained.
The International Council on Clean Transportation (2017) and
The discount rate is used to account for the decreased value of Reinhart (2015) studied the fuel consumption of two different types
money over time. It works as an opportunity cost for the purchase of trucks, a 40t gross vehicle weight (GVW) HDT and a 12t GVW
of the vehicle, evaluating the return that money could have had if MDT, across different speed cycles that represent different appli-
invested elsewhere. For the means of this study, the discount rate is cations such as urban, regional and long-haul distributions. Other
set at 10%, as it was considered as a reasonable number for the studies and reports (Harrington and Keupnick, 2012; International
region. Energy Agency, 2017b; U.S. Energy Information Administration,
2016) also present the average fuel consumption of different
3.3. Costs of ownership types of trucks, that include heavy, medium, and light duty weight
classes.
In this TCO analysis all costs incurred during the vehicle lifespan For the HDT segment, a representative 40-ton GVW tractor
trailer is selected for the regional and long-haul distributions, while
a 24t GVW rigid truck is considered for urban distribution. For the
2
Weight-class definitions: HDT: class 8 (>15 tons); MDT: class 4e7 (6.4e15 MDT and LDT segments, 12 and 5,5t GVW rigid trucks are selected
tons); LDT: class 2e3 (3.5e6.4 tons). respectively for all applications. With the defined weight classes
1358 M. Tanco et al. / Journal of Cleaner Production 228 (2019) 1354e1367

450
400 BNEF (2017)
Ba ery pack Price (USD/kWh)

350 DOE (2016)

300 IEA (2017)

250 Tesla (2018)

200 Mareev et al.

(2018)
150 Berckman et al.
(2017)
100
50
0 Year
015

017

019

021

023

025

027

029

031
Fig. 1. Battery pack price projections in USD/kWh.

and the studies previously mentioned, Table 2 is created, present- their data was compiled in Table A1-3, in which battery life is
ing the fuel consumption of these diesel trucks. expressed as cycle number and calendar life. During the 10-year
To determine the BET's energy consumption using the ICETs fuel lifespan considered for the trucks in the TCO calculation, there
consumption, several studies were considered (CalHEAT Research will be no need for a battery replacement. This can be concluded
Center, 2013; California Air Resources Board, 2018; European from Fig. 3, where the battery pack life of a truck that travels
Federation for Transport and Environment, 2018; Mareev et al., 200 km per day is shown in total kilometres covered (information
2018; Zhou et al., 2017), in which efficiencies of BETs and ICETs from Table A1-3). The average 910,000 km can also be expressed as
are compared through different applications. According to the an 18-year lifespan for the battery.
California Air Resources Board (2018), a correlation exists between
the consumption ratio (ICEV energy consumption divided by the 3.3.1.4. Charging infrastructure. Charging will be considered to
BEV energy consumption) and the average speed at which the ve- occur during night time, and each charger will be distributed be-
hicles are tested. The information from the studies previously tween two trucks, so each one of them will have 4 h of charging.
mentioned was gathered in Fig. 2 in order to determine this cor- The cost of the EVSE for each truck is defined in equation (4) and
relation. With this correlation, the consumption ratio for the three depends on charger power (depends on battery size). The value of
applications can be found and thus the energy consumption of the 700 USD/kW3 was taken from the European Federation for
BETs can be determined. The average speeds chosen for the appli- Transport and Environment (2018) study on long-haul BETs.
cations are 20 km/h for urban distribution, 40 km/h for regional and
lastly 80 km/h for long-haul, following the same line as the speed Battery Size
EVSEcost ¼  700USD=kW (4)
cycles used by Reinhart (2015) in its study. Results are presented in 4h
Table 3.
Consumption at 100% payload will be used to dimension the size
of the battery packs, while consumption at 50% payload will be 3.3.1.5. Import tariffs. To calculate the final BET purchase price, it is
used to determine fuel and energy operating expenses. For the necessary to add import tariffs and local taxes. EVSE are also subject
battery sizing an extra capacity will be contemplated to cope with to local tariffs. Each country's tariffs are expressed in Table 4. Tariffs
the degradation factor and as a safety factor. This will be 20% for the are applied based on an importation regime of 50/50 from China
regional and long-haul applications. For the urban application a and the EU.
200 km minimum available range will be contemplated because For the particular Brazilian case, taxes and tariffs are applied
existing BETs do not have ranges of less than 200 km. with the calculation base expressed in equation (5). The value used
for ICMS is 18%.
3.3.1.3. Li-ion battery useful life and replacement. Several studies
and papers in which battery EOL is approached were reviewed and CIF  ð1 þ ICMS  ðII þ IPI  ð1 þ IIÞÞÞ
(5)
ð1  PIS  COFINSÞ  ð1  ICMSÞ
Table 2
Fuel consumption in litres of diesel equivalent per 100 km (lde/100 km) for all ICETs
segments and applications.
3.3.2. Operational costs
Payload/Cycle Urban Regional Long-Haul 3.3.2.1. Fuel and electricity costs. Diesel prices were obtained from
HDT Global Petrol Prices (2018) (excluding VAT) while energy prices
50% 43.5 41.3 35.9 were gathered from governmental sources from each of the coun-
100% - 12.5/25/25t 50.8 50.0 41.4
tries (excluding VAT); both sets of data expressed in Table 5. Due to
MDT
50% 26.1 22.3 20.6 the unpredictability of fuel prices, they will be subject to a 2%
100% - 5.5t 30.0 25.0 22.4 annual price escalation, while different scenarios are presented for
LDT
50% 17.6 15.9 14.9
100% - 2.5t 20.0 17.6 16.3 3
Assuming an average exchange rate for 2018 of 1.165 USD for V.
 M. Tanco et al. / Journal of Cleaner Production 228 (2019) 1354e1367 1359

7.0

6.0

Consump on Ra os
5.0

4.0

3.0

2.0

1.0

0.0
0.0 20.0 40.0 60.0 80.0 100.0
Cycle Average Speed (km/h)

Fig. 2. Energy consumption ratios.

