Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=ujst20
International Journal of Sustainable Transportation
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/ujst20
Life cycle assessment as decision-support
in choice of road corridor: case study and
stakeholder perspectives
Carolina Liljenström , Sofiia Miliutenko , Reyn O’Born , Helge Brattebø ,
Harpa Birgisdóttir , Susanna Toller , Kristina Lundberg & José Potting
To cite this article: Carolina Liljenström , Sofiia Miliutenko , Reyn O’Born , Helge Brattebø , Harpa
Birgisdóttir , Susanna Toller , Kristina Lundberg & José Potting (2020): Life cycle assessment as
decision-support in choice of road corridor: case study and stakeholder perspectives, International
Journal of Sustainable Transportation, DOI: 10.1080/15568318.2020.1788679
To link to this article: https://doi.org/10.1080/15568318.2020.1788679
© 2020 The Author(s). Published with
license by Taylor and Francis Group, LLC
Published online: 14 Jul 2020.
Submit your article to this journal
View related articles
View Crossmark data
Life cycle assessment as decision-support in choice of road corridor: case study
and stakeholder perspectives
Carolina Liljenstr€
om
a
, Sofiia Miliutenko
b
, Reyn O’Born
c
, Helge Brattebø
d
, Harpa Birgisd
ottir
e
, Susanna Toller
f
,
Kristina Lundberg
g
, and Jos
e Potting
a,h,i
a
Division of Sustainability Assessment and Management, Department of Sustainable Development, Environmental Science and Engineering,
KTH Royal Institute of Technology, Stockholm, Sweden;
b
IVL Swedish Environmental Research Institute, Stockholm, Sweden;
c
Faculty of
Engineering and Science, University of Adger, Grimstad, Norway;
d
Department of Energy and Process Engineering, Norwegian University of
Science and Technology, Trondheim, Norway;
e
Danish Building Research Institute, Aalborg University, København, SV, Denmark;
f
Swedish
Transport Administration, Solna, Sweden;
g
Ecoloop, Strategic Services & Sustainable Development, Stockholm, Sweden;
h
EnviroSpotting,
Wageningen, the Netherlands;
i
ECO2 Vehicle Design, Stockholm, Sweden
ABSTRACT
Use of life cycle assessment (LCA) in choice of road corridor could reduce environmental impacts
of traffic and infrastructure. This paper explores how the LCA model LICCER, designed to compare
life cycle climate impact and energy use of alternative road corridors, fulfills practitioners’require-
ments concerning data availability and usefulness for decision-making. Results are based on a case
study where the model was applied to a Swedish road reconstruction project and a workshop
with potential users of the model. In the case study, the shorter construction alternatives had the
lowest traffic related impacts and the highest infrastructure related impacts. Earthworks, soil stabil-
ization, and pavement contributed most to infrastructure related impacts. For the stakeholders,
the LICCER model was considered useful because it includes both traffic and infrastructure,
includes default data that the user can replace by project specific data, identifies possible
improvements, and presents results relative to a reference alternative. However, the model could
be improved by including further nation specific default data, different traffic scenarios depending
on the road corridor, more detailed traffic scenarios, and an uncertainty assessment of the model
output. These findings may be useful in the development and improvement of LCA models and
when evaluating the suitability of existing models for use in early planning.
ARTICLE HISTORY
Received 19 March 2018
Revised 7 February 2020
Accepted 21 June 2020
KEYWORDS
greenhouse gas emissions;
infrastructure planning; life
cycle assessment; primary
energy use; road corridor;
stakeholder participation
1. Introduction
The transport sector is a significant contributor to climate
change (Blanco et al., 2014), accounting for 24% of global
energy related greenhouse gas (GHG) emissions
(International Energy Agency, 2018). The majority of these
emissions (74%) are due to road transport (International
Energy Agency, 2018). In addition, construction, operation,
maintenance, and demolition of road infrastructure results
in GHG emissions. In 2009, the construction sector emitted
5.7 billion tonnes CO
2
globally, making construction one of
the largest carbon emitting sectors (Huang et al., 2018).
Several studies have assessed the contribution of road infra-
structure to the overall environmental impacts of road trans-
port. Based on a review of such studies, Hill et al. (2012)
concluded that road infrastructure accounted for 10-40% of
road transport GHG emissions, depending on factors such
as traffic volume, surface type, and maintenance measures.
Policy-makers in the transport sector show a growing
interest in life cycle assessment (LCA) as a decision-support
tool to reduce the climate impact of road infrastructure.
This interest can be seen for example in the financing of
research projects under the ERA-NET road program
Sustainability and Energy Efficient Management of Roads
(Carlson & Folkeson, 2014). Other examples are the devel-
opment of EU Green Public Procurement Criteria for road
infrastructure (Garbarino et al., 2016), and use of LCA in
planning and procurement of road infrastructure by the
Swedish Transport Administration (Toller & Larsson, 2017).
To reduce the environmental impact of road transport, a
life cycle perspective is called for throughout the whole
planning process of a construction project (Huang et al.,
2015): (1) in choice of transport mode on national and pro-
ject level, (2) in choice of road corridor and construction
type (plain road, bridge, and tunnel) of a specific project,
and (3) in choice of specific construction design (Miliutenko
et al., 2014).
The greatest opportunity to influence life cycle impacts of
transport occurs in early planning stages (Hammervold,
2014; Karlsson et al., 2017;O’Born et al., 2016), such as
choice of road corridor and construction type. The choice of
ß2020 The Author(s). Published with license by Taylor and Francis Group, LLC
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/
4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in
any way.
CONTACT Carolina Liljenstr€
om carlil@kth.se Division of Sustainability Assessment and Management, Department of Sustainable Development,
Environmental Science and Engineering, KTH Royal Institute of Technology, Teknikringen 10B, 114 28 Stockholm, Sweden.
INTERNATIONAL JOURNAL OF SUSTAINABLE TRANSPORTATION
https://doi.org/10.1080/15568318.2020.1788679
road corridor influences route length and construction type
and has thereby a large influence on environmental impacts
from future traffic on the road and on impacts from road
construction, operation, and maintenance. It was found that,
when comparing alternative routes for road construction,
energy savings of up to 47% could be achieved in construc-
tion when choosing the route option with the lowest energy
requirements (‘Energy conservation in road pavement
design’[ECRPD], 2010).
At the same time there is limited access to project spe-
cific data, such as type and quantity of construction materi-
als, required to complete the LCA (Butt et al., 2015; Kluts &
Miliutenko, 2012; Oduro & Lautala, 2017). Such information
does not become available until the design stage (Butt et al.,
2015), when opportunities to influence life cycle impacts are
lower. This problem can be referred to what Bhander et al.
(2003) call the paradox of eco-design: while product know-
ledge increases over time, the possibilities for improving
environmental performance decrease. Product knowledge is
greatest by the end of the process, but then only minor
changes can be made to product design. Consequently, in
early planning of road construction, environmental perform-
ance must be evaluated based on a limited amount of
input data.
Several LCAs have been conducted for road infrastructure
projects that are either under construction or have already
been completed (see for example Barandica et al., 2013;
Bouhaya et al., 2009; Du et al., 2014; Gulotta et al., 2018;
Guo et al., 2019; Huang et al., 2015; Liu et al., 2019;
Miliutenko et al., 2012;O’Born, 2018; Stripple, 2001; Trunzo
et al., 2019). Such studies are useful in showing which
aspects are most significant to include in an LCA made in
early planning and in providing a source of inventory data
(Huang et al., 2015); however, they do not consider the spe-
cific challenges occurring in early planning.
To support the integration of LCA in decision-making,
several LCA models with different scope have been devel-
oped, for example: asPECT (Wayman et al., 2012), CEREAL
(Spriensma et al., 2014), CHANGER (Huang et al., 2013),
CO
2
NSTRUCT (Fern
andez-S
anchez et al., 2015), Dubo-Calc
(Cenosco & Royal HaskoningDHV, n.d.), ECORCE (Jullien
et al., 2015), EFFEKT (Sandvik & Hammervold, 2011;
Straume & Bertelsen, 2015), Carbon Tool (Highways
England, 2016), Joulesave (ECRPD, 2010), Klimatkalkyl
(Toller & Larsson, 2017), LICCER (Potting, Birgisd
ottir,
Brattebø, Kluts et al., 2013), as well as others reviewed by
Miliutenko et al. (2014) and Santos et al. (2017).
However, many of these models require input data not
available in early planning of road construction. Previous
research (Miliutenko et al., 2014) has found that only
EFFEKT, Joulesave, LICCER, and Klimatkalkyl have been
developed for choice of road corridor. EFFEKT is used by
The Norwegian Public Roads Administration for cost-bene-
fit analysis (including life cycle GHG emissions and energy
use) of construction projects (Sandvik & Hammervold,
2011; Straume & Bertelsen, 2015). Joulesave was produced
in the research project Integration of Energy Into Road
Design for calculating energy use of traffic and
infrastructure in alternative road corridors (ECRPD, 2010).
The LICCER model was developed in the research project
Life Cycle Considerations in Environmental Impact
Assessment of Road Infrastructure (LICCER) (2012-2013)
to quantify GHG emissions and energy use of alternative
road corridors (Potting, Birgisd
ottir, Brattebø, Kluts et al.,
2013). Klimatkalkyl was developed by the Swedish
Transport Administration (STA) in 2013 and is now used
to quantify climate impact and primary energy use of
Swedish road and railway infrastructure throughout the
whole planning process: from route selection to follow-up
(Toller, 2018).
A limited number of studies have been found that reports
the use of LCA in early planning of transport infrastructure.
The majority of those studies concern the application of the
aforementioned models in case studies. Sandvik and
Hammervold (2011) used EFFEKT to show how GHG emis-
sions and energy use vary for alternative road corridors and
how this affects the benefits of a construction project. In the
research project ECRPD, Joulesave was used to investigate
possible energy savings that could be achieved by applying
the model in choice of road corridor (ECRPD, 2010). The
LICCER model was applied in Sweden to test model robust-
ness and to develop the scope and outline of the model
(Liljenstr€
om, 2013; Liljenstr€
om et al., 2013). Later, the model
was also used in Norway to show how a simplified LCA can
be conducted in early planning of a road construction pro-
ject and how results from the LCA can help road planners
in deciding between construction alternatives (O’Born
et al., 2016).
The models EFFEKT, LICCER, and Klimatkalkyl have
also been compared to each other with the purpose of iden-
tifying similarities and differences between them
(Chrysovalantis Lemperos & Potting, 2015; Ebrahimi
et al., 2016).
Additional work has been made by Oduro and Lautala
(2017) who assessed GHG emissions of three alternative
routes for transporting nickel and copper ore from mine to
refinery, aiming to compare a simplified and a detailed
LCA approach.
