See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/389355980

Lifecycle Assessment and Sensitivity Analysis of Roller Compacted Concrete,

Jointed Plain Concrete, and Hot Mix Asphalt Pavements

Conference Paper · February 2025

CITATIONS READ

0 1

8 authors, including:

Haoran Li Hessam AzariJafari

Massachusetts Institute of Technology Massachusetts Institute of Technology
11 PUBLICATIONS 44 CITATIONS 58 PUBLICATIONS 1,101 CITATIONS

SEE PROFILE SEE PROFILE

Jeremy R Gregory Randolph Kirchain

Massachusetts Institute of Technology Massachusetts Institute of Technology
107 PUBLICATIONS 2,147 CITATIONS 247 PUBLICATIONS 6,332 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Haoran Li on 26 February 2025.

The user has requested enhancement of the downloaded file.

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

Lifecycle Assessment and Sensitivity Analysis of Roller

Compacted Concrete, Jointed Plain Concrete, and Hot Mix
Asphalt Pavements
Li, H. 1, Levin, J.R. 2, Azarijafari, H. 3, Gregory, J. 4, Kirchain, R. 5, Mack, J.W. 6,
Zollinger, C.J. 7, and Mahdi, M. 8

Abstract
Transport is the second-largest contributor to global greenhouse gas (GHG)
emissions. Therefore, it is crucial to establish a benchmark for the carbon intensity of
the road transportation sector. This assessment should encompass not only the impact
of paving materials and road construction practice, but also consider the entire
lifecycle of roads, including maintenance schedules and use phase. Roller-Compacted
Concrete Pavements (RCCPs) have gained popularity in low-traffic roads, expected to
offer a similar lifespan and maintenance requirement as jointed plain concrete
pavement (JPCP). However, they are constructed with different equipment, lower
cement content, and different mix proportions, and lack steel reinforcement. Due to
the lack of sufficient field data and analytical models, RCCPs’ long-term performance
and environmental impact are not well studied, as compared to JPCP and hot mix
asphalt (HMA) pavements. In this paper, we conducted life-cycle assessments (LCAs)
to evaluate the environmental impact of RCCPs, employing reasonable assumptions.
Sensitivity analyses were conducted to quantify the impact of various factors,
including joint faulting modeling, enhanced surface smoothness degradation models,
and diamond grinding, providing insights into the environmental considerations for
each pavement type. We used equivalent designs of JPCP and HMA conducted by
AASHTOWare Pavement ME Design as reference points for comparison. Our results
reveal that RCCP is a viable, low-global warming potential pavement option when the
surface receives appropriate treatment and maintenance. Specifically, RCCP with
diamond grinding as an initial surface treatment can significantly reduce total GHG
emissions, by up to 50%, in comparison to the HMA design alternative.

Introduction
Construction Sustainability. While critical to society, the built environment is a
major contributor to both carbon emissions and waste (Montgomery et al., 2015). The
transport sector is the second largest contributor to global greenhouse gas (GHG)

1
Haoran Li; PhD; Department of Civil and Environmental Engineering, Massachusetts Institute of
Technology; USA; [email protected]
2
Jessica R. Levin; PhD; Dow Inc.; USA; [email protected]
3
Hessam Azarijafari; PhD; Massachusetts Institute of Technology Concrete Sustainability Hub; USA;
[email protected]
4
Jeremy Gregory; PhD; Massachusetts Institute of Technology Climate and Sustainability Consortium;
USA; [email protected]
5
Randolph Kirchain; PhD; Massachusetts Institute of Technology Materials Research Laboratory;
USA; [email protected]
6
James W. Mack; PE, MBA, MS; Cemex; USA; [email protected]
7
Corey Zollinger; PE, MS; Cemex; USA; [email protected]
8
Moinul Mahdi; PhD; Louisiana Transportation Research Center; USA; [email protected]

13th International Conference on Concrete Pavements 1

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

emissions, and 90-95% of these emissions are attributed to roads. Of this, road
construction is responsible for merely 5-10%, while the bulk of emissions originate
from the use phase, including fuel consumption (World Bank, 2005). Compounding
the materials consumption problem is the limited recycling of road materials and
construction waste in general. According to the U.S. Environmental Protection
Agency (EPA), the United States generated 600 million tons of construction and
demolition waste in 2018 alone, which is more than two times the weight of
municipal waste produced in the same year (Townsend et al., 2019).