Table 3 Table 4
Energy Consumption Ratios (ICET/BET) for three different applications. Local taxes and tariffs for BETs.

Urban Cycle Regional Cycle Long-Haul Cycle Argentina Brazil Chile Colombia Uruguay

Average Speed (km/h) 20 40 80 Import duties 35% 35% 0% 0% 0%

Ratio 5.09 4.06 2.73 Tax on personal goods 0.25%
Statistics fee 0.5%
COFINS* 7.60%
PIS* 1.65%
electricity costs: 2% inflation, 0.5% deflation and constant price. Of Consular tax 5%
these scenarios, the last two seem the most likely since all Latin-
American countries follow the model shown in Section 1, i.e.
lower prices due to lower generation costs. the data found in Table A1-5. Maintenance costs are expressed in
US$/1000 km and will be taken as a percentage of the truck price.
The values that will be used are 0.115% for the HDT, 0.172% for the
3.3.2.2. Maintenance costs of BETs and ICETs. The maintenance MDT and 0.250% for the LDT as seen in Table A1-5.
costs of electric versus conventional vehicles are presented in
Table A1-4. According to the numbers on the table, the average cost
of maintenance for a BET can be taken as 60.5% of the maintenance 3.3.2.3. Insurance and motor tax. Insurances are complicated to
cost in an ICET. To determine the maintenance costs of the ICETs, quantify since their price tend to vary according to the insurance
several TCO and life cycle cost papers were examined, arriving to company, driver expertise, vehicle model and contract

Field Study Experimental Model Average = 910.000km

1,600,000

1,400,000
ery Useful Life (km)

1,200,000

1,000,000

800,000

600,000

400,000
B

200,000

Fig. 3. Battery useful life in km.

1360 M. Tanco et al. / Journal of Cleaner Production 228 (2019) 1354e1367

Table 5
Fuel and electricity prices.

Prices as of September 2018 (excl. VAT) Argentina Brazil Chile Colombia Uruguay

Fuel (USD/l) 0.744 0.797 0.815 0.613 1.008

Electricity (USD/kWh) 0.111 0.129 0.131 0.135 0.156

characteristics. Nonetheless, for this research the compulsory in- can be seen in Fig. 5. Chile and Uruguay remain the most adamant
surances were considered to have a common baseline. These countries regarding changes in the electricity price, mainly due to
amounts were specified in some cases (Brazil, Chile, Colombia) or the high diesel prices that already enable an early breakthrough of
calculated through an average of the market values in the BETs. Brazil and Argentina continue to be the most difficult markets
remaining cases (Argentina and Uruguay). No differentiation was to penetrate. In the other three countries, break-even is reached
done between ICET and BET. between 2019 and 2026 for all regional distribution weight classes.
Lastly, the motor tax is expressed as a percentage of the vehicle Nonetheless, LDTs and MDTs used in regional distribution seem like
purchase price stipulated by local legislation, with the exception of a feasible possibility for Argentina and Brazil, reaching equilibrium
Chile.4 It is considered to be the same for both vehicle types. as soon as 2023 for Argentina and 2024e2025 for Brazil. Long-haul
transportation is significantly affected by an energy price deflation
and makes the adoption of BETs a likely scenario, although it re-
4. Results mains a distant threshold. HDTs remain a challenge to overcome in
most of the countries -specially in long-haul distribution-, with the
In first place, applying the TCO model developed in MATLAB parity of TCOs far from being achieved. However, the difference
(explained in Section 3), the break-even year was found for each of between the weight classes is smaller than previously expected.
the 45 segments, these can be seen in Fig. 4 and Fig. 5. It should be This can be explained by the fact that larger trucks besides greater
remarked that for the model developed, no stochastic behaviour is purchase prices also have greater operational costs, and BETs
considered, and all values are predetermined for coming years. In benefit significantly from this. Uruguay could be the first to achieve
Fig. 4, the break-even years are shown for two different scenarios: an economically sound transition by 2022 in a best-case scenario
one that considers energy price as constant over time and the other (2026 with both fuel and energy inflation) and thus it will be
one that considers a 2% inflation in energy prices. The left end of studied in greater detail with a sensitivity analysis. Also, in
each arrow expresses the break-even year for the constant energy Uruguay, 4 out of 9 segments break even as early as 2019. Similarly,
price scenario. The right end of the arrow conveys the break-even Chile also presents 6 segments where BETs can breakthrough in a
year for the 2% inflation. Consequently, each arrow represents the short-term basis.
interval of potential TCO break-even points for the segment
analysed.
The wide array of parameters such as import tariffs, mainte- 5. Sensitivity analysis
nance costs, fuel and electricity costs for the different vehicles differ
between the countries in question, although the same period of The results obtained are limited to specific values and con-
ownership, discount rate, and inflation rates are applied. Factors as strained to various assumptions expressed in the methodology
vehicle availability, subjectivity of prices to the supply and demand section. Assumptions that might be subject to some variations in
law and any financial costs associated with the initial investment local market conditions (affecting purchase price) and uncertainty
are excluded. The opportunity of having regional production of of prices’ escalation.
BETs in future years is also dismissed. Lastly, the model will be run In order to evaluate the effect of the different variables in the
starting in the year 2019. TCO, the Uruguayan case study is selected to apply a sensitivity
The main result that can be extracted from Fig. 4 is the similar analysis by means of a Monte Carlo simulation. Uruguay is chosen
trend that the different weight classes are subject to within the since it has the closest break-even points to the actual date as well
different countries and distribution types, i.e. the HDTs are the last as a practically 100% renewable energy matrix which makes it an
to reach break-even while LDTs reach first the equilibrium. ideal location for developing electric mobility. The main goal of this
Furthermore, in countries like Argentina and Brazil (local manu- sensitivity analysis is to determine and understand the impact of
facturers of trucks) most cases result in BETs requiring a longer specific variables in the break-even years for the different segments
period of time to be able to compete with ICETs, mainly due to present in Uruguay.
heavy tariffs on foreign truck production. An interesting result is BETs' purchase price, energy price inflation and fuel price
found at Colombia's long-haul application, where all weight classes inflation are the parameters set to vary in the sensitivity analysis
face a long period of time before being competitive with conven- according to the values stated in Table 6. The BET price is assumed
tional trucks. The main reason behind this is the high electricity to vary between the actual market costs of BETs in Uruguay (an
costs versus the low diesel costs which are increased with longer elevated cost due to the lack of offer) and the theoretically calcu-
distances. Finally, regional distribution shows the earliest TCO lated one by this paper's model. An additional assumption is made
break-even year for all of the cases. This is due to the batteries being in which the gap between both prices narrows until equilibrium is
tailor-made for the required distance, while urban distribution reached in 2030, year in which sufficient offer is expected to arrive
requires a battery capable of delivering a minimum range of to the region.
200 km. These results are furtherly analysed and discussed in Fig. 6 shows the results of the Monte Carlo simulation when
Section 6. simultaneously varying the parameters as detailed in Table 6. It
In a 0.5% deflation on electricity prices scenario, results are portrays the probability for each year on achieving a smaller TCO
exacerbated, and the break-even is positioned in earlier years as for BETs than ICETs.
The most evident conclusions from Fig. 6 sustain the results
exhibited in Section 4, i.e. regional distribution is the most likely to
4
Decree 2385/96: https://www.leychile.cl/Navegar?idNorma¼18967. have an earlier break-even (long-haul last), and LDTs are the
 M. Tanco et al. / Journal of Cleaner Production 228 (2019) 1354e1367 1361