This paper builds on research conducted within the
research project LICCER. The aim is to explore the demands
that practitioners have on LCA-based models used in early
planning (concerning availability of input data to the model
and the usefulness of model outputs for decision-making)
and how well the LICCER model corresponds to those
demands. This is done in order to further evaluate the
LICCER model and to provide recommendations for devel-
opment and improvement of LCA based models in
early planning.
2. The LICCER model
The LICCER model (Potting, Birgisd
ottir, Brattebø, Kluts
et al., 2013) is an Excel based model that quantifies life cycle
GHG emissions and energy use of road infrastructure ele-
ments (bridges, tunnels, and plain roads) and traffic in alter-
native road corridors in a planned construction project. The
2 C. LILJENSTRÖM ET AL.
model can compare life cycle impacts of up to three road
corridors with each other and a reference alternative repre-
senting the situation if the project is not undertaken.
Although the LICCER model was specifically developed for
use in Sweden, Norway, Denmark, and the Netherlands, it
could be used in any country by supplementing the data
included in the model.
This section briefly describes the LICCER model. For
additional details, the reader is referred to the LICCER
model guideline report (Lundberg et al., 2013) and the
LICCER model technical report (Brattebø et al., 2013).
2.1. Life cycle assessment methodology
The calculations in the LICCER model are based on LCA
methodology. LCA methodology consists of four phases
(International Organization for Standardization, 2006):
1. goal and scope definition, including choice of functional
unit and system boundaries (section 2.2)
2. data inventory, involving data collection and processing
for all life cycle stages (section 2.3)
3. impact assessment, involving translation of inventory
data to environmental impact contribution (section 2.4)
4. interpretation of results from the data inventory and
impact assessment (section 2.5).
2.2. Functional unit and system boundaries
The functional unit in the LICCER model is “Road infra-
structure enabling annual traffic between ‘A’and ‘B’over an
analysis time horizon of a defined number of years”
(Brattebø et al., 2013).
The LICCER model quantifies annual GHG emissions
and energy use of material production (including extraction
of raw materials), construction, operation, and end-of-life
for road infrastructure elements and of traffic on the road
during operation (Figure 1). Each life cycle stage is
described below:
Production: includes raw material extraction and proc-
essing, material production, manufacturing of construc-
tion components, and transportation of raw materials to
production and manufacturing.
Construction: includes transportation from production
and manufacturing to the construction site, fuel use in
construction machinery, earthworks, transport of exca-
vated masses within the project, and electricity use on
site during construction.
Operation: includes production of materials for resurfac-
ing and maintenance of the road, transportation from
material production to the construction site, road light-
ing, and ventilation of tunnels. Carbon sequestration of
concrete and lime during road operation is not
accounted for.
End-of-life: includes energy use for road deconstruction
and material removal, transportation to depot, and earth-
works to restore the land area. No GHG emissions from
landfilling of construction materials are accounted for.
Environmental burdens and benefits of material recycling
are allocated to the system that uses the
recycled materials.
Traffic: includes operation of traffic on the road
(accounting for traffic volume and different types of
vehicles and fuels).
The model includes default data representative of con-
struction conditions in Sweden and Norway. Annual GHG
emissions and energy use is calculated over an analysis
period set by the user, for example 20 years, based on aver-
age traffic conditions during that analysis period and by
dividing life cycle GHG emissions and energy use of a road
element with a fixed service life (Brattebø et al., 2013).
2.3. Data inventory
The LICCER model automatically quantifies material, fuel,
and electricity consumption in all life cycle stages based on
user input and a set of default data included in the model.
User input (Table A1 in Appendix) covers project specific
data for infrastructure (for example length and width of
road elements) and traffic (for example traffic volume and
projected annual traffic increase) in the road corridors
(Brattebø et al., 2013). Default data (Table A2 in Appendix)
includes for example transport distance of construction
materials and fuel consumption for traffic.
2.4. Impact assessment
The environmental impact categories included in the
LICCER model are cumulative energy demand (CED) and
climate impact. The aim was to complement environmental
assessments (Brattebø et al., 2013) that include local envir-
onmental impacts but, at the time the LICCER model was
developed, often excluded life cycle GHG emissions and
energy use (Brattebø et al., 2013; Finnveden & Åkerman,
2014; Miliutenko et al., 2014).
Figure 1. System boundaries of the LICCER model (adapted from Brattebø
et al., 2013).
INTERNATIONAL JOURNAL OF SUSTAINABLE TRANSPORTATION 3
Based on the quantities of material, fuel, and electricity
(see section 2.3), the LICCER model automatically calculates
GHG emissions and energy use in each life cycle stage based
on default values (Table A2 in Appendix) for GHG emis-
sions and energy use of resource input. The datasets for spe-
cific GHG emissions of materials and energy use per unit of
resource input (Table A2 in Appendix) include already char-
acterized data, expressed in kilogram CO
2
equivalents (for
climate change) and MJ (for cumulative energy demand).
Climate impact was measured as Global Warming Potential
(GWP100) (Goedkoop et al., 2013). CED represents direct
and indirect energy use (Jungbluth & Frischknecht, 2010).
In this paper, the indirect energy use includes the feed-
stock energy.
Since the LICCER model was designed for use at national
road authorities, no license is needed to access any of the
defaultdata.Henceecoinvent(Wernetetal.,2016), a common
LCA database, was not used as a source of emission data.
2.5. Interpretation
Results from the LICCER model are intended to help in the
choice of road corridor. The LICCER model presents annual
impacts in terms of GHG emissions (kg CO
2
equivalents/
year) and energy use (MJ/year) for each road corridor
(including the reference alternative) in several ways
(Lundberg et al., 2013):
impacts of traffic and infrastructure in each
road corridor,
impacts of traffic and infrastructure in each road corri-
dor relative to the reference alternative,
impacts of each road element and life cycle stage, and
impacts of different material and energy inputs.
The model can help in decision-making by showing
which road corridor has the lowest life cycle GHG emissions
and energy use. By showing environmental hotspots (life
cycle stages and resource inputs that contribute most to life
cycle impacts) of each road corridor the model can also help
planners identifying measures to reduce environmen-
tal impacts.
3. Methods
3.1. Case study using the LICCER model
The LICCER model
1
was used to assess life cycle GHG
emissions and energy use, environmental hotspots (life cycle
stages, construction activities, and materials that contribute
most to GHG emissions and energy use), and critical
parameters (project specific and default parameters that
have the largest influence on the outcome of the study) of
alternative road corridors in a construction project. Results
from the case study are used to discuss data availability in
relation to a specific construction project and to suggest
when to use project specific data in early planning.
The case study used was the reconstruction of road 55
between Yxtatorpet and Malmk€
oping in south-eastern
Sweden. This case study had also been used previously to
test previous versions of the LICCER model (Liljenstr€
om,
2013; Liljenstr€
om et al., 2013). There were several reasons
for choosing this case. It provided good data availability
compared to other construction projects in the early plan-
ning stage and also involved several construction measures
(plain road, extended road, and bridge), which allowed the
testing of several features of the LICCER model. The project
had also been analyzed previously by Shamoon (2012) using
the model Joulesave, which allowed comparison of case
study results to a previous study.
The road section was reconstructed between April 2012
and September 2014. This case study was not conducted as
a part of the actual planning process and had no influence
on the choice of road corridor.
The feasibility study (Englund & Dahlin, 2006) conducted
in planning of the construction project compared three road
corridors (Alternative 1-3) (Figure 2) with a reference alter-
native (Alternative 0) that represented the existing situation.
The construction alternatives compared in the case study are
presented below:
Alternative 0: The reference alternative includes only
measures needed for maintaining ordinary function of the
road (Englund & Dahlin, 2006) and so it includes no new
construction or major reinvestment projects. The road sec-
tion is about 7.5 km long.
Alternative 1: The existing road, about 7.5 km long, is
widened from 9 to 14 meters and the road profile
is adjusted.
Alternative 2: The beginning and end of the existing
road is widened from 9 to 14 meters. A new road section,
about 2.6 km long, and a new bridge are constructed. The
total road length between Yxtatorpet and Malmk€
oping in
this alternative is about 6.9 km.
Alternative 3: The beginning and end of the existing
road is widened from 9 to 14 meters. A new road section,
about 3.0 km, and a new bridge are constructed. The total
road length between Yxtatorpet and Malmk€
oping in this
alternative is about 6.6 km. This alternative was the one
selected for construction.
The functional unit was, according to the LICCER model
(see section 2.2) “Road infrastructure enabling annual traffic
between Yxtatorpet and Malmk€
oping over an analysis time
horizon of 20 years”. The time horizon 20 years was chosen
since it is the common dimensioning period for road infra-
structure in Sweden.
Different approaches were used to find project specific
input data for infrastructure (Table A3 in Appendix):
Data were compiled directly from the feasibility study
(Englund & Dahlin, 2006).
Estimations were based on qualitative descriptions and
scenarios in the feasibility study (Englund &
Dahlin, 2006).
1
Modelling was made using the LICCER model version 1.0 from December
2013; however, the model was updated to correct for some calculation errors.
4 C. LILJENSTRÖM ET AL.
Estimations were based on previous road construction
projects (Karlsson & Carlson, 2010) with similar traffic
density and road width.
Project specific input data on traffic (Table A4 in
Appendix) at the start of the analysis period were provided
by the Swedish Road Administration (via Shamoon, 2012),
reflecting the conditions that were the basis for road design.
Future fuel use was estimated from a previous scenario ana-
lysis (Hansson & Grahn, 2013) representing the target that
the Swedish Transport Administration had in 2012 for the
Swedish transport system in 2030.
Available data were modified to fit the scope of the
LICCER model. Material, fuel, and electricity use in each
road corridor (Table A5 in Appendix) was quantified based
on the project specific data and default data.
A sensitivity analysis was conducted to identify critical
parameters, i.e. the parameters where a change in input data
and default data has the biggest influence on the results. In
the sensitivity analysis, parameters that influenced resulting
GHG emissions and energy use of hotspots were, one at a
time, increased by 10%. The sensitivity analysis included
only changes to infrastructure parameters since these
differed significantly between road corridors. While changes
to traffic related parameters influence the quantitative
results, the same traffic scenario is used in all road corridors
(because the roads are dimensioned for a specific traffic
scenario) so changes to traffic related parameters do not
influence ranking of road corridors. A sensitivity analysis for
traffic in the case study was conducted previously using an
earlier version of the LICCER model (Liljenstr€
om, 2013;
Liljenstr€
om et al., 2013).
3.2. Stakeholder involvement in model development
During development of the LICCER model, a workshop was
held aiming to discuss relevance and applicability of the
model in the decision-making process with potential users of
the LICCER model: national road administrations, research-
ers, and consultants working with environmental assessment
of road infrastructure in Sweden, Norway, Denmark, and the
Netherlands. The workshop was held in Stockholm, Sweden
in September 2013. Of about 85 persons invited, 12 attended
the workshop, most of them working at Swedish consultan-
cies or with research at Swedish universities and research
institutes (Table 1). The majority of the participants were
familiar with LCA in the context of road infrastructure plan-
ning; however, none of them was directly involved in deci-
sions on road corridor at a road administration.