Actions to Reduce Emissions. Building strong and resilient mobility networks should
significantly reduce transportation GHG emissions, and correlate to more sustainable
roads. Strong and resilient roadways impact GHG emissions and waste in two ways.
First, excess fuel consumption is minimized by maintaining an excellent road surface
condition, such as International Roughness Index (IRI) and road stiffness
(Louhghalam et al., 2015; Robbins & Tran, 2015; Xu et al., 2019; Izevbekhai, 2022).
Second, durable roads require less maintenance and rehabilitation (M&R). This
reduces waste, materials consumption, and user delay. Lifecycle assessments (LCAs)
over the road’s entire lifetime should help identify strong and resilient road options
and reduce GHG emissions. Collection of environmental product declarations (EPDs)
and completion of LCAs of various road materials and construction practices will
allow road owners to benchmark current practices and identify options to improve in
the future.

Roller Compacted Concrete Pavement (RCCP). Given the increasing importance of

LCAs in selecting road designs globally, there is a need to identify sustainable, robust
roads that minimize carbon emissions and waste generation across the pavement’s
entire lifecycle. To address this issue, we have completed an LCA of a lesser-used
road design, roller compacted concrete pavement (RCCP). RCCP was developed for
heavy industry with similar durability to JPCP, but with a much faster return to
service (Holderbaum & Schweiger, 2007; Harrington et al., 2010; Ceylan, 2010). The
low user delay can be attributed to the mix design. The RCC mix is a zero slump, high
volume fraction aggregate, and low cement design that relies on aggregate interlock,
similar to asphalt, for early strength. The pavement still has joints but contains no
dowels, whose surface is more similar to HMA than to JPCP (Burwell et al., 2018).

Historically, because of its rough surface, RCCP was relegated to low-speed,

heavy-load areas such as ports, logging, and intermodal facilities (Piggott, 1999;
Zollinger, 2015). With improved technique and surface finishing, RCCP has been
growing in the pavement sector, particularly in low-volume roads or shoulders
(Zollinger, 2015). As a durable, low cement, low water design, RCCP could be a good
option for those investing in new, sustainable road surfaces. An LCA comparing an
RCCP design to HMA in Canada demonstrated that RCCP could have lower
embodied emissions (Lemieux & Thibodeau, 2017). However, this study excluded the
use phase when conducting an LCA (Lemieux & Thibodeau, 2017).

Given the relatively higher IRI (1.55-3.75 m/km depending on the base and
paving quality) of RCCP, it is important to assess the impact of surface treatment and
maintenance on the use phase to get a better understanding of the global warming
potential across the roads’ entire lifespans (Harrington et al., 2010; Zollinger, 2015;
Chhorn et al., 2017). However, as a less used, low-volume road pavement, there is

13th International Conference on Concrete Pavements 2

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

insufficient data in several key areas that are needed for a comprehensive LCA,
including lifetime data for high-quality RCCP roads, surface degradation models, and
IRI measurements over time. In this paper, we leveraged a recently developed
framework for evaluating the full LCA of RCCP and compared it to more common
pavement designs: jointed plain concrete pavements (JPCP) and hot mix asphalt
(HMA). The goal of this analysis is to compare the carbon intensity of the three
different pavements and understand the scenarios in which one pavement might be
lower in GHG emissions relative to the others.

Research Significance. This research holds significant value in pavement

sustainability by first establishing boundary scenarios for evaluating the potential
deterioration of RCCP in terms of joint faulting and IRI. We outline a methodology to
assess the long-term performance of RCCP using Pavement ME and the
corresponding GHG emission from the use phase, which was not possible before.
Additionally, we employ sensitivity analysis to assess the relative importance of
factors such as embodied carbon and the impact of the use phase in the LCA of
RCCP. Furthermore, the study offers valuable insights by recommending strategies to
mitigate GHG emissions over the entire life cycle of RCCP, encompassing enhanced
design and M&R strategies, and low-carbon concrete mix design approaches.