LDT MDT HDT

Fig. 4. Break-even year for each of the 45 segments defined. A constant energy price is considered alongside a 2% inflation scenario.

segment leading the transition towards electric mobility. A con- Argentina and Brazil charge on foreign trucks (BETs are assumed to
spicuous result is the fact that the most likely break-even year for come from China and EU) to protect their industry; making the
the urban application (exemplified by the highest peak of each initial BET price extremely high compared to the ICET's (that has no
curve) stands at least 5 years further away than those seen in import duties since they are produced locally), as shown on Fig. 7.
Section 4. Similarly, for regional distribution, MDTs' and HDTs’ Additionally, Brazil charges some other important taxes, such as PIS
break-even is also at least 5 years away. In long-haul distribution, (1,65%) and COFINS (7,6%), that ultimately position it as the last
the most probable results run closer to those seen in the Results country to reach a DTCO < 0. On a counterpart, Chile, Colombia and
section, with only a 2-year deviation. Therefore, the sensitivity Uruguay have no import duties on BETs, with Uruguay the only one
analysis displays the urban and regional distributions as more of this three that charges a 5% tax on imported EVs.
volatile to variations in parameters, while long-haul distribution, Fig. 7 exemplifies a specific case of a TCO analysis of the model
though sensitive, is less elastic to variations and the impact on the developed. It summarizes the total costs of ownership in all the
break-even is abated. countries for a particular segment, i.e. a MDT in a long-haul dis-
Some of the probability density functions present two crests. tribution application. It is easily perceived that for diesel trucks
This can be explained by the fact that the year 2019 accumulates the with big mileage, the main costs are accounted by fuel -which are
probability that the electric truck breaks-even, not only in that greater in Chile and Uruguay, favouring an earlier BET break-
same year, but also on previous years (2018, 2017, etc.). That is the through-, followed with similar weight in the TCO by maintenance
reason why these two crests can be seen mainly on the regional costs (trucks travel 1,250,000 km in 10 years) and purchase price.
application, where the break-even year for the base cases was 2019. Nevertheless, for BETs, the main cost is the purchase price which is
accentuated in the Argentinian and Brazilian cases, accounting for
6. Discussion over 50% of the lifetime cost of ownership. The same can be
perceived for a regional distribution portrayed in Fig. 8. In the latter,
Although uncertainty exists in future values of BETs and elec- the purchase price gains even more weight as the number of kil-
tricity and fuel costs, this study helps to achieve a wide perspective ometres covered over the vehicle's lifetime are reduced, together
on the insertion of this technology in South America. Firstly, the with fuel costs. If import duties were eliminated from these
paper finds that Chile and Uruguay -followed by Colombia-are the countries, or largely reduced, the TCO for BETs would reduce
main countries that could experience an early break-even of BETs' enormously. A tariff exemption for BETs (as applied in Chile,
TCO. The main reason behind this are the 35% import duties that Colombia and Uruguay) would allow Argentina and Brazil to
1362 M. Tanco et al. / Journal of Cleaner Production 228 (2019) 1354e1367

LDT MDT HDT

Fig. 5. Break-even year for each of the 45 segments. A constant 0.5% deflation scenario is considered alongside a 2% inflation scenario for energy prices.

Table 6
Parameters varied for sensitivity analysis.

Parameter Base case assumption Sensitivity analysis assumption

BET purchase Function of the segment studied. It is the addition of the truck price BET price stochastically varies between the theoretically modelled value and
price plus the battery selling price actual Uruguayan market price
Electricity price Electricity price escalation is set at 2% and specific scenarios at 0% Between 1 and 2 percent annual inflation
inflation and 0.5% are also studied
Fuel price Fuel price escalation is established at 2% Between 1 and 3 percent annual inflation
inflation

Fig. 6. Sensitivity analysis of input parameters on break-even year by segment.

diminish the impact of the purchase price on TCO, ultimately even though, as can be seen in Fig. 8, the TCO for the BET is larger
resulting in an earlier break-even year and adoption of a cleaner than in Colombia and Chile. This is justified by higher energy prices
transportation alternative, aligned with the Paris conference ob- and CIF cost for the BET.
jectives for 2030. Lastly, Uruguay is the first to achieve TCO parity For the example portrayed in Fig. 8, a tariff exemption would
 M. Tanco et al. / Journal of Cleaner Production 228 (2019) 1354e1367 1363

Fig. 7. Total costs of ownership for a long-haul MDT purchased in 2019. Comparison between Argentina, Brazil, Chile, Colombia and Uruguay.