The case presented in section 3.1 was used as a basis for
discussions at the workshop. An interactive exercise was
held to guide the participants through each step needed to
conduct an LCA with the LICCER model (Potting et al.,
2013). Each step of the exercise was followed by a
Table 1. Participants at the workshop (not including developers of the
LICCER model).
Sweden Norway Denmark
University 2
Research institute 2
Consultancy 3
Construction company 1
Road authority 2 1 1
Figure 2. The map shows the location of the three road corridors (Alternative 1, Alternative 2, and Alternative 3) compared in the case study (Englund & Dahlin,
2006). #Lantm€
ateriet.
INTERNATIONAL JOURNAL OF SUSTAINABLE TRANSPORTATION 5
questionnaire on data availability, usefulness of model out-
puts, and ease of use. The questionnaire is available in the
workshop report by Potting et al. (2013). At the end of the
workshop, participants and developers of the LICCER model
discussed the questions in a plenary discussion. Notes taken
by the LICCER team during the workshop and answers
from the questionnaires are summarized in the work-
shop report.
After the workshop, the LICCER model was further
developed based on workshop participants’comments on
ease of using the model. This paper only discusses partici-
pants’answers on accessibility of input data and usefulness
of results for choice of road corridor.
4. Results
4.1. Stakeholder views
4.1.1. Data availability
A problem expressed generally at the workshop was that
data needed to use the LICCER model are rarely available in
early planning. However, workshop participants believed it
would be possible to make estimates of this data.
Participants appreciated the default data in the LICCER
model since they considered it particularly difficult to quan-
tify those parameters, for instance transport distance, spe-
cific material consumption, and specific GHG emissions. A
general request was the access to default data that are nation
specific and approved by the national road authority.
Participants also suggested that a model containing default
data on material and energy use for construction of roads,
tunnels, and bridges could simplify use of LCA in choice of
road corridor. By providing the opportunity to insert project
specific data instead of default data, the model provides
enough flexibility.
About traffic related parameters, participants expressed
that many of these are available at national road authorities
or from other sources, but they also noted the high uncer-
tainty in these parameters. For example, while road author-
ities measure the traffic density on the infrastructure stock,
the data are often relatively old. Prognoses and scenario
analyses for future traffic increase and proportion of fuels
depend on factors such as market intervention by authorities
and assumed development of transport needs. If road traffic
should be allowed to increase in the future is a politically
dependent question. Since results depend a lot on traffic
related parameters, workshop participants suggested that
model users should test the outcome by changing these
input parameters in a sensitivity analysis.
4.1.2. Usefulness of results
The workshop participants thought the LICCER model pro-
vides results that are useful in choice of road corridor par-
ticularly because the model includes both traffic and
infrastructure, identifies possible improvements, and
presents results relative to a reference alternative. However,
it was pointed out that including the same traffic scenario in
all road corridors may not be realistic.
To understand the influence of uncertain data on differ-
ences between road corridors, participants suggested that the
user should include an uncertainty analysis. At the same
time, including an uncertainty analysis increases the model’s
complexity. Participants also pointed out that, for use in
decision-making, national road authorities must integrate
the model in the planning process. Therefore, the model
must complement tools already used for other purposes.
Some participants expressed that they would benefit from
a model that can also guide in choice of transport mode.
Currently, the LICCER model cannot be used for
that purpose.
4.2. Case study
4.2.1. Life cycle energy use and GHG emissions
Traffic operation accounted for 94-99% of the life cycle
GHG emissions (about 3 330-3 770 tonnes CO
2
equivalents
per year) and 90-97% of the life cycle energy use (about 58
780-66 590 GJ per year) in the three road corridors and the
reference alternative (Figure 3). Infrastructure accounted for
40-200 tonnes CO
2
equivalents and 2 350- 6 420 GJ per
year, depending on construction alternative. The significant
influence of traffic depends on the traffic scenario used, for
example the proportion of electric vehicles on the road.
With a larger proportion of electric vehicles, the GHG emis-
sions of traffic would be lower. In the LICCER model, GHG
emissions and energy use of traffic operation are directly
related to the road length and they were therefore highest in
Alternative 0 and Alternative 1 that are longer than the
other road corridors.
Figure 4 shows annual GHG emissions and energy use of
infrastructure and traffic in the three road corridors relative to
the reference alternative, for example DAlt.3 ¼Alt.3-Alt.0.
Alternative 2 and 3, which have shorter driving distance for
vehicles than Alternative 0, showed traffic related emission sav-
ings (290-440 tonnes CO
2
equivalents per year) as well as
energy savings (5 060-7 810 GJ per year) compared to the refer-
ence alternative. However, all new road corridors (Alternative 1-
3) had higher infrastructure related GHG emissions (80-160
tonnes CO
2
equivalents per year) and energy use (4 030-4 070
GJ per year) than the reference alternative because they require
construction work (widening the road and building new road
sections) and more asphalt for resurfacing (since they
are wider).
Hence, because Alternative 3 is the shortest construction
alternative, it had lowest overall GHG emissions and energy
use (Figure 3) and the largest savings of GHG emissions
and energy compared to the reference alternative (Figure 4).
However, Alternative 3 had higher infrastructure related
GHG emissions than the other alternatives and about the
same infrastructure related energy use (Figure 5). Since
Alternative 3 have more difficult construction conditions
and therefore require larger quantities of earthworks and
soil stabilization than the other road corridors, GHG emis-
sions and energy use of the production and construction
stages were highest in Alternative 3. However, since
Alternative 3 is the shortest road corridor, less material is
6 C. LILJENSTRÖM ET AL.
required for resurfacing and there is a smaller road area to
restore at end-of-life; hence GHG emissions and energy use
of operation and end-of-life were lower for Alternative 3
than for the other alternatives.
Energy use was dominated by the operation stage in all
new road corridors as well as the reference alternative
(Figure 5). The operation stage had higher relative import-
ance to energy use than to GHG emissions because feed-
stock energy (i.e. the chemical energy stored in the bitumen)
was included for bitumen used in resurfacing.
Figures 6 and 7show the contribution of different construc-
tion parts and processes to infrastructure related GHG emis-
sions and energy use. Production based GHG emissions are
influenced by the construction conditions in each road corridor.
In Alternative 1, which involves only widening of the existing
road, materials for the pavement, base layer, and sub-base layer
(i.e. bitumen and aggregates) had the highest contribution to
production related GHG emissions. However, in Alternative 2
and 3 that have more difficult construction conditions, a larger
proportion of GHG emissions are due to soil stabilization (i.e.
cement and lime) and explosives. In Alternative 3, which
involves the largest volumes of rock excavation, explosives con-
tributed more to production based GHG emissions than the
pavement layer. The bridge in Alternative 2 and 3 contributed
to about 1% of the production related GHG emissions.
Production based energy use was however dominated by
the pavement layer, base layer, and sub-base layers in all
construction alternatives, due to the feedstock energy
included for bitumen.
GHG emissions and energy use of the construction stage
were mainly due to earthworks in Alternative 2 and 3.
Material transport had a more significant contribution to
emissions and energy use in Alternative 1, which involves
less earthworks. Common for all alternatives, GHG emis-
sions and energy use of infrastructure operation was mainly
due to the pavement layer required for resurfacing. Also in
Figure 4. Annual GHG emissions (tonne CO2 equivalents) and energy use (GJ) of infrastructure and traffic in the three road corridors (Alt. 1-Alt.3) and the reference
alternative (Alt.0) relative to the reference alternative (DAlt.X ¼Alt.X-Alt.0).
Figure 3. Annual GHG emissions (tonne CO2 equivalents) and energy use (GJ) of infrastructure and traffic in the three road corridors (Alt. 1-Alt.3) and the reference
alternative (Alt.0).
INTERNATIONAL JOURNAL OF SUSTAINABLE TRANSPORTATION 7
all alternatives, GHG emissions and energy use of end-of-
life was mainly due to diesel use for earthworks required to
restore the road area.
While infrastructure related impacts are influenced by the
road corridor characteristics, traffic related impacts depend
on a traffic scenario that is the same in all road corridors.
Hence, the proportion of impacts due to different types of
vehicles and fuels is the same in all new road corridors as
well as in the reference alternative (Figure 8). Light vehicles
accounted for 63% of the traffic related GHG emissions and
65% of the traffic related energy use. Both for light vehicles
and lorries, fossil fuels (diesel and petrol) accounted for the
largest share of emissions and energy use. In total, biofuels
and electric vehicles accounted for 7% of the GHG emis-
sions and 23% of the energy use. This difference is due to
electric vehicles and biofuels having relatively low GHG
emissions per MJ.
4.2.2. Critical parameters
In all road corridors, including the reference alternative, the
resulting infrastructure GHG emissions and energy use were
sensitive
2
to changes in parameters related to the quantity of
asphalt needed during the road lifetime (assumption of
resurfacing period) (Figures 9 and 10). When the resurfacing
period was decreased by 10%, the resulting GHG emissions
increased by 2.5-8.4%, and the resulting energy use
increased by 5.7-10.5%, depending on construction alterna-
tive. Assumption of resurfacing period was especially
important for Alternative 0 that have no impacts from pro-
duction or construction and for Alternative 1 where little
construction work was made.
Figure 5. Annual GHG emissions (tonne CO2 equivalents) and energy use (GJ) of infrastructure life cycle stages in the three road corridors (Alt.1-Alt.3) and the refer-
ence alternative (Alt.0).
Figure 6. Contribution of construction parts and processes to the annual GHG emissions of each life cycle stage (production, construction, operation, and end-of-
life) for the different road corridors (Alt.1-Alt.3) and the reference alternative (Alt.0).
2
Here defined as an increase by at least 2% when the parameter value was
increased or decreased by 10%.
8 C. LILJENSTRÖM ET AL.
Resulting GHG emissions and energy use were also sensitive
to changes in parameters that affect the impacts of earthworks
(specific GHG emissions of diesel) and pavement (energy use
per unit of bitumen input), respectively (Figures 9 and 10).
When the value of these parameters were increased by 10%,
the resulting GHG emissions increased by 1.9-2.7% and the
resulting energy use increased by 6.5-8.0%, depending on con-
struction alternative. Assumption of specific GHG emissions of
diesel was most important for the construction alternatives
using the largest quantities of diesel for earthworks in the con-
struction stage, as well as for Alternative 0 that had a relatively
high proportion of impacts from the end-of-life stage where
diesel is used for earthworks.