Methodology
Life-Cycle Assessment Framework of Pavement. As shown in Figure 1, a full
pavement LCA framework was employed for assessing the pavement life-cycle
environmental impacts, specifically focusing on GHG emissions as a measure of
Global Warming Potential (GWP). It starts with defining the pavement context. The
functional systems are used for describing the traffic of the proposed road, which
includes traffic loads, traffic volume, growth rate, traffic speed, etc. Pavement design
is a crucial process affecting the pavement’s long-term structural performance. An
overly conservative pavement design (e.g., too thick surface layer or overly
substantial base) usually leads to better long-term performance (Li & Khazanovich,
2022) but extensive initial cost in a life cycle cost analysis (LCCA) perspective (Umer
et al., 2017). Similarly, some trade-offs exist between embodied and use phase
emissions from an LCA perspective. We employed the AASHTO mechanistic-
empirical pavement design guide (MEPDG) (AASHTO, 2015) and the companion
AASHTOWare Pavement ME Design (Pavement ME) software (AASHTO, 2023) for
pavement design, to better understand the correlations between pavement design and
the LCA.

For pavement thickness design, the pavement geometry and material

properties are needed. For concrete pavement, parameters encompass portland cement
concrete (PCC) compressive strength, PCC coefficient of thermal expansion, PCC
elastic modulus, joint spacing, slab width, etc. For asphalt pavement, key factors
include asphalt concrete (AC) binder type, AC air voids, effective binder content, etc.
Our framework permits the examination of three pavement structures: JPCP, RCCP,
and asphalt (HMA) pavements. Since Pavement ME is not capable of designing
RCCP, we therefore assumed that RCCP has the same PCC thickness as JPCP for the
same pavement contexts. As a key component affecting long-term pavement
rideability, IRI is used to describe the surface condition, where lower numbers
indicate smoother surfaces.

13th International Conference on Concrete Pavements 3

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

Figure 1: Flowchart of pavement design and LCA framework and comparative LCA
analysis for alternative pavement structures: JPCP, RCCP, and HMA

The full LCA framework includes both the embodied and use-phase GHG
emissions. As shown in Figure 1, the embodied emissions encompass emissions from
the initial construction of the pavement, as well as those arising from M&R activities
and end-of-life (EOL) processes. Specifically, these emissions originate from the
production of materials (A1-A3) and the construction phase of the pavement (A4-A5).
To quantify the emissions attributable to material production, material emission
factors for both asphalt and concrete materials were calculated, using asphalt EPD
(NAPA EPDs, 2023) and the NRMCA industry-averaged emission data (NRMCA,
2019) respectively. Regarding the use phase, multiple submodules that account for
GHG emissions were applied during the pavement’s use phase, illustrated in Figure 1.
It includes GHG emissions attributed to excessive fuel consumption resulting from
the pavement vehicle interaction (PVI) mechanism, i.e., pavement roughness
(Louhghalam et al., 2017) and deflection (Louhghalam et al., 2014). The use phase
LCA also considers the global warming potential impacts arising from changes in
albedo for various surfaces, which the subsequent model was adopted from Xu et al.
(2020), leveraging the climatic data from NASA (NASA, 2022). To quantify the
GHG emission offset by the carbonation of concrete pavements, this work adopted the
models developed by AzariJafari et al. (2021). Recognizing the substantial influence
of initial construction and M&R activities, the transportation-related emission (A4-

13th International Conference on Concrete Pavements 4

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

A5) was computed using data obtained from the ecoinvent database and other
resources (Skolnik et al., 2013; US EPA, 2020).