Fig. 8. Total costs of ownership for a regional MDT purchased in 2019. Comparison between Argentina, Brazil, Chile, Colombia and Uruguay.

allow Argentina to break-even as soon as 2020 and 2024 for Brazil Despite the important percentage of the initial BET price in the TCO,
-contrasted with 2026 and 2029 respectively-. a greater mileage enables the possibility to assimilate a rise in the
Secondly, in accordance with previous studies (Lee et al., 2013; purchase cost and mitigate its impact.
Taefi et al., 2017), this study has found that vehicle mileage heavi- A third important result found by the paper, is the challenge for
ly impacts the TCO. Figs. 4 and 5 evidence the sensitivity that en- these countries to successfully and profitably incorporate HDTs into
ergy price evolution in time has on the TCO. This is furtherly their fleets, specifically in a long-haul distribution. Assuming that
emphasized by the sensitivity analysis made, in which the effect of the technology reaches the continent, Uruguay would be the first
energy and fuel evolution together with the impact of initial pur- country to break-even in this segment and it would be by 2025,
chase price is studied. The model expresses great sensitivity to considering a 2% inflation scenario. Opposed to what other studies
these parameters which, when varied simultaneously, can affect have shown, high usage and larger BEVs are not more cost-
the break-even within a 7-year range. For example, trucks used in competitive than ICEVs when considering the long-term costs of
an urban distribution in Uruguay achieve TCO parity by 2019e2020 ownership. The main cause behind this is again the soaring pur-
according to the base model, whereas the sensitivity analysis chase price that every country must face for HDTs in long-haul
demonstrates that break-even is most likely to be reached by application, since the battery's capacity needed is 1000 kWh -see
2026e2027. This is explained by the high percentage that the initial Table 1-. Due to the high battery prices in current years, the size of
purchase price represents in the TCO of a BET in comparison to the battery results on a high initial investment, negatively affecting
other costs, mainly due to the low mileage of trucks in this appli- the TCO. The relatively low taxes on diesel also affect negatively the
cation. Hence, growths in the purchase price will negatively affect TCO. Taxes on fuel in Latin America are lower than in most Euro-
the BETs TCO, distancing the break-even year further into the pean countries, resulting in lower fuel costs and finally lower
future. Oppositely, in a long-haul application, trucks are more operative costs; not allowing a grander impact with greater usage.
resilient to changes in parameters in the sensitivity analysis. As Mckinsey's (2017) study states, a high price differential between
1364 M. Tanco et al. / Journal of Cleaner Production 228 (2019) 1354e1367

fuel costs and electricity prices has a stronger effect in accom- in Latin America are scarce or do not exist at all. More importantly,
plishing faster TCO parity. this study analyses which segments will be the first to break-even
In the particular Colombian case, long-haul applications appear in the different countries studied, using a TCO framework. It also
as unachievable. At a first glance it might seem odd, since regional finds and portrays the break-even year for each segment consid-
and urban distribution resulted as a reliable option before 2026. ering different variations in electricity prices. The main results
Nonetheless, Colombia presents the lowest diesel costs and the obtained are that Chile and Uruguay present themselves as the
second highest energy costs of the examined countries, thus front-runners in achieving TCO parity between BETs and ICETs
resulting in a lower absolute price differential. This relationship while Argentina and Brazil lag behind due to high import duties
between fuel costs, gains momentum the longer the distances and low purchase costs for conventional trucks. The penetration of
travelled are. Therefore, the existent gap between ICETs' and BETs’ the weight classes follows the same pattern in every application:
purchase price becomes a steeper hill to overcome when the cost first LDTs, then MDTs and lastly HDTs, with the latter struggling to
difference between the fuel alternatives is not as effective as with reach break-even in long-haul distribution. Regional distribution is
the other countries. A reduction in energy costs, or alternatively, a the first application to achieve TCO parity, basically due to low
raise in diesel costs would result positively in the final DTCO. upfront costs resulting from smaller batteries -that currently have
Analogically, LDTs and MDTs have the possibility of breaking elevated prices- and long enough distances to narrow the differ-
even earlier as a consequence of a lower initial investment. This is ence in initial investment through operational costs.
exemplified in Fig. 9 where the initial investment (purchase price Another key conclusion of this article is the leverage that the
plus EVSE) embodies 59% of the BET's TCO, while operative costs initial investment has on the BETs' TCO, representing an important
(electricity and maintenance) represent 33%. Meanwhile, for the portion of lifetime costs. A sensitivity analysis was performed to
ICET, the initial investment is just 29% of the lifetime costs, while evaluate the impact of the initial investment confirming its rele-
operative costs are 62%. This evidences the wide gap between the vance on BETs' competitiveness. The insights obtained through the
purchase costs of both alternatives. sensitivity analysis show that TCO equilibrium is extremely sensi-
Therefore, this paper confirms what other TCO calculations in tive to variations on the purchase price (together with energy and
literature have identified as the main aspect negatively affecting fuel costs), causing delays of up to 7 years in the break-even points.
BETs’ lifetime costs: initial investment. More specifically, the Findings suggest that subsidizing EVs purchase price or relieving
upfront purchase price. This issue is emphasized as the weight class import duties in the Argentinian and Brazilian cases would be
and mileage grow. This can also be seen in the sensitivity analysis successful policies to accelerate BETs' adoption. Independently of
where a variation on the purchase price prolongs the break-even this, the model used in this study exhibits that despite current
between 3 and 7 years. Surely, operative costs play their role and difficulties to financially compete with diesel trucks, BETs will
indeed encompass an important portion of the TCO but tackling the become a profitable option in the short-term, mainly in Chile,
purchase price would have the greatest impact in achieving Colombia and Uruguay for urban and regional distribution. This
improved competitiveness for BETs. This insight might be inter- new technology's competitiveness is not as distant as it once was.
esting for Latin-American policy makers and governments who Moreover, If Latin America is able to translate its growing renew-
could promote tax incentives and/or adopt subsidies for the pur- able energy matrix into lower energy prices, BETs will be able to
chase of BETs thus reducing the upfront costs and enabling a penetrate every segment in each country within one or two
greater return on investment for fleet managers. The options decades.
highlighted in Taefi et al. (2017) such as the V4.000 subsidy scheme A final aspect to consider in this paper are the key assumptions
adopted by the German government in 2016 for EVs with a price up made which eventually contribute to add uncertainty into the
to V60.000, or the high purchase price subsidies for medium-duty analysis. One key assumption is the evolution of fuel and electricity
EVs in Netherlands are sound examples that Latin America could prices which undoubtedly have a heavy impact on TCO calculation.
follow. Nonetheless, fuel prices are constantly shifting and dependent on
global events which make them unpredictable in the long-term.
7. Conclusions Neglecting a battery replacement cost in the lifetime of the BETs
is also a key assumption scarcely seen on literature. Still, with the
This study conducts a literature review through which the state- increasing technology and increased battery guarantees offered by
of-the-art BEV's aspects are presented. A current gap on literature is manufacturers, TCO studies will start to exclude this cost. This
fulfilled by this paper since TCO studies comparing BETs and ICETs assumption positively affects the TCO for BETs, bringing the parity