Particularly for Alternative 3, which required larger vol-
umes of rock excavation and soil stabilization than the other
alternatives, resulting GHG emissions were also sensitive to
changes in the volume of excavated rock as well as volume
of soil stabilization, and quantity of lime and cement needed
for soil stabilization (Figure 9).
Because infrastructure had a small contribution to life
cycle GHG emissions and energy use compared to traffic,
increasing the value of infrastructure related parameters did
not have a significant influence on the overall results and
did not change the ranking of road corridors.
5. Discussion
Using LCA for decision support in choice of road corridor
provides an opportunity to influence significantly the life
cycle impacts of traffic and infrastructure. However, because
of limited access to project specific data, environmental per-
formance must be evaluated based on a limited amount of
input data. LCA models for use in choice of road corridor
Figure 7. Contribution of construction parts and processes to the annual energy use of each life cycle stage (production, construction, operation, and end-of-life)
for the different road corridors (Alt.1-Alt.3) and the reference alternative (Alt.0).
Figure 8. Annual GHG emissions (tonne CO2 equivalents) and energy use (GJ) of different types of vehicles and fuels in the three road corridors (Alt.1-Alt.3) and
the reference alternative (Alt.0).
INTERNATIONAL JOURNAL OF SUSTAINABLE TRANSPORTATION 9
are available and have been applied in case studies; however,
it has not been investigated whether they fulfill LCA practi-
tioners’requirements on such models. Identifying such
requirements can give recommendations for development of
LCA based models and aid in evaluation of existing models
for suitability in early planning. Based on a workshop with
LCA practitioners and a case study, this paper explores how
the LCA model LICCER corresponds to practitioners’
demands concerning availability of input data to the model
and the usefulness of model outputs for decision-making.
5.1. Availability of project specific and default
input data
It was clear from the workshop that the project specific data
required to use the LICCER model is limited in early
planning and that default data is required to complete an
LCA. Naturally, the variation in such default data is large
(because, by necessity is has been collected from different
other sources). For example, the assumed default diesel use
for soil excavation is several times higher in Klimatkalkyl
3.0 than in the LICCER model and the assumed default
quantity of explosives for rock blasting is twice as high
(Karlsson et al., 2017). In terms of traffic, default data (for
example fuel consumption) may vary between different
roads depending on local conditions such as road incline
and speed limit. Thus, the default data included in the
model may not be representative of the actual construc-
tion project.
Uncertainty in default data may affect the outcome of the
LCA; however, in case default data is common for all road
corridors, they do not necessarily influence the ranking
between road corridors. In the case study of this paper for
example, some default data in the LICCER model, particu-
larly specific emission and energy use of diesel and bitumen,
had high importance for results. Since these default data
were common for all road corridors (and there was not a
very significant difference in quantity), they did not influ-
ence ranking of construction alternatives. However, in other
construction projects, uncertainty in default data may vary
between road corridors and affect the resulting ranking. For
example, in another LICCER case study (O’Born et al.,
2016) one construction alternative included a tunnel, which
follows straightforward construction guidelines, whereas the
other construction alternative included a bridge, which has
larger variations in construction design. Thus, the uncer-
tainty in concrete related emissions was lower in the con-
struction alternative that included a tunnel.
By increasing the number of default parameters in the
LICCER model, data collection could be further simplified.
At the workshop, participants suggested that models like
Klimatkalkyl, containing default data for different construc-
tion types in Sweden, may be easier to use by reducing the
complexity of input data. When the LICCER model was
developed, such default data were not yet available so this
approach was not implemented in the LICCER model
(Potting et al., 2013). However, the model may now be
extended by such data.
At the same time, default data on construction types
could possibly make the model too simple to differentiate
the road corridors, thereby making the model unsuitable for
decision-making. Especially for volumes of excavated rock
and soil, type of soil stabilization method, and quantity of
soil stabilization needed, default data should be used with
caution. Because requirements on earthwork and soil stabil-
ization vary largely between projects, site-specific data are
preferred to get reliable results for decision-support in early
planning. LCA based models could be combined with meth-
ods to provide site-specific volumes of rock and soil excava-
tion. By using available tools such as geographic information
systems, mass haul optimization models, and software for
road alignment, usability of LCA models for early planning
support may be improved without losing simplicity.
Figure 9. Sensitivity analysis for life cycle GHG emissions of infrastructure.
Figure 10. Sensitivity analysis for life cycle energy use of infrastructure.
10 C. LILJENSTRÖM ET AL.
To some extent, characteristics of the specific construc-
tion project determine what parameters should be project
specific. Including the possibility to replace default data with
project specific data is therefore required for simple but flex-
ible models, for improved usability of the model, and for
reliability of results for decision-making. According to the
workshop participants, the possibility to replace the default
data was also one of the benefits of the LICCER model.
As noted above, project specific data is required for the
parameters that differentiate the road corridors, in other
words the parameters that do not have the same value in
each road corridor. In addition, parameters that can be
influenced in early planning should be project specific so
that possibilities for improvements can be identified. Some
characteristics, for example land use, are important to con-
sider in early planning LCA of road infrastructure because
more precise calculations in later planning stages may not
help to reduce environmental impacts of the project design
(Butt et al., 2015). Other aspects, for example impacts of
specific construction materials, are influenced mainly by
decisions taken in the design stage and these are therefore
of smaller relevance in the choice of road corridor.
Some parameters may, in early planning, be influenced
mainly by changing the road length. In the case study, this
was the case for asphalt quantities (more asphalt is required
for a longer road) and traffic (more emissions from traffic
the longer the driving distance). Thus, the shortest road cor-
ridor will likely have the lowest impact of these parameters.
Project specific data is therefore especially important for the
parameters that are not specifically related to the road
length. However, it should be noted that for traffic, this con-
clusion depends on the level of detail in the traffic scenarios.
A more detailed assessment of traffic related impacts than
what is included in the LICCER model may be warranted
(as discussed in section 5.2) and in that case, the traffic
related impacts may not be directly related to the
road length.
It should be noted that project specific data is not neces-
sarily less uncertain than default data. In the case study for
instance, the volumes of excavated rock and soil in each
road corridor were estimated in the feasibility study and
were therefore considered parameters with good data avail-
ability. Even so, the uncertainty in these parameters is high.
Karlsson et al. (2017) found that the actual volume of exca-
vated soil during construction of Alternative 3 in this case
study was 16% lower than estimated in the feasibility study
and that the actual volume of excavated rock was 62% lower
than estimated in the feasibility study. For traffic, many of
the project specific parameters, for example annual traffic
increase and fuel types used at the end of the analysis
period, are based on assumptions on future traffic develop-
ment and technologies.
The project specific parameters may be environmental
hotspots of the system, but not necessarily. For example, in
the case study, earthworks and soil stabilization had low
emissions and energy use compared to traffic on the road
and did not influence the ranking of road corridors.
However, the environmental impacts of earthworks and soil
stabilization are aspects that planners can influence by
choice of road corridor. Additionally, earthworks and soil
stabilization are naturally site specific (they differentiate the
road corridors) and the volumes of excavated materials and
stabilized soil do not necessarily decrease by choosing a
shorter road corridor. In the case study, the shortest road
corridor involved more rock excavation and soil stabilization
than the other alternatives. Considering the general variabil-
ity in earthworks and soil stabilization, it is recommended
that default data is not used for earthwork volumes, soil sta-
bilization method, and volume of stabilized soil in early
planning LCA of road infrastructure.
5.2. Usefulness of model output for decision-making
A requirement for use of LCA models in decision-making is
that they provide reliable results. Because road construction
projects are unique, results from road LCAs cannot be dir-
ectly compared (Barandica et al., 2013). Literature studies
have however indicated general patterns regarding important
life cycle stages and construction activities. These include
the dominance of traffic related impacts (Hill et al., 2012),
the relative importance of material production and construc-
tion compared to maintenance and end-of-life (Barandica
et al., 2013; Gulotta et al., 2018), and the importance of
earthworks in case of difficult construction conditions
(Barandica et al., 2013). Generally, results from the case
study comply well with such findings. The ranking of road
corridors is also the same as in the case study using
Joulesave (Shamoon, 2012), although the LICCER model
resulted in significantly higher energy use (Liljenstr€
om,
2013). Other studies have indicated that the LICCER model
results in lower GHG emissions and energy use than
Klimatkalkyl version 2.0 and 3.0 (Chrysovalantis Lemperos
& Potting, 2015) and EFFEKT (Ebrahimi et al. 2016).
The workshop participants thought the LICCER model
was useful for decision-making because it compares impacts
of traffic and infrastructure in the road corridors. However,
the model can be improved in this respect. The importance
of traffic related impacts (as seen in the case study) indicates
a need to include more detailed traffic scenarios that allow
identifying improvement measures related to traffic.
One way to improve the model is to include additional
details on road geometry, such as speed limit and incline.
The model could also be improved by including the possibil-
ity to consider different traffic scenarios (for example traffic
volume and annual traffic increase) depending on the road
corridor. Accounting for different traffic scenarios becomes
important in case the reference alternative cannot sustain
the traffic levels predicted on the new road. This was likely
the event in the case study; hence, potentially, no emission
savings were made compared to the reference alternative.
Another reason to include different traffic scenarios is to
analyze potential tradeoffs of constructing shorter roads. For
example, if the reference alternative is replaced by a shorter
road, traffic related emissions may decrease (due to shorter
driving distance); however, emissions may likewise increase
due to new and shorter roads inducing more traffic (see for
INTERNATIONAL JOURNAL OF SUSTAINABLE TRANSPORTATION 11
example the overview by Streimikiene et al., 2013). If the
shortest road corridor also requires construction of tunnels
and bridges, total life cycle impacts may not decrease com-
pared to the reference situation.
Including more detailed traffic scenarios in LCA models
for early planning would probably not increase the complex-
ity of the model (considering the traffic data available at
road authorities); however, the results would be more rele-
vant for decision-making (considering that other benefits
highlighted at the workshop was the opportunity to compare
with the reference alternative and identify possible improve-
ment measures).
Due to the presence of assumptions in infrastructure LCA,
decisions are inevitably made under significant uncertainty.
Understanding and communicating this uncertainty is necessary
for the reliability of results and for the acceptance of LCA as
part of the decision-making process (Igos et al., 2019;Zhao
et al., 2017). For comparative studies, the practitioner must
understand whether he/she should question the preference for
onealternativeduetouncertainties(Igosetal.,2019).
At the workshop, it was recommended that scenarios and
uncertainty analysis should be included when using the
LICCER model, but it was also recognized that this might
increase the complexity of the model and thereby conflict
with the goal of keeping the model simple and easy to use.
Thus, practical ways to handle uncertainty treatment in the
LICCER model as well as in other LCA models are required.