Roller Compacted Concrete (RCC) Life Cycle Assessment (LCA). To complete an

LCA on RCCP, several assumptions needed to be made. Previous work by Wu,
Mahdi, and Rupnow showed that thin RCCPs had a much longer-than-expected
lifetime based on accelerated pavement loading (Wu et al., 2016). The anticipated
RCCP lifetime was modeled using the same assumptions as traditional JPCPs, but this
model was found to be an underestimate of the pavement’s strength at a given
thickness. Therefore, assuming the same maintenance plans and lifetime of an
equivalently designed JPCP is likely valid and enabled us to use Pavement ME to
model the degradation of RCCP. Indeed, if the RCCP and JPCP designs are at the
same thickness, it is possibly an overestimate of maintenance needed and an
underestimate of the lifetime of an RCCP, which will result in a potentially
conservative estimate of the LCA. The pavement designs in this paper were initially
based on a previous study focused on comparing JPCP to HMA (Swei et al., 2014).

Because RCCP is historically a rougher, higher IRI, road versus either JPCP or
asphalt, it is therefore important to estimate PVI during the pavement use phase
(Harrington et al., 2010; Zollinger, 2015; Chhorn et al., 2017). There is limited IRI
data on freshly prepared RCCPs (Chhorn et al., 2017), and we are unaware of any
measurements of the IRI of an aged RCCP to indicate how smoothness varies over
time. Therefore, sensitivity analysis was completed using Pavement ME to better
understand RCCP impact with varying IRI values, degradation models, and
maintenance plans on carbon emissions during the pavement use phase.

Finally, two different analyses were completed for the study. Most of the work
focuses on ordinary portland cement (OPC) as the sole cement in the mix design. We
also examined the impact of utilizing a low-carbon cement, a 50:50 blend of OPC and
slag cement, in both the JPCP and RCCP cases. The cement industry has been
pursuing options to lower carbon intensity, which has led to the fast implementation
of blended cements like Type 1L. RCC is typically produced in a pugmill mobile
plant with only one cement silo, which makes it challenging to produce concretes with
locally available supplementary cementitious materials (SCMs) or pozzolans to lower
carbon impact. As low clinker blends become available, the embodied emissions for
the cement-based pavements will decrease. Therefore, we also include one example to
examine how the addition of SCMs, or low-carbon cement blends like Type 1L or
LC3, could impact the GHG emissions of either RCCP or JPCP. The blend utilized
was 50% slag cement and 50% ordinary portland cement, which is atypical of RCCP
mix designs, but represents a possibility of what the future could look like.

Case Studies
General Assumptions. Several hypothetical case studies were employed to investigate
the sensitivity of parameters for RCCP’s LCA. These pavements were assumed to
serve as a rural state highway near Miami, Florida. We focused on Florida for this
study for two main reasons: (i) they diamond grind all concrete pavements to their
specifications to standardize initial smoothness, and (ii) the state Department of
Transportation has an emerging interest in RCCPs. A 30-year pavement design life
(DL) was generally assumed for all case studies based on Florida Department of
Transportation (FDOT) guidelines (FDOT, 2022). A 50-year analysis period (AP) (or

13th International Conference on Concrete Pavements 5

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

service life) was applied for life cycle analysis. Other pavement context assumptions
were adopted from a previous study (Swei et al., 2014) with necessary modifications
according to Pavement ME (version 3.0) (AASHTO, 2023), as summarized in Table
1. The JPCP and HMA cases served as the baselines for evaluating the LCA of
various RCCP case studies.

The significant variation in pavement materials and performance over time

makes direct comparisons challenging without a clearly defined functional unit
(Santero et al., 2010). In this study, the functional unit is specified as the life-cycle
GHG emissions (expressed in CO2-eq) of a four-lane pavement per mile over a 50-
year analysis period. Given that all three pavement types share the same base and
subgrade, and are subjected to identical traffic and environmental conditions, they are
considered equivalent in functionality (Santero et al., 2011). Additionally, our LCA
methodology integrates the effects of performance degradation on the pavement
during the use phase and facilitates the scheduling of M&R activities. This approach
ensures the pavement is well-maintained throughout its lifecycle, thereby improving
the ability to compare different alternatives (Inyim et al., 2016). Therefore, we set a
performance criterion for all pavement types, stipulating that they must maintain an
International Roughness Index (IRI) within a range of 1.0 to 3.75 m/km.