Fig. 9. Total costs of ownership for a regional MDT purchased in 2020, Colombia.
 M. Tanco et al. / Journal of Cleaner Production 228 (2019) 1354e1367 1365

nearer to the present, even more so, for vehicles that include bat- purchase prices become available, more accurate economic analysis
teries with larger capacity (Weldon et al., 2018). Finally, as Morrison can be developed. Real-world consumptions and maintenance data
et al. (2018) remark, several relevant variables were not considered, will also become available with the introduction of BETs in the
such as the importance of non-monetary factors like symbolic and continent, eventually resulting in more refined studies.
functional benefits of vehicles to consumers. This paper makes a
strictly economic analysis.
Future studies could validate and update the results of this pa- Appendix I
per and utilize this study as the cornerstone for more investigations
on BETs' penetration in Latin America. As more data on BETs’

Abbreviations ICET Internal combustion engine truck

BET Battery-electric truck LCV Light commercial vehicle
BEV Battery-electric vehicle LDT Light-duty truck
DOD Depth of discharge MDT Medium duty truck
EOL End of life NCA Niquel-cobalt-aluminum
EV Electric vehicle NMC Niquel-cobalt-manganese
EVSE Electric vehicle supply equipment SOH State of health
GVW Gross vehicle weight TCO Total cost of ownership
HDT Heavy-duty truck VAT Value added tax

Appendix II

Table A1
.1. Market Price of ICETs in USD (excluding VAT)

Argentina Brazil Chile Colombia Uruguay

LDT 27,824 20,551 20168 25,910 27,869

MDT 43,526 38,489 44,118 36,245 42,623
HDT - Rigid 89,426 83,206 81,611 88,385 102,254
HDT - Tractor 115,702 103410 98,863 109,847 125,137

Table A2
2. BET price as a % of ICET cost

Truck type ICET price ($) BET price w/o battery ($) BET price % of ICET Source

HDT 107,362 129,040 120% Sen et al. (2017)

HDT 108,000 100,000 92,6% (California Air Resources Board (CARB), 2015)
HDT 120,000 75,000 62,5% Mareev et al. (2018)
MDT 75,000 90,000 120% Lee & Thomas (2017)
Average 98.8% Own

Li-ion Batteries Useful Life

Table A3
3. Battery life expectancy till reaching 80% SOC

Li-ion Battery life Description Source Classification

(80% SOH)

6000 cycles Cycle life of 94Ah Samsung cells at 25  C, 100% DOD, 1C/1C charge/discharge rate and 80% SoC Samsung SDI (2018) Experimental
840.000 km Field study of on the road Tesla's Model S and X (aprox. 340 vehicles) in which 90% capacity is achieved Steinbach, Coumans, & Belgium Field Study
after 300.000 km Tesla Forum (2018)
1000-8000 Gathered from the bibliographic research done Ahmadian et al. (2018) Model
cycles
3600 cycles Results of a dynamic stress test done in order to illustrate the battery's performance in mixed-cycle Xu et al. (2018) Experimental
operations. 25e100% SoC at 20  C
1000-2000 Battery electric trucks cycled with large variations of SOC den Boer et al. (2013) Experimental
cycles
>3500 cycles For a 60e70% SOC. Deduced from a model for the calculations of battery life cycle depending on battery Pelletier et al. (2017) Model
SoC at 25  C and using parameters form; Sarasketa-Zabala et al. (2013).
5.5 years EV battery degradation is explored in four different cases. Driving and uncontrolled charging results Wang et al. (2016) Model
shown.
6.8 years Experimental result with following parameters: 7 kW charging, 17A charge/discharge rate and 40% Lacey et al. (2013) Experimental
change in SOC
1366 M. Tanco et al. / Journal of Cleaner Production 228 (2019) 1354e1367

ICE and BE Trucks Maintenance Costs.