Recently, Igos et al. (2019) have provided recommendations
for uncertainty treatment including approaches applicable
when little data is available. Additionally, Larsson Ivanov
et al. (2019) discuss methods to evaluate uncertainty specif-
ically in LCA of infrastructure, using the model
Klimatkalkyl as an example. It needs to be further evaluated
whether these recommendations are suitable for integration
in the LICCER model as well as for other models used in
early planning, depending on the decision-making context
and who will use the model.
The scope of the LICCER model is limited to climate
impact and primary energy use, thus conclusions cannot be
drawn on what road corridor is best from a life cycle per-
spective. Further research is required to verify whether con-
clusions from this paper are valid also for other impact
categories, regarding both data availability and requirements
on model output.
In addition, the road corridor with lowest life cycle envir-
onmental impacts will not necessarily perform best also
from other perspectives, such as technical, local environ-
ment, social, etc. At the workshop, participants pointed out
that an LCA model must complement other forms of deci-
sion support used in planning. The LICCER model was
developed as a stand-alone tool that can be used in parallel
to the environmental impact assessment process (Potting,
Birgisd
ottir, Brattebø, Kluts et al., 2013). LCA can also be
integrated in the environmental assessment (Miliutenko
et al., 2014)–an option that could be suitable in other geo-
graphical contexts.
Multi-criteria decision-making has been suggested as a
method to integrate several different environmental impacts
(del Mar Casanovas-Rubio & Ramos, 2017), social sustain-
ability of transport infrastructure (Sierra et al., 2018), and in
general for sustainable decision-making in construction,
when a number of criteria must be evaluated (Zavadskas
et al., 2017). Other requirements may be placed on LCA
when it is integrated with other forms of decision support;
these have not been considered in this paper.
5.3. Limitations of the study
Results from this paper reflect the views of the workshop
participants: mainly Swedish researchers and consultants
familiar with LCA of road infrastructure, however not dir-
ectly involved in decisions on choice of road corridor. The
perception of data availability, the acceptance and know-
ledge of uncertainty, and the importance of model output
likely depend on the user of the model. Therefore, it needs
to be further confirmed whether findings in this paper also
reflect requirements from the road administrations in the
participating countries as well as potential users of the
LICCER model from Norway, Denmark, and the
Netherlands.
Results also reflect the views of the workshop participants
in 2013. Since then, an increased focus has been placed on
the climate impact of transport infrastructure. For example,
since 2016, the STA places climate requirements in procure-
ment of construction and maintenance of road and rail
infrastructure and in procurement of construction materials
(Toller & Larsson, 2017). Additionally, the Swedish con-
struction sector has adapted the target to reduce their GHG
emissions, from a life cycle perspective, with 50% to 2030
and to be climate neutral by 2045 (Fossilfritt Sverige, 2018).
Such initiatives could potentially lead to different require-
ments on model outputs to help achieve these targets.
At the same time, several of the requirements put for-
ward at the workshop have been recognized in other con-
texts. For example, Meex et al. (2018) analyzed user
requirements on LCA based tools for use in early building
design. Among them are: a limited amount of input data
consistent with the specific decision-making stage; use of
default values for missing data (unknown material types and
quantities) that should, when possible, be based on national
averages for representative results; include both construction
and operation (in that case operational energy use); simple
output adapted to the specific decision-making stage. While
another decision context than described in this paper, this
may point to some general requirements for usefulness of
models in early planning stages with little specific data and
several construction options.
6. Conclusion
Including LCA in early planning of road construction may
give planners decision support to reduce life cycle impacts
of a road at a planning stage where there are large opportu-
nities to reduce the life cycle impacts of the road. This paper
has explored demands that practitioners have on LCA mod-
els used in early planning (concerning availability of input
12 C. LILJENSTRÖM ET AL.
data and the usefulness of model outputs for decision-mak-
ing), and how the LCA model LICCER fulfills these
demands. Findings from this paper could be useful in the
development and improvement of LCA models for use in
early planning of road construction (in choice of road corri-
dor) and in the evaluation of suitability of existing models
for use in early planning.
Based on results from this paper, it is recommended that
an LCA model for use in early planning fulfills the following
requirements:
It includes default data that is nation specific and prefer-
ably approved by the national road authority. If possible,
the model should include national default data for con-
struction measures such as different types of roads,
bridges, and tunnels.
It provides the ability to replace default data by project
specific data.
It includes traffic as well as infrastructure for alternative
road corridors and is able to differentiate the road corri-
dors both in terms of traffic and in terms of
infrastructure.
It identifies possible improvements both for traffic and
for infrastructure.
It presents results relative to a reference alternative.
It includes opportunities to assess the uncertainty in
the results.
It provides results that complement other decision-mak-
ing tools.
Planners are recommended to find project specific data
for parameters that can be influenced in early planning, that
differentiate the alternative road corridors, and that are not
specifically related to the road length. These parameters may
be hotspots of the system, but not necessarily. In particular,
it is recommended that default data should not be used for
site-specific parameters such as volumes of earthworks, soil
stabilization method, or volume of soil stabilization.
The LICCER model fulfills several of the requirements
above, but can be improved by including further nation spe-
cific default data for different construction measures (now
made available in Sweden through the model Klimatkalkyl),
different traffic scenarios depending on the road corridor,
more detailed traffic scenarios, and an assessment of uncer-
tainty in the model output.
Further research is needed to understand how to best
include uncertainty assessment and how model requirements
change when LCA is combined with other tools in the deci-
sion-making process. Additional research is also required to
confirm whether findings are generally applicable also to
other groups of LCA practitioners and to other case studies
than included in this paper.
Acknowledgements
This study was performed as part of the project “Life Cycle
Considerations in Environmental Impact Assessment of Road
Infrastructure (LICCER)”. We would like to thank the ERA NET Road
Programme for financing the project. Additional support was provided
through the VINNMER programme from Vinnova –Swedish Agency
for Innovation Systems.
Funding
European Research Area (ERA-NET Road Programme), Vinnova -
Swedish Agency for Innovation Systems (the VINNMER programme).
References
Barandica, J. M., Fern
andez-S
anchez, G., Berzosa,
A., Delgado, J. A., &
Acosta, F. J. (2013). Applying life cycle thinking to reduce green-
house gas emissions from road projects. Journal of Cleaner
Production,57,79–91. https://doi.org/10.1016/j.jclepro.2013.05.036
Bhander, G. S., Hauschild, M., & McAloone, T. (2003). Implementing
life cycle assessment in product development. Environmental
Progress,22(4), 255–267. https://doi.org/10.1002/ep.670220414
Blanco, G., Gerlagh, R., Suh, S., Barrett, J., de Coninck, H. C., Diaz
Morejon, C. F., Mathur, R., Nakicenovic, N., Ahenkora, O.A., Pan,
J., Pathak, H., Rice, J., Richels, R., Smith, S.J., Stern, D.I., Toth, F.L.,
& Zhou, P. (2014). Drivers, trends and mitigation. In: O. Edenhofer,
R. Pichs-Madruga, Y. Sokona, E. Farahani, S. Kadner, K. Seyboth, A.
Adler, I. Baum, S. Brunner, P. Eickemeier, B. Kriemann, J.
Savolainen, S. Schl€
omer, C. von Stechow, T. Zwickel & J.C. Minx
(Eds.), Climate Change 2014: Mitigation of Climate Change.
Contribution of Working Group III to the Fifth Assessment
Report of the Intergovernmental Panel on Climate Change (pp.
351–412). International Panel of Climate Change.
Bouhaya, L., Le Roy, R., & Feraille-Fresnet, A. (2009). Simplified envir-
onmental study on innovative bridge structure. Environmental
Science & Technology,43(6), 2066–2071. https://doi.org/10.1021/
es801351g
Brattebø, H., O’Born, R., Miliutenko, S., Birgisd
ottir, H., Lundberg, K.,
Toller, S., Potting, J. (2013). LICCER Model Technical Report.
Account of technical backgrounds of the LICCER model (Report No.
4.2 [final report] December 2013). Retrieved from https://www.cedr.
eu/strategic-plan-tasks/research/era-net-road/call-2011-energy/liccer-
project-results/.
Butt, A. A., Toller, S., & Birgisson, B. (2015). Life cycle assessment for
the green procurement of roads: a way forward. Journal of Cleaner
Production,90, 163–170. https://doi.org/10.1016/j.jclepro.2014.11.068
Carlson, A., & Folkeson, L. (2014). Sustainability and energy efficient
management of roads. Retrieved from https://www.cedr.eu/strategic-
plan-tasks/research/era-net-road/call-2011-energy/.
Cenosco & Royal HaskoningDHV. (n.d.). DuboCalc. Retrieved from
https://www.dubocalc.nl/en/.
Chrysovalantis Lemperos, X., & Potting, J. (2015). Comparison of
Klimatkalkyl, LICCER & SimaPro –Three models to quantify life
cycle energy and carbon dioxide in early road infrastructure plan-
ning. KTH Royal Institute of Technology. Department of sustainable
development, environmental science and engineering.
del Mar Casanovas-Rubio, M., & Ramos, G. (2017). Decsion-making
tool for the assessment and selection of construction processes based
on environmental criteria: Application to precast and cast-in-situ
alternatives. Resources, Conservation & Recycling,126, 107–117.
https://doi.org/10.1016/j.resconrec.2017.07.035
Delft University of Technology. (2012). Idemat2012 [Dataset].
Retrieved from http://www.ecocostsvalue.com/EVR/model/theory/
subject/5-data.html.
Du, G., Safi, M., Pettersson, L., & Karoumi, R. (2014). Life cycle assess-
ment as a decision support tool for bridge procurement: environ-
mental impact comparison among five bridge designs. The
International Journal of Life Cycle Assessment,19(12), 1948–1964.
https://doi.org/10.1007/s11367-014-0797-z
Ebrahimi, B., Wallbaum, H., Brattebø, H., Vingnisdottir, H. R., Bohne,
R. A., & Booto, G. K. (2016). Environmental Life Cycle Assessment
(LCA) of road pavements: Comparing the quality and point of
INTERNATIONAL JOURNAL OF SUSTAINABLE TRANSPORTATION 13
application of existing software tools on the basis of a Norwegian
Case Study. In N. Achour (Ed.), Proceedings of the CIB World
Building Congress 2016: Volume V –Advancing products and services
(pp. 749–760). Department of Civil Engineering.
Energy conservation in road pavement design [ECRPD]. (2010). Energy
conservation in road pavement design, maintenance and utilisation.
Retrieved from https://ec.europa.eu/energy/intelligent/projects/en/
projects/ecrpd.
Englund, J., & Dahlin, A. (2006). V€
ag 55, delen Yxtatorpet-
Malmk€
oping. V€
agutredning. Samrådshandling 2006-01-25 [Road 55,
section Yxtatorpet-Malmk€
oping. Feasability study. Consultation
report 2006-01-25]. Swedish Road Administration.