Table 2 shows the applied JPCP concrete and RCCP concrete mix designs,
respectively, which both reached a 27.6 MPa (4,000 psi) compressive strength at 28
days. As the cement contributes to 75-90% of the mix’s GWP (Gregory et al., 2021),
we adjusted the cement loading and emissions in Table 2 to conform with prevailing
Florida practices and ensure a more precise and localized implementation. It should
be highlighted that despite the differences in construction equipment between RCCP
and JPCP, the analysis assumes equivalent GHG emissions for RCCP construction as
those reported for JPCP, due to limitations in available data. As shown in Table 2,
NRMCA industry averages of aggregate proportions (NRMCA, 2019) were applied
for computing the material emission factors, 299.6 and 282.9 kg of CO2-eq/m3,
respectively. Regarding HMA, the material emission factor was 154.2 kg of CO2-
eq/m3, based on the NAPA EPD database (NAPA EPDs, 2023).

13th International Conference on Concrete Pavements 6

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

Table 1: General pavement design, pavement contexts, and traffic assumptions for JPCP,
RCCP, and HMA
Parameters Inputs
Traffic State Highway (Rural)
• Two-way Annual Average Daily Traffic (AADT) = 17,000
• Two-way Annual Average Daily Truck Traffic (AADTT) = 1,000
• Truck percent (%): 5.88
• Traffic growth rate: 3% linear growth
• Number of lanes for two directions: 4 (2 lanes in each direction)
• Directional Distribution Factor (DDF) = 0.5
• Lane Distribution Factor (LDF) = 0.9
• Design Lane AADTT = 450
• Traffic Speed: 56-72 km/h (35-45 mph)
Pavement JPCP RCCP HMA
Pavement • Joint Spacing: 3.66 m • Joint Spacing: • AC30 and PG64-
Design and • 28-day Modulus of 3.66 m 22
Material Rupture (MR): • 28-day MR: • Air Voids: 7%
Properties 4.1 MPa 4.1 MPa • Effective Binder
• PCC coefficient of • PCC coefficient of Content: 11.6%
thermal expansion: thermal expansion:
9.0 × 10−6 (1/oC) 9.0 × 10−6 (1/oC)
• Elastic modulus of • Epcc at 28
PCC (Epcc) at 28 days: 28.9 GPa
days: 28.9 GPa • Dowel bar: varies
• Dowel bar: 31.75 mm based on cases

Surface 6.35 cm surface layer

Thickness based on cases based on cases and 8.89 cm binder
layer
Base Layer 15cm Aggregate (Elastic Modulus=206.8 MPa)
Subgrade A-2-4 (Subgrade MR=113.7 MPa)
Layer
Maintenance
6.35 cm AC Mill and
& based on cases based on cases
Fill @ 20 and 37 years
Rehabilitation

Table 2: Mix design for JPCP and RCC.

Pavement
Mix Design Components
JPCP RCCP
Portland Cement ( 𝑘𝑘𝑘𝑘/𝑚𝑚 ) 3
297.23 281.80
Mixing Water ( 𝑘𝑘𝑘𝑘/𝑚𝑚3 ) 143.66 125.78
Coarse Aggregate ( 𝑘𝑘𝑘𝑘/𝑚𝑚3 ) 1168.74 984.25
Fine Aggregate ( 𝑘𝑘𝑘𝑘/𝑚𝑚3 ) 787.33 984.25
Air Entraining Admixture ( 𝑘𝑘𝑘𝑘/𝑚𝑚3 ) 0.030 0
Water Reducer ( 𝑘𝑘𝑘𝑘/𝑚𝑚3 ) 0.089 0
High Range Water Reducer ( 𝑘𝑘𝑘𝑘/𝑚𝑚3 ) 0 0
Accelerator ( 𝑘𝑘𝑘𝑘/𝑚𝑚3 ) 0.44 0
Material Emission Factor (kg of CO2-eq/m3) 299.6 282.9

Worst Case Scenario. We examined worst-case scenarios to simulate the poorest

possible performance of RCCPs from an LCA perspective. In JPCP Case 1, we
studied the original JPCP with a 225mm PCC slab thickness and 31.25mm dowel
bars. Maintaining the same slab thickness, we explored three more scenarios with
varying maintenance schedules for RCCPs. We assumed that RCCP joint faulting
could match undoweled JPCP faulting performance in the worst cases. To model

13th International Conference on Concrete Pavements 7

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

surface smoothness, we assumed RCCP would have poor smoothness, deteriorating

10% faster than undoweled JPCP with the same thickness. RCCPs were also assumed
to start with a rougher surface, registering an initial IRI of 1.74 m/km, while JPCP and
asphalt pavements had an initial IRI of 1.0 m/km.