Table A4
4. BET maintenance costs as a % of ICET costs

BEV Maintenance Costs as a % of Description Source

ICEV costs

50% Estimation of electric truck maintenance costs Feng & Figliozzi

(2012)
75% Maintenance costs for electric freight vehicles. Study analyses collected data from the current Quak et al. (2016)
European project FREVUE.
40% Electric truck maintenance costs compared to those of diesel trucks Zhou et al. (2017)
70% Values obtained using GREET model (Greenhouse gases, Regulated Emissions, and Energy use in Lee & Thomas
Transportation Model) (2017)
50% - Lebeau et al. (2015)
82% - Weldon et al.
(2018)
57% Maintenance cost for diesel trucks derived from real tests Zhao et al. (2016a)

Average 60.5% Own

Table A5
5. ICET's maintenance costs in $/1000 km as a % of their price

Truck Purchase price Maintenance costs % of purchase Comments Source

class ($) ($/1000 km) price

HDT 107,362 121 0.113 Modelled diesel truck Sen et al. (2017)
114,000 132 0.116 Real diesel truck (European Federation for Transport and
Environment, 2018)
MDT 50,000 81 0.162 Real diesel truck Zhao et al. (2016b)
36,000 50 0.140 Dyna diesel truck Taefi et al. (2017)
70,000 170 0.243 Ford F-650 diesel Zhou et al. (2017)
75,000 125 0.167 Real diesel truck Lee & Thomas (2017)
50,000 84 0.168 Freightliner P70 diesel Zhao et al. (2016a)
LDT 50,000 125 0.250 Isuzu diesel. When truck gets.to age 9: Feng & Figliozzi (2012)
0,700%
LCV 16,500 47 0.284 Real diesel vehicle Lebeau et al. (2015)

HDT 0.115 Own

MDT 0.175 Own
LDT 0.250 Own

References electric delivery trucks. Transport. Res. E Logist. Transport. Rev. 49 (1), 8e23.
https://doi.org/10.1016/j.tre.2012.07.003.
den Boer, E., Aarnik, S., Kleiner, F., Pagenkopf, J., 2013. Zero Emissions Trucks: an
Ahmadian, A., Sedghi, M., Elkamel, A., Fowler, M., Aliakbar Golkar, M., 2018. Plug-in
Overview of State-Of-The-Art Technologies and Their Potential. The Intena-
electric vehicle batteries degradation modeling for smart grid studies: review,
tional Council for Clean Transportation (ICCT), Delft, Netherlands. https://doi.
assessment and conceptual framework. Renew. Sustain. Energy Rev. 81,
org/13.4841.21.
2609e2624 (July 2016). https://doi.org/10.1016/j.rser.2017.06.067.
U.S. Department of Energy (DOE), 2017. Cost and Price Metrics for Automotive
The Boston Consulting Group (BCG), 2010. Batteries for Electric Cars: Challenges,
Lithium-Ion Batteries. Retrieved from. https://energy.gov/sites/prod/files/2017/
Oportunities, and the Outlook to 2020. Retrieved from. http://www.bcg.com/
02/f34/67089 EERE LIB cost vs price metrics r9.pdf.
documents/file36615.pdf.
Douris, C., 2017. The Bottom Line on Electric Cars: They’re Cheaper to Own.
Berckmans, G., Messagie, M., Smekens, J., Omar, N., Vanhaverbeke, L., Mierlo, J. Van,
Retrieved July 17, 2018, from. https://www.forbes.com/sites/constancedouris/
2017. Cost projection of state of the art lithium-ion batteries for electric vehicles
2017/10/24/the-bottom-line-on-electric-cars-theyre-cheaper-to-own/
up to 2030. Energies 10 (9). https://doi.org/10.3390/en10091314.
#3057ae1910b6.
Bloomberg New Energy Finance (BNEF), 2017. Lithium-ion Battery Costs and Mar-
U.S. Energy Information Administration (EIA), 2016. Motor Vehicle Mileage, Fuel
ket. Retrieved from. https://data.bloomberglp.com/bnef/sites/14/2017/07/BNEF-
Consumption and Fuel Economy. Retrieved September 3, 2018, from. https://
Lithium-ion-battery-costs-and-market.pdf.
www.eia.gov/opendata/qb.php?category¼711246.
Bloomberg New Energy Finance (BNEF), 2018. Electric cars may Be cheaper than gas
Feng, W., Figliozzi, M.A., 2012. Conventional vs electric commercial vehicle fleets: a
guzzlers in seven years. Retrieved September 20, 2018, from. https://www.
case study of economic and technological factors affecting the competitiveness
bloomberg.com/hyperdrive.
of electric commercial vehicles in the USA. In: Procedia - Social and Behavioral
CalHEAT Research Center, 2013. Battery Electric Parcel Delivery Truck Testing and
Sciences, vol. 39, pp. 702e711. https://doi.org/10.1016/j.sbspro.2012.03.141.
Demonstration. Prepared for California Energy Comission, Pasadena. Retrieved
Global Petrol Prices, 2018. Diesel Prices. Retrieved September 17, 2018, from.
from. http://www.calstart.org/Libraries/CalHEAT_2013_Documents_
https://es.globalpetrolprices.com/diesel_prices/.
Presentations/Battery_Electric_Parcel_Delivery_Truck_Testing_and_
Gustafsson, T., Johansson, A., 2015. Comparison between Battery Electric Vehicles
Demonstration.sflb.ashx.
and Internal Combustion Engine Vehicles Fueled by Electrofuels. Master's
California Air Resources Board (CARB), 2015. Technology Assessment: Medium- and
Thesis in Sustainable Energy Systems. Chalmers University of Technology.
Heavy- Duty Battery Electric Trucks and Buses. Retrieved from. http://www.arb.
Retrieved from. http://publications.lib.chalmers.se/records/fulltext/218621/
ca.gov/msprog/tech/techreport/bev_tech_report.pdf.
218621.pdf.
California Air Resources Board (CARB), 2018. Battery Electric Truck and Bus Energy
Harrington, W., Keupnick, A., 2012. Improving Fuel Economy in Heavy-Duty Vehi-
Efficiency Compared to Conventional Diesel Vehicles. Retrieved from. https://
cles. Resources for the Future (RFF). https://doi.org/10.2139/ssrn.2038842.
www.arb.ca.gov/msprog/actruck/docs/HDBEVefficiency.pdf.
Helbig, C., Bradshaw, A.M., Wietschel, L., Thorenz, A., Tuma, A., 2018. Supply risks
da Silva, R.E., Sobrinho, P.M., de Souza, T.M., 2018. How can energy prices and
associated with lithium-ion battery materials. J. Clean. Prod. 172, 274e286.
subsidies accelerate the integration of electric vehicles in Brazil? An economic
https://doi.org/10.1016/j.jclepro.2017.10.122.
analysis. Electr. J. 31 (3), 16e22. https://doi.org/10.1016/j.tej.2018.03.007.
Holland, M., 2018. $100/kWh Tesla Battery Cells This Year, $100/kWh Tesla Battery
Davis, B.A., Figliozzi, M.A., 2013. A methodology to evaluate the competitiveness of
Packs in 2020. Retrieved September 20, 2018, from. https://cleantechnica.com/
 M. Tanco et al. / Journal of Cleaner Production 228 (2019) 1354e1367 1367