Fern
andez-S
anchez, G., Berzosa,
A., Barandica, J. M., Cornejo, E., &
Serrano, J. M. (2015). Opportunities for GHG emissions reduction
in road projects: a comparative evaluation of emissions scenarios
using CO
2
NSTRUCT. Journal of Cleaner Production,104, 156–167.
https://doi.org/10.1016/j.jclepro.2015.05.032
Finnveden, G., & Åkerman, J. (2014). Not planning a sustainable trans-
port system. Environmental Impact Assessment Review,46,53–57.
https://doi.org/10.1016/j.eiar.2014.02.002
Fossilfritt Sverige. (2018). F€
ardplan f€
or fossilfri konkurrenskraft. Bygg-
och anl€
aggningssektorn [Roadmap for fossil free competitiveness.
The building and construction sector]. Retrieved from http://fossil-
fritt-sverige.se/wp-content/uploads/2018/01/ffs_bygg_anlggningssek-
torn181017.pdf.
Fritsche, U. (2005). Process-oriented Basic Data for Environmental
Management Instruments - www.probas.umweltbundesamt.de. The
International Journal of Life Cycle Assessment,10(3), 225–225.
https://doi.org/10.1065/lca2005.03.004
Garbarino, E., Rodriguez Qunitero, R., Donatello, S., & Wolf, O.
(2016). Revision of green public procurement criteria for road design,
construction and maintenance. Procurement practice guidance docu-
ment. (EUR 28028 EN). Joint Research Centre.
Gode, J., Martinsson, F., Hagberg, L., €
Oman, A., H€
oglund, J., Palm, D.,
Ekvall, T., Hallberg, L., H€
ogberg, J., Ljunggren S€
oderman, M.,
Jerksj€
o, M. & Rydberg, T. (2011). Milj€
ofaktaboken 2011: Uppskattade
emissionsfaktorer f€
or br€
anslen, el, v€
arme och transporter.
[Milj€
ofaktaboken 2011: Estimated emission factors for fuels, electri-
city, heat, and transport in Sweden]. V€
armeforsk Service AB.
Goedkoop, M., Heijungs, R., Huijbregts, M., De Schryver, A., Struijs, J.,
& van Zelm, R. (2013). ReCiPe 2008 –A life cycle impact assessment
method which comprises harmonised category indicators at the mid-
point and the endpoint level. First edition (version 1.08). Report I:
Characterisation. Ministry of Housing, Spatial Planning and
Environment.
Gulotta, T. M., Mistretta, M., & Pratic
o, F. G. (2018). Life cycle assess-
ment of roads: material and process related energy savings.
Modelling, Measurement and Control C,79(3), 146–153. https://doi.
org/10.18280/mmc_c.790313
Guo, C., Xu, J., Yang, L., Guo, X., Liao, J., Zheng, X., Zhang, Z., Chen,
X., Yang, K., & Wang, M. (2019). Life cycle evaluation of green-
house gas emissions of a highway tunnel: A case study in China.
Journal of Cleaner Production,211, 972–980. https://doi.org/10.1016/
j.jclepro.2018.11.249
Hammervold, J. (2014). Towards greener road infrastructure: life cycle
assessment of case studies and recommendations for impact reductions
and planning of road infrastructure [Doctoral disertation,
Norweigian University of Science and Technology, Norway].
Retrieved from https://brage.bibsys.no/xmlui/handle/11250/274029.
Hammond, G., & Jones, C. (2011). ICE Inventory of Carbon and
Energy version 2.0. [Dataset]. Retrieved from http://www.circulare-
cology.com.
Hansson, J., & Grahn, H. (2013). Utsikt f€
or f€
ornybara drivmedel i
Sverige. Uppdatering och utvidgning av studien M€
ojligheter f€
or
f€
ornybara drivmedel i Sverige till år 2030 av Grahn och Hansson,
2010. [Outlook for renewable fuels in Sweden. Update and expan-
sion of the study Possibilites for renewable fuels in Sweden by 2030
by Grahn and Hansson, 2010]. (IVL Report B2083). IVL Swedish
Research Institute.
Highways England. (2016). Carbon emissions calculation tool: Highways
England. Retrieved from https://www.gov.uk/government/publica-
tions/carbon-tool.
Hill, N., Brannigan, C., Wynn, D., Milnes, R., van Essen, H., den Boer,
E., van Grinsven, A., Lighthart, T. & van Gijlswijk, R. (2012). EU
Transport GHG: Routes to 2050 II. Final report Appendix 2: The role
of GHG emissions from infrastructure construction, vehicle manufac-
turing, and ELVs in overall transport sector emissions. (Task 2 paper
produced as part of a contract between European Commission
Directorate-General Climate Action and AEA Technology plc; (see
website www.eutransportghg2050.eu).
Huang, L., Bohne, R. A., Bruland, A., Drevland Jakobsen, P., & Lohne,
J. (2015). Life cycle assessment of Norweigian Road Tunnel. The
International Journal of Life Cycle Assessment,20(2), 174–184.
https://doi.org/10.1007/s11367-014-0823-1
Huang, L., Krigsvoll, G., Johansen, F., Liu, Y., & Zhang, X. (2018).
Carbon emission of global construction sector. Renewable and
Sustainable Energy Reviews,81, 1906–1916. https://doi.org/10.1016/j.
rser.2017.06.001
Huang, Y., Hakim, B., & Zammataro, S. (2013). Measuring the carbon
footprint of road construction using CHANGER. International
Journal of Pavement Engineering,14(6), 590–600. https://doi.org/10.
1080/10298436.2012.693180
Igos, E., Benetto, E., Meyer, R., Baustert, R., & Othoniel, B. (2019).
How to treat uncertainies in life cycle assessment studies? The
International Journal of Life Cycle Assessment,24(4), 794–807.
https://doi.org/10.1007/s11367-018-1477-1
International Energy Agency. (2018). CO2 Emissions from Fuel
Combustion 2018 Highlights [Excel file]. Retreived from https://web-
store.iea.org/statistics-data.
International Organization for Standardization. (2006). Environmental
management –Life cycle assessment –Principles and framework.
(ISO standard 14040: 2006). International Organization for
Standardization.
Jullien, A., Dauvergne, M., & Proust, C. (2015). Road LCA: the dedi-
cated ECORCE tool and database. The International Journal of Life
Cycle Assessment,20(5), 655–670. https://doi.org/10.1007/s11367-
015-0858-y
Jungbluth, N. & Frischknecht, R. (2010). Cumulative energy demand.
In: R. Hischier & B. Weidema (Eds.), Implementation of life cycle
impact assessment methods. Data v2.2 (2010). ecoinvent report No.
3, v2.2. Swiss Centre for Life Cycle Inventories.
Karlsson, C., Miliutenko, S., Bj€
orklund, A., M€
ortberg, U., Olofsson, B.,
& Toller, S. (2017). Life cycle assessment in road infrastructure plan-
ning using spatial geological data. The International Journal of Life
Cycle Assessment,22(8), 1302–1317. https://doi.org/10.1007/s11367-
016-1241-3
Karlsson, R., & Carlson, A. (2010). Ber€
akningar av energiåtgång och
koldioxidutsl€
app vid byggande, drift och underhåll av v€
agar.
Ber€
akningar enligt fyra typfall [Calculations of energy use and carbon
dioxide emission in construction, operation, and maintenance of
roads. Calculations according to four typical cases]. (VTI Notat 3-
2010). Swedish National Road and Transport Research Institute.
Kluts, I., & Miliutenko, S. (2012). Overview of road infrastructure plan-
ning process and the use of Environmental Assessments in the
Netherlands and Sweden. KTH Royal Institute of Technology,
Deparment of Urban Planning and Environment, Division of
Environmental Strategies Research.
Larsson Ivanov, O., Honfi, D., Santandrea, F., & Stripple, H. (2019).
Consideration of uncertainties in LCA for infrastructure using prob-
abilistic methods. Structure and Infrastructure Engineering,15(6),
711–724. https://doi.org/10.1080/15732479.2019.1572200
Liljenstr€
om, C. (2013). Life cycle assessment in early planning of road
infrastructure - Application of The LICCER-model [Master’s thesis].
KTH Royal Institute of Technology. Retrieved from http://www.
diva-portal.org/smash/search.jsf?dswid=5059.
Liljenstr€
om, C., Miliutenko, S., Brattebø, H., O’Born, R., Birgisd
ottir,
H., Lundberg, K., Toller, S. & Potting, J. (2013). LICCER Model
Case Study Report: Application of the LICCER-model to a Swedish
road section between Yxtatorpet and Malmk€
oping. (Report Nr 5.1
14 C. LILJENSTRÖM ET AL.
November 2013). Retrieved from https://www.cedr.eu/strategic-plan-
tasks/research/era-net-road/call-2011-energy/liccer-project-results/.
Liu, Y., Wang, Y., Li, D., & Yu, Q. (2019). Life cycle assessment for
carbon dioxide emissions from freeway construction in mountainous
area: Primary source, cut-off determination of system boundary.
Resources, Conservation and Recycling,140,36–44. https://doi.org/10.
1016/j.resconrec.2018.09.009
Lundberg, K., Miliutenko, S., Birgisd
ottir, H., Toller, S., Brattebø, H.,
Potting, J. (2013). LICCER Model Guidelines Report. (Report Nr 4.1).
Retrieved from https://www.cedr.eu/strategic-plan-tasks/research/era-
net-road/call-2011-energy/liccer-project-results/.
Meex, E., Hollberg, A., Knapen, E., Hildebrand, L., & Verbeeck, G.
(2018). Requirements for applying LCA-based environmental impact
assessment tools in the early stages of building design. Building and
Environment,133, 228–236. https://doi.org/10.1016/j.buildenv.2018.
02.016
Miliutenko, S., Åkerman, J., & Bj€
orklund, A. (2012). Energy use and
greenhouse gas emissions during the life cycle stages of a road tun-
nel –the Swedish case Norra L€
anken. European Jounral of Transport
and Infrastructure Research,12,39–62. Retrieved from https://www.
tudelft.nl/tbm/over-de-faculteit/afdelingen/engineering-systems-and-
services/research/ejtir/.
Miliutenko, S., Kluts, I., Lundberg, K., Toller, S., Brattebø, H.,
Birgisd
ottir, H., & Potting, J. (2014). Consideration of life cycle
energy use and greenhouse gas emissions in road infrastructure
planning processes: Examples of Sweden, Norway, Denmark and the
Netherlands. Journal of Environmental Assessment Policy and
Management,16(04), 1450038. https://doi.org/10.1142/
S1464333214500380
O’Born, R. (2018). Life cycle assessment of large scale timber bridges: A
case study from the world’s longest timber bridge design in Norway.