To assess RCCP LCA sensitivity to M&R activities, we employed three

strategies, as shown in Table 3. RCCP Case 1 follows the same M&R as JPCP (Swei
et al., 2014), with diamond grinding (DG) and full-depth replacement (FDR)
scheduled at 30 years. RCCP Cases 2 and 3 consider additional DG and FDR for
improved smoothness. Asphalt pavements typically receive two AC mill and fill
treatments, as detailed in Table 1 (Swei et al., 2014).

Table 3: Surface thickness and dowel bar design as well as M&R schedule for JPCP and
RCCP case studies in the worst-case scenario
Parameters JPCP RCCP
JPCP Case 1:
Case Original JPCP Case 1 Case 2 Case 3
Design
PCC thickness (mm) 225 225 225 225
Dowel Bar (mm) 31.25 Undoweled Undoweled Undoweled
Initial IRI (m/km) 1.0 1.74 1.74 1.74
IRI after DG (m/km) 1.0 1.0 1.0 1.0
IRI deterioration Pavement ME 10% faster than 10% faster than 10% faster than
rate JPCP modeling JPCP JPCP JPCP
DG and 5% DG and 5%
DG and 5% DG and 5%
M&R Schedule FDR @ 15 and FDR @ 20 and
FDR @ 30 year FDR @ 30 year
30 years 40 years

Figure 2 (a) and (b) depict the degradation in IRI and the cumulative GHG emissions
for each case, respectively, during the analysis period. Figure 2 (c) employs stacked
bars to represent the contributions of various LCA components to the total GHG
emissions over a 50-year lifespan. In Figure 2 (a), it is evident that JPCP Case 1
maintains a consistently low IRI throughout the analysis period. This is due to
negligible transverse cracking and joint faulting, indicating an overly conservative
thickness design for this low-volume road. The IRI of the HMA baseline was
effectively kept below 1.6 m/km through two sequential M&R activities. On the other
hand, for all the RCC cases, there is a rapid deterioration in the Pavement ME-
predicted IRI as the pavement ages. For RCC Case 1, DG and FDR at year 30
significantly enhanced surface smoothness, reducing IRI from the highest value of
3.2 m/km to 1.0 m/km. However, as depicted in Figure 2 (b), M&R occurs too late
and is insufficient for GHG emission reduction during the use phase, resulting in the
highest GHG emissions among all cases. Moreover, RCC Case 1 GHG emission is
always higher than the baseline HMA case, indicating it is not an effective choice for
reducing the pavement’s carbon footprint.

In RCC Case 2, with M&R activities at years 15 and 30, IRI (Figure 2 (a))
shows reasonable improvement, and GHG emission (Figure 2 (b)) increases at a
slower rate compared to RCC Case 1, crossing with the baseline HMA at 32 years
indicating a lower GHG emissions after that point. In RCC Case 3, the crossover with
the HMA baseline occurs after 40 years due to the relatively late M&R schedule. The
analysis of Figure 2 (b) suggests that M&R timing is less sensitive to total GHG
emissions when a sufficient number of M&R treatments (e.g., 2 in this case) are

13th International Conference on Concrete Pavements 8

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

performed. Nevertheless, the total GHG emissions show a substantial dependence on

the frequency of M&R treatments.