2018/06/09/100-kwh-tesla-battery-cells-this-year-100-kwh-tesla-battery- Quantifying uncertainties in reusing lithium-ion batteries from electric vehi-

packs-in-2020/. cles. Procedia Manuf. 8, 603e610. October 2016. https://doi.org/10.1016/j.
International Council on Clean Transportation (ICCT), 2017. Fuel Efficiency Tech- promfg.2017.02.077.
nology in European Heavy-Duty Vehicles: Baseline and Potential for the 2020- Samsung, S.D.I., 2018. ESS batteries by samsung SDI: top safety & reliability solu-
2030 Time Frame. Berlin. Retrieved from. https://www.theicct.org/publications/ tions. Retrieved from. http://www.samsungsdi.com/upload/ess_brochure/
fuel-efficiency-technology-european-heavy-duty-vehicles-baseline-and- 201803_SamsungSDI ESS_EN.pdf.
potential-2020. Sen, B., Ercan, T., Tatari, O., 2017. Does a battery-electric truck make a difference? e
International Energy Agency (IEA), 2017a. Global EV Outlook 2017: Two Million and life cycle emissions, costs, and externality analysis of alternative fuel-powered
Counting. https://doi.org/10.1787/9789264278882-en. Class 8 heavy-duty trucks in the United States. J. Clean. Prod. 141, 110e121.
International Energy Agency (IEA), 2017b. The Future of Trucks: Implications for https://doi.org/10.1016/j.jclepro.2016.09.046.
Energy and the Environment, second ed. Retrieved from. https://www.iea.org/ Speirs, J., Contestabile, M., Houari, Y., Gross, R., 2014. The future of lithium avail-
publications/freepublications/publication/ ability for electric vehicle batteries. Renew. Sustain. Energy Rev. 35, 183e193.
TheFutureofTrucksImplicationsforEnergyandtheEnvironment.pdf. https://doi.org/10.1016/j.rser.2014.04.018.
Inter-American Development Bank, 2016. The Incorporation of Electric Cars in Latin Sripad, S., Viswanathan, V., 2017. Performance metrics required of next-generation
America. Transport Division. https://doi.org/https://doi.org/10.18235/0000394. batteries to make a practical electric semi truck. ACS Energy Lett. 2 (7),
Jiao, N., Evans, S., 2016. Business models for sustainability: the case of second-life 1669e1673. https://doi.org/10.1021/acsenergylett.7b00432.
electric vehicle batteries. Procedia CIRP 40, 250e255. In: https://doi.org/10. Steinbach, M., Coumans, M., Belgium, Tesla Forum, 2018. Tesla Model S Battery
1016/j.procir.2016.01.114. Degradation Data. Retrieved June 14, 2018, from. https://steinbuch.wordpress.
Lacey, G., Jiang, T., Putrus, G., Kotter, R., 2013. The effect of cycling on the state of com/2015/01/24/tesla-model-s-battery-degradation-data/.
health of the electric vehicle battery. In: 48th International Universities Power European Federation for Transport and Environment (T&E), 2018. Analysis of Long
Engineering Conference (UPEC), p. 12. https://doi.org/10.1109/UPEC.2013. Haul Battery Electric Trucks in EU. Amended paper. Graz, 17-18 May 2018.
6715031. Retrieved from. https://www.transportenvironment.org/publications/analysis-
Lebeau, P., Macharis, C., Mierlo, J. Van, Lebeau, K., 2015. Electrifying light com- long-haul-battery-electric-trucks-eu.
mercial vehicles for city logistics?A total cost of ownership analysis. Eur. J. Taefi, T.T., Stütz, S., Fink, A., 2017. Assessing the cost-optimal mileage of medium-
Transp. Infrastruct. Res. 15 (4), 551e569. Retrieved from. http://www.tbm. duty electric vehicles with a numeric simulation approach. Transport. Res.
tudelft.nl/fileadmin/Faculteit/TBM/Onderzoek/EJTIR/Back_issues/15.4/2015_ Transport Environ. 56 (September), 271e285. https://doi.org/10.1016/j.trd.2017.
04a_03.pdf. 08.015.
Lee, D.Y., Thomas, V.M., 2017. Parametric modeling approach for economic and United Nations Environment Programme (UNEP), 2011. Recycling Rates of Metals.
environmental life cycle assessment of medium-duty truck electrification. Retrieved from. http://wedocs.unep.org/handle/20.500.11822/8702.
J. Clean. Prod. 142, 3300e3321. https://doi.org/10.1016/j.jclepro.2016.10.139. United Nations Environment Programme (UNEP), 2016. Electric Mobility: Opor-
Lee, D.Y., Thomas, V.M., Brown, M.A., 2013. Electric urban delivery trucks: energy tunities for Latin America. Retrieved from. http://www.pnuma.org/cambio_
use, greenhouse gas emissions, and cost-effectiveness. Environ. Sci. Technol. 47 climatico/publicaciones/informe_movilidad_electrica.pdf.
(14), 8022e8030. https://doi.org/10.1021/es400179w. United States Advanced Battery Consortium (USABC), 1996. Electric Vehicle Battery
March
an, E., Viscidi, L., 2015. Green Transportation: the Outlook for Electric Vehicles Test Procedures Manual. https://doi.org/DOE/ID-10597.
in Latin America. The Dialogue: Leadership for the Americas. City of Mexico. U.S. Geological Survey (USGS), 2018. Mineral Commodity Summaries 2018. Reston,
Retrieved from. https://www.thedialogue.org/analysis/green-transportation- Virginia. https://doi.org/10.3133/70194932.