Transportation Research Part D: Transport and Environment,59,
301–312. https://doi.org/10.1016/j.trd.2018.01.018
O’Born, R., Brattebø, H., Kålas Iversen, O. M., Miliutenko, S., &
Potting, J. (2016). Quantifying energy demand and greenhouse gas
emissions of road infrastructure projects: An LCA case study of the
Oslo fjord crossing in Norway. European Journal of Transport and
Infrastructure Research,16, 445–466. Retrieved from https://www.
tudelft.nl/tbm/over-de-faculteit/afdelingen/engineering-systems-and-
services/research/ejtir/.
Oduro, S. K., & Lautala, P. (2017). Incorporating life cycle assessment
(LCA) in freight transportation infrastructure project evaluation
[Paper presentation]. 2017 Joint Rail Conference, April). In
Proceedings of the JRC2017, April 4-7, 2017, Philadelphia, PA, USA.
https://doi.org/10.1115/JRC2017-2303
Olofsson, T. (2010). Delrapport: Bro. [Progress report: Bridge].
Retrieved from http://www.bygginnovationen.se/.
Potting, J., Birgisd
ottir, H., Brattebø, H., Kluts, I., Liljenstr€
om, C.,
Lundberg, K., Miliutenko, S., O’Born, R., Iversen, O.M., Toller, S. &
van Oirschot, R. (2013). LICCER Final Report. (Report Nr 6
December 2013). Retrieved from https://www.cedr.eu/strategic-plan-
tasks/research/era-net-road/call-2011-energy/liccer-project-results/.
Potting, J., Birgisd
ottir, H., Brattebø, H., Liljenstr€
om,C.,Lundberg,K.,
Miliutenko, S., O’Born, R. & Toller, S. (2013). Report from second work-
shop. (Report Nr 3 October 2013). Retrieved from https://www.cedr.eu/
strategic-plan-tasks/research/era-net-road/call-2011-energy/liccer-pro-
ject-results/.
Rydberg, T., & Andersson, R. (2003). Milj€
oeffektbed€
omning (LCA) f€
or
markstabilisering [Life cycle assessment (LCA) of ground stabilisa-
tion]. (Report). Swedish Deep Stabilisation Research Centre.
Salkangas, L. (2009). ETSI Project (Stage 2) –Bridge Life Cycle
Optimisation. Helsinki University of Technology.
Sandvik, K. O., Hammervold, J. (2011). Incorporating greenhouse gas
emissions in Benefit-Cost Analysis in the transport sector in Norway.
Paper presented at the 24th World Road Congress, Mexico.
Santos, J. M., Thyagarajan, S., Keijzer, E., Fernandez Flores, R., &
Flintsch, G. (2017). Pavement life cycle assessment –a comparison
of American and European tools. In I. L. Al-Quadi, H. Ozer, & J.
Harvey (Eds.), Pavement Life-Cycle Assessment. Proceedings of the
Symposium on Life-Cycle Assessment of Pavements (Pavement LCA
2017) April 12-13, 2017, Champaign, Illinois, USA (pp. 281–287).
Taylor & Francis Group. doi https://doi.org/10.1201/9781315159324
Shamoon, H. (2012). Energieffektiv v€
agdesign [Energy efficient road
design]. [Master’s thesis]. KTH Royal Institute of Technology,
Sweden. Retrieved from http://www.diva-portal.org/smash/search.
jsf?dswid=5059.
Sierra, L. A., Yepes, V., & Pellicer, E. (2018). A review of multi-criteria
assessment of the social sustainability of infrastructures. Journal of
Cleaner Production,187, 496–513. https://doi.org/10.1016/j.jclepro.
2018.03.022
Simonsen, M. (2010). Transport, energi og miljø [Transport, energy,
and environment]. (Report nr 2/2010). Western Norway Research
Institute.
Spriensma, R., van Gurp, C., Anker Larsen, M. R. (2014). CEREAL
Final report. (Deliverable Nr 6 & 7 January 2014). Retrieved from
https://www.cedr.eu/wpfb-file/04_final-report-cereal-2014-pdf/.
Straume, A., & Bertelsen, D. (2015). Brukerveiledning GS-EFFEKT 6.6.
[User manual GS-EFFEKT 6.6] (Statens vegvesens rapporter Nr.
357). Norweigian Public Roads Administration.
Streimikiene, D., Bale
zentis, T., & Bale
zentien_
e, L. (2013). Comparative
assessment of road transport technologies. Renewable and
Sustainable Energy Reviews,20, 611–618. https://doi.org/10.1016/j.
rser.2012.12.021
Stripple, H. (2001). Life cycle assessment of road.A pilot study for
inventory analysis. (IVL Report B 1210 E). IVL Swedish
Environmental Research Institute.
Stripple, H. (2009). Kompletterande underlag f€
or tillåtlighetspr€
ovning –
en €
oversiktlig milj€
ostudie av v€
aginfrastrukturen i projekt F€
orbifart
Stockholm [Complementing basis for consideration of permissibility –
a general environmental study of the road infrastructure in the pro-
ject Bypass Stockholm. ] (U2380). IVL Swedish Environmental
Research Institute.
Swedish Transport Administration. (2013). Klimatkalkyl (version 1.0)
[Excel-file].
Toller, S. (2018). Klimatkalkyl –Calculating energy use and greenhouse
gas emissions of transport infrastructure from a life cycle perspective
(TRV 2018:115). The Swedish Transport Administration.
Toller, S., & Larsson, M. (2017). Implementation of life cycle thinking
in planning and procurement at the Swedish Transport
Administration. In I. L. Al-Quadi, H. Ozer, & J. Harvey (Eds.),
Pavement Life-Cycle Assessment. Proceedings of the Symposium on
Life-Cycle Assessment of Pavements (Pavement LCA 2017) April 12-
13, 2017, Champaign, Illinois, USA., (pp. 281–287). Taylor & Francis
Group. doi https://doi.org/10.1201/9781315159324
Trunzo, G., Moretti, L., & D’Andrea, A. (2019). Life cycle analysis of
road construction and use. Sustainability,11(2), 377. https://doi.org/
10.3390/su11020377
Wayman, M., Sciavi Mellor, I., Cordell, B., James, D., Gossling, R.,
Loveday, C., Simms, M. & Southwell, C. (2012). The asphalt pave-
ment embodied carbon tool (asPECT): Developing a carbon footprint-
ing methodology for asphalt products. Paper presented at the 5th
Eurasphalt & Eurobitume Congress, 13-15th June 2012, Istanbul.
Wernet, G., Bauer, C., Steubing, B., Reinhard, J., Moreno-Ruiz, E., &
Weidema, B. (2016). The ecoinvent database version 3 (part I): over-
view and methodology. The International Journal of Life Cycle
Assessment,21(9), 1218–1230. https://doi.org/10.1007/s11367-016-
1087-8
Zapata, P., & Gambatese, J. A. (2005). Energy consumption of asphalt
and reinforced concrete pavement materials and construction.
Journal of Infrastructure Systems,11(1), 9–20. https://doi.org/10.
1061/(ASCE)1076-0342(2005)11:1(9)
Zavadskas, E. K., Antucheviciene, J., Vilutiene, T., & Adeli, H. (2017).
Sustainable decision-making in civil engineering, construction and
building technology. Sustainability,10(2), 14. https://doi.org/10.
3390/su10010014
Zhao, K., Ng, T. S., Kua, H. W., & Tang, M. (2017). Modeling environ-
mental impacts and risk under data uncertainties. IISE Transactions,
49(12), 1150–1159. https://doi.org/10.1080/24725854.2017.1342054
INTERNATIONAL JOURNAL OF SUSTAINABLE TRANSPORTATION 15
Appendix
Table A1. User input (project specific data) required to use the LICCER model.
Parameter Unit
INFRASTRUCTURE (For each road corridor and road element)
Length m
Share length with road lighting %
Share length with side guardrails as well as centre guardrails %
Type of side and centre guardrails (steel or concrete) –
Volume of excavated soil m
3
Volume of excavated rock m
3
Volume of stabilised soil m
3
Type of soil stabilisation method (in-situ, concrete mass, LC columns or soil replacement) –
Tunnel walls and lining methods (shotcrete, PE foam, concrete, concrete EI.) –
Number of lanes (driving lanes, hard and soft shoulders, central reserve, cycling/pedestrian lanes, road ditch) Nr
Width of each lane m
Height of sub-base, base, and pavement layers m
Type of material in sub-base and base layers (aggregate or sand) –
TRAFFIC (The same traffic scenario for all road corridors)
Traffic volume (annual average daily traffic [AADT]) Nr of vehicles
Annual traffic increase %
Analysis time horizon Years
Share of biofuel in end year %
Share of electric cars in end year %
Table A2. Default data of the LICCER model (LICCER_LCA-model_v1 0) used in the case study.
Parameter Unit Value Source
SERVICE LIFE OF ROAD INFRASTRUCTURE
Superstructure components in roads Years 60 Karlsson and Carlson (2010); Stripple (2009)
Superstructure components in bridges Years 60 Karlsson and Carlson (2010); Stripple (2009)
Resurfacing (of pavement layer; calculated from AADT) Years 10 Calculated according to input variables
TRANSPORT DISTANCE OF MATERIALS (LORRY ON ROAD ONLY)
Materials from outside suppliers
Aggregate/gravel, all usage except pavement asphalt km 20 Stripple (2001)
Asphalt membrane km 150 Personal communication with the Norwegian Public Roads
Administration
Asphalt, pavement (incl. bitumen and aggregate) km 30 Stripple (2001)
Sand/soil, all usage km 20 Stripple (2001)
Concrete, bridges km 150 Miliutenko et al. (2012)
Cement, soil stabilisation km 300 Stripple (2001)
Lime for lime pillars, soil stabilisation km 300 Stripple (2001)
Explosives km 240 Miliutenko et al. (2012)
Rebar, bridges km 500 Miliutenko et al. (2012)
Steel, guardrails km 500 Miliutenko et al. (2012)
Internal transportation of masses
Internal transportation of masses from earthwork km 0.5 Stripple (2001)
Transport, pavement materials to depot at end-of-life km 2.0 Stripple (2001)
Transport, base & subbase materials to depot at end-of-life km 2.0 Stripple (2001)
Transport, concrete materials to depot at end-of-life km 2.0 Stripple (2001)
Transport, rebar materials to depot at end-of-life km 2.0 Stripple (2001)
Transport, steel materials to depot at end-of-life km 2.0 Stripple (2001)
FUEL CONSUMPTION FROM TRAFFIC IN USE STAGE
Diesel fuel, traffic use stage, lorries, no trailer litre/10km 2.32 Personal communication with the Swedish Transport
Administration
Diesel fuel, traffic use stage, lorries, with trailer litre/10 km 4.09 Personal communication with the Swedish Transport
Administration
Diesel fuel, traffic use stage, light vehicles litre/10 km 0.67 Personal communication with the Swedish Transport
Administration
Petrol fuel, traffic use stage, light vehicles litre/10 km 0.87 Personal communication with the Swedish Transport
Administration
Electricity, traffic use stage, light vehicles MJ/km 1.607 Simonsen (2010)
Proportion of lorry traffic, no trailer % of AADT 6.6 Shamoon (2012)
Proportion of lorry traffic, with trailer % of AADT 5.4 Shamoon (2012)
Proportion of light vehicle traffic % of AADT 88 Shamoon (2012)
Proportion light vehicles on diesel fuel, traffic use stage % of light vehicles 17 Shamoon (2012)
Proportion of biofuel in diesel/petrol fuel today % of total fuel use 7.0 Shamoon (2012)
Proportion of electric cars in use today, light vehicles % of light vehicle stock 0.5 Shamoon (2012)
(continued)
16 C. LILJENSTRÖM ET AL.
Table A2. Continued.