Figure 2: Worst-case scenario: (a) IRI deterioration vs. analysis period, (b) the
corresponding GHG emissions over the same period, and (c) Stacked bar plot expressing
each component contributing to GWP

As shown in Figure 2 (c), the GHG emissions encompasses embodied

emissions (including initial, M&R, and EOL phases) and use-phase emissions
(including GHG emissions stemming from PVI deflection and roughness effects,
radiative forcing, and concrete carbonation). Notably, the initial emissions,
specifically those associated with materials and construction, significantly dominate
the embodied emissions across the lifecycle for concrete pavements. As for the HMA
baseline, the embodied emissions were evenly split between initial construction and
M&R. In the context of use-phase emissions, PVI roughness emerges as the dominant
factor for the three RCC cases, while PVI deflection plays a pivotal role in
determining use-phase emissions for HMA. Conversely, for JPCP, GHG emissions
attributed to PVI roughness are comparatively negligible when contrasted with the
other cases.

In terms of the total GHG emissions, the original JPCP exhibits the lowest
GHG emissions at 1.74 million kg of CO2-eq, owing to its consistently well-
maintained surface smoothness, as seen in Figure 2 (a). In contrast, RCC Case 1
records the highest GHG emissions, primarily driven by a substantial contribution
from PVI roughness. RCC Cases 2 and 3 exhibit similar total emissions at 3.73

13th International Conference on Concrete Pavements 9

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

million and 3.63 million kg of CO2-eq, respectively, underscoring the significant

impact of initial IRI after construction that contributes to excessive use-phase
emissions. The HMA baseline displays moderate GHG emissions over the life cycle,
totaling 4.05 million kg of CO2-eq. Overall, it is noteworthy that RCC pavements with
timely M&R treatments exhibit lower total GHG emissions than the HMA baseline.
Furthermore, the data indicates that reducing the PVI roughness component
significantly contributes to the reduction of both use-phase and overall emissions.

Best Case Scenario. In the best-case scenario, we aimed to establish upper-

performance bounds for RCC in the context of LCA. Since the original JPCP design
(JPCP Case 1) recommended a somewhat conservative PCC thickness (Swei et al.,
2014), we proposed an optimized PCC thickness for JPCP using Pavement ME
(AASHTO, 2023). This led to a final design of a 212.5mm PCC slab with 31.25mm
dowel bars. For RCC, we assumed the same PCC slab thickness with various potential
faulting scenarios, detailed in Table 4. Given that RCC mixes usually contain a higher
amount of aggregate, it was assumed that RCC pavements have a superior aggregate
load transfer efficiency (LTE) compared to that of JPCP at transverse joints
(Khazanovich et al., 2004; Harrington et al., 2010). Therefore, given the low traffic
speed, it was reasonably posited that the faulting performance of RCC at transverse
joints would be intermediate between that of undoweled JPCP and JPCP equipped
with 31.25mm dowels. To simulate the possible faulting performance of RCC,
different dowel diameters were employed in RCCP cases, as detailed in Table 4. Table
4 also includes the associated M&R treatments and the assumptions regarding the IRI.

Table 4: Surface thickness dowel bar design as well as M&R schedule for JPCP and
RCCP case studies in the best-case scenario
Parameters JPCP RCCP
Case JPCP Case 2 Case 4 Case 5 Case 6 Case 7

Dowel Bar (mm) 31.25 31.25 25 18.75 Undoweled

DG after DG after DG after DG after DG after
construction construction construction construction construction
M&R Schedule and DG & 1% and DG & and DG & and DG & and DG &
FDR @ 30 5% FDR @ 5% FDR @ 5% FDR @ 5% FDR @
years 30 years 30 years 30 years 30 years
Surface
212.5 212.5 212.5 212.5 212.5
Thickness (mm)
Initial IRI and
IRI after DG 0.79 0.79 0.79 0.79 0.79
(m/km)

Figure 3 (a) and (b) present the predictions for IRI and total GHG emissions
over the pavement life cycle, respectively, for each case in our best-case scenario.
Figure 3 (c) illustrates the contributions of each LCA component to the total GHG
emissions. As shown in Figure 3 (a), As anticipated, the IRI increases with
diminishing dowel bar diameter throughout the best-case scenario. The JPCP Case 2
and RCC Case 4, featuring 31.25mm dowel bars, maintain a consistently low IRI
throughout their 50-year lifespan. In