electric-vehicles-in-latin-america/. Veps€al€ainen, J., 2017. Driving style comparison of city buses: electric vs. Diesel. In:
Mareev, I., Becker, J., Sauer, D.U., 2018. Battery dimensioning and life cycle costs 2017 IEEE Vehicle Power and Propulsion Conference (VPPC) (P. 5). Belfort,
analysis for a heavy-duty truck considering the requirements of long-haul France. https://doi.org/10.1109/VPPC.2017.8330942.
transportation. Energies 11 (1). https://doi.org/10.3390/en11010055. Vikstro€ m, H., Davidsson, S., Ho €o€k, M., 2013. Lithium availability and future pro-
Mckinsey & Company, 2017. What's Sparking Electric-Vehicle Adoption in the Truck duction outlooks. Appl. Energy 110, 252e266. https://doi.org/10.1016/j.
Industry? Retrieved April 10, 2018, from. https://www.mckinsey.com/ apenergy.2013.04.005.
industries/automotive-and-assembly/our-insights/whats-sparking-electric- Wang, D., Coignard, J., Zeng, T., Zhang, C., Saxena, S., 2016. Quantifying electric
vehicle-adoption-in-the-truck-industry. vehicle battery degradation from driving vs. vehicle-to-grid services. J. Power
Morrison, G., Stevens, J., Joseck, F., 2018. Relative economic competitiveness of light- Sources 332, 193e203. https://doi.org/10.1016/j.jpowsour.2016.09.116.
duty battery electric and fuel cell electric vehicles. Transport. Res. C Emerg. Weldon, P., Morrissey, P., O'Mahony, M., 2018. Long-term cost of ownership
Technol. 87, 183e196 (June 2017). https://doi.org/10.1016/j.trc.2018.01.005. comparative analysis between electric vehicles and internal combustion engine
Narins, T.P., 2017. The battery business: lithium availability and the growth of the vehicles. Sustain. Cities Soc. 39, 578e591. October 2017. https://doi.org/10.1016/
global electric car industry. Extr. Ind. Soc. 4 (2), 321e328. https://doi.org/10. j.scs.2018.02.024.
1016/j.exis.2017.01.013. Xu, B., Oudalov, A., Ulbig, A., Andersson, G., Kirschen, D.S., 2018. Modeling of
Nykvist, B., Nilsson, M., 2015. Rapidly falling costs of battery packs for electric ve- lithium-ion battery degradation for cell life assessment. IEEE Trans. Smart Grid
hicles. Nat. Clim. Change 5 (4), 329e332. https://doi.org/10.1038/nclimate2564. 9 (2), 1131e1140. https://doi.org/10.1109/TSG.2016.2578950.
Pehlken, A., Albach, S., Vogt, T., 2017. Is there a resource constraint related to lithium Yun, L., Linh, D., Shui, L., Peng, X., Garg, A., LE, M.L.P., et al., 2018. Metallurgical and
ion batteries in cars? Int. J. Life Cycle Assess. 22 (1), 40e53. https://doi.org/10. mechanical methods for recycling of lithium-ion battery pack for electric ve-
1007/s11367-015-0925-4. hicles. Resour. Conserv. Recycl. 136, 198e208. https://doi.org/10.1016/j.
Pelletier, S., Jabali, O., Laporte, G., Veneroni, M., 2017. Battery degradation and resconrec.2018.04.025.
behaviour for electric vehicles: review and numerical analyses of several Zhao, Y., Noori, M., Tatari, O., 2016a. Vehicle to Grid regulation services of electric
models. Transp. Res. Part B Methodol. 103, 158e187. https://doi.org/10.1016/j. delivery trucks: economic and environmental benefit analysis. Appl. Energy
trb.2017.01.020. 170, 161e175. https://doi.org/10.1016/j.apenergy.2016.02.097.
Quak, H., Nesterova, N., Van Rooijen, T., Dong, Y., 2016. Zero emission city logistics: Zhao, Y., Onat, N.C., Kucukvar, M., Tatari, O., 2016b. Carbon and energy footprints of
current practices in freight electromobility and feasibility in the near future. electric delivery trucks: a hybrid multi-regional input-output life cycle assess-
Transport. Res. Procedia 14, 1506e1515. https://doi.org/10.1016/j.trpro.2016.05. ment. Transport. Res. Transport Environ. 47, 195e207. https://doi.org/10.1016/j.
115. trd.2016.05.014.
Rahman, A., Afroz, R., Safrin, M., 2017. Recycling and disposal of lithium batteries : Zhou, T., Roorda, M.J., MacLean, H.L., Luk, J., 2017. Life cycle GHG emissions and
an economical and environmental approach. IIUM Eng. J. 18 (2), 238e252. lifetime costs of medium-duty diesel and battery electric trucks in Toronto,
Reinhart, T.E., 2015. Commercial Medium- and Heavy-Duty Truck Fuel Efficiency Canada. Transport. Res. Transport Environ. 55, 91e98. https://doi.org/10.1016/j.
Technology Study. Report #1. National Highway Traffic Safety Administration, trd.2017.06.019.
Washington, DC. Retrieved from. https://www.nhtsa.gov/sites/nhtsa.dot.gov/ Zubi, G., Dufo-Lo
pez, R., Carvalho, M., Pasaoglu, G., 2018. The lithium-ion battery:
files/812146-commercialmdhd-truckfuelefficiencytechstudy-v2.pdf. state of the art and future perspectives. Renew. Sustain. Energy Rev. 89,
Rohr, S., Müller, S., Baumann, M., Kerler, M., Ebert, F., Kaden, D., Lienkamp, M., 2017. 292e308. October 2017. https://doi.org/10.1016/j.rser.2018.03.002.