Parameter Unit Value Source
SPECIFIC MATERIAL CONSUMPTION
Asphalt, depth of layer replaced in each reasphaltation m thickness replaced 0.04 Swedish Transport Administration (2013)
Asphalt membrane, bridge surface kg/m
2
surface area 26 EFFEKT (Sandvik & Hammervold, 2011)
Concrete, concrete bridges m
3
/m
2
surface area 0.78 Calculated average from Olofsson (2010)
Cement, lime-cement columns in soil stabilisation tonne/ m
3
stabilised soil 0.05 Recalculated from Rydberg and Andersson (2003)
Lime, lime-cement columns in soil stabilisation tonne/m
3
stabilised soil 0.05 Recalculated from Rydberg and Andersson (2003
Diesel, transportation of masses in earthwork litre/tkm mass transport 0.035 EFFEKT (Sandvik & Hammervold, 2011)
Diesel, machinery for earthwork in road construction litre/m
3
loose materials 0.80 EFFEKT (Sandvik & Hammervold, 2011)
Diesel, earthworks, blasted rock litre/m
3
loose materials 0.80 Assumed
Diesel, earthworks, simple soil excavation litre/m
3
loose materials 0.09 EFFEKT (Sandvik & Hammervold, 2011)
Diesel, end-of-life pavement removal litre/m
3
loose materials 0.80 EFFEKT (Sandvik & Hammervold, 2011)
Diesel, end-of-life base & subbase removal litre/m
3
loose materials 0.80 EFFEKT (Sandvik & Hammervold, 2011)
Diesel, end-of-life concrete structures demolition litre/m
3
concrete structure 5.0 Assumed
Diesel, end-of-life earthworks litre/m
2
total road area 2.0 EFFEKT (Sandvik & Hammervold, 2011)
Explosives, road construction kg/m
3
rock in situ 1.0 EFFEKT (Sandvik & Hammervold, 2011)
Rebar, concrete bridges tonne/m
2
surface area 0.13 Calculated average from Olofsson (2010)
Steel, guardrails tonne/m guardrail 0.025 Swedish Transport Administration (2013)
SPECIFIC GREENHOUSE GAS EMISSIONS OF MATERIALS
Aggregate kg CO
2
e/ tonne 5.2 Hammond and Jones (2011)
Bitumen kg CO
2
e/ tonne 430 Delft University of Technology (2012)
Asphalt membrane kg CO
2
e/ tonne 310.89 Salkangas (2009) (considered as 72.3 % of impact
from bitumen)
Asphalt mixing kg CO
2
e/ tonne 14.04 Zapata and Gambatese (2005) (Swedish electricity mix)
Concrete kg CO
2
e/m
3
256.8 Hammond and Jones (2011); Swedish Transport
Administration (2013)
Diesel kg CO
2
e/m
3
3 093 Delft University of Technology (2012)
Biofuel kg CO
2
e/m
3
636 Recalculated from Gode et al. (2011) for ethanol
from wheat
Electricity kg CO
2
e/ kWh 0.036 Gode et al. (2011)
Explosives kg CO
2
e/ tonne 2 310 Probas Database (Fritsche, 2005)
Petrol kg CO
2
e/m
3
2 925 Recalculated from Delft University of Technology (2012) for
Petrol including combustion (assuming the density of
719.7 kg/m
3
)
Rebar (reinforcement steel) kg CO
2
e/ tonne 720 Swedish Transport Administration (2013)
Steel kg CO
2
e/ tonne 1 500 Swedish Transport Administration (2013)
Lime, soil stabilisation kg CO
2
e/ tonne 780 Hammond and Jones (2011)
Cement kg CO
2
e/ tonne 715 Swedish Transport Administration (2013)
Transport work kg CO
2
e/ tkm 0.15 Hammond and Jones (2011)
TOTAL ENERGY CONSUMPTION PER UNIT OF RESOURCE INPUT
Aggregate MJ/tonne 83 Hammond and Jones (2011)
Bitumen MJ/tonne 52 000 Delft University of Technology (2012)
Asphalt membrane MJ/tonne 37 596 Salkangas (2009) (considered as 72.3 % of impact
from bitumen)
Asphalt mixing MJ/tonne 390 Zapata and Gambatese (2005)
Concrete MJ/m
3
1 800 Swedish Transport Administration (2013)
Diesel MJ/m
3
45 575 Delft University of Technology (2012)
Biofuel MJ/m
3
31 376 Recalculated from Gode et al. (2011) for ethanol
from wheat
Electricity MJ/kWh 7.56 Gode et al. (2011)
Explosives MJ/tonne 26 312 Probas Database (Fritsche, 2005)
Petrol MJ/m
3
42 975 Recalculated from Delft University of Technology (2012)
Petrol including combustion (assuming the density of
719.7 kg/m3)
Rebar (reinforcement steel) MJ/tonne 13 100 Swedish Transport Administration (2013)
Steel MJ/tonne 20 100 Swedish Transport Administration (2013)
Cement MJ/tonne 4 135 Swedish Transport Administration (2013)
Lime in soil stabilisation MJ/tonne 5 300 Hammond and Jones (2011)
Transport work MJ/tkm 2.4 Hammond and Jones (2011)
INTERNATIONAL JOURNAL OF SUSTAINABLE TRANSPORTATION 17
Table A3. Project specific data for infrastructure in the different road corridors (Alt. 1-3) and the reference alternative (Alt. 0).
Input parameter Road element Alt. 0 Alt. 1 Alt. 2 Alt. 3 Reference
Road elements, length (m) Existing road 7 500 –––Englund and Dahlin (2006)
New road ––2 580 2 980
Extended road –7 500 4 330 3 620
Concrete bridge –– 20 20
Road elements, width (m) Existing road 9 –––Englund and Dahlin (2006)
New road –– 12 12
Extended road –888
Concrete bridge –– 12 12
Centre guardrails, steel (% of road length) Existing road 0 –––Englund and Dahlin (2006)
New road ––100 100
Extended road –100 100 100
Concrete bridge ––100 100
Side guardrails, steel (% of road length) Existing road ––––Estimated from Englund and Dahlin (2006)
New road –– 41 64
Extended road –16 12 11
Concrete bridge ––100 100
Soil excavation (m
3
) Total –184 000 430 000 207 000 Englund and Dahlin (2006)
Rock excavation (m
3
) Total –50 000 219 000 535 000 Englund and Dahlin (2006)
Soil stabilisation, lime-cement columns (m
3
) Total –3 925 20 822 38 055 Estimated from Englund and Dahlin (2006)
Pavement layer, 94% aggregate, 6% bitumen (m) Roads, bridges 0.08 0.08 0.08 0.08 Karlsson and Carlson (2010)
Base layer, 98% aggregate, 2% bitumen (m) Roads –0.15 0.15 0.15
Subbase layer, 100% aggregate (m) Roads –0.42 0.42 0.42
The total road width is 14 m after reconstruction; however, only 8 metres was modelled, since the centre of the old road could be retained.
Table A4. Project specific data for current and future traffic operation on the road (common for all road corridors and the reference alternative).
Input parameter Value References
Traffic density in start year (number of vehicles per day) 4 894 Swedish Road Administration as cited in Shamoon (2012)
Annual increase in traffic (%) 1.0 Swedish Road Administration as cited in Shamoon (2012)
Proportion of biofuel in end year (%) 43 Hansson and Grahn (2013)
Proportion of electric cars in end year (%) 12 Hansson and Grahn (2013)
Table A5. Resulting data inventory for the different life cycle stages of road corridors.
Material/Activity Unit Alt.0 Alt.1 Alt.2 Alt.3
PRODUCTION STAGE
Asphalt membrane tonne/year ––0.1 0.1
Aggregate/gravel (base layer, sub-base layer, and pavement layer) tonne/year –1 670 1 657 1 609
Bitumen (base layer and pavement layer) tonne/year –30 32 27
Concrete, concrete bridges tonne/year ––7.5 7.5
Cement, soil stabilisation tonne/year –3.0 16 26
Lime for lime pillars, soil stabilisation tonne/year –3.0 16 26
Explosives tonne/year –0.8 3.7 8.9
Steel, guardrails tonne/year –4.1 4.2 4.7
CONSTRUCTION STAGE
Transport of external materials from supplier to the construction site
Asphalt membrane tkm/year ––16 16
Aggregate/gravel (base layer, sub-base layer, pavement layer) tkm/year –36 777 36 108 34 984
Bitumen (base layer and pavement layer) tkm/year –897 946 797
Concrete, concrete bridges tkm/year ––1 119 1 119
Cement, soil stabilisation tkm/year –906 4 806 7 878
Lime from lime pillars, soil stabilisation tkm/year –906 4 806 7 878
Explosives tkm/year –200 876 2140
Steel, guardrails tkm/year –2 063 2101 2 340
Diesel used for handling and transport of internal masses in construction activities
Diesel, excavation & uploading rock m
3
/year –1.1 4.7 11.4
Diesel, excavation & uploading simple soil m
3
/year –0.3 0.7 0.3
Diesel, transport masses from earthwork m
3
/year –0.1 0.4 0.5
OPERATION STAGE
Aggregate/gravel, pavement resurfacing tonne/year 569 1 011 889 840
Bitumen, pavement resurfacing tonne/year 36 65 57 54
Transport, pavement resurfacing materials tkm/year 18 144 32 256 28 361 26 807
END-OF-LIFE STAGE
Diesel used for materials (pavement, base & subbase) removal, earthworks m
3
/year 2.9 5.9 6.3 5.0
Transport of materials (pavement, base & subbase, concrete, steel) to depot tkm/year 2 471 3 409 3 386 3 281
TRAFFIC ON ROAD DURING OPERATION STAGE
Electricity, transport by light vehicle MWh/year 365 365 337 322
Total diesel, traffic use stage m
3
/year 520 520 480 459
Total biofuel, traffic use stage m
3
/year 390 390 360 344
Total petrol, traffic use stage m
3
/year 650 650 600 573
18 C. LILJENSTRÖM ET AL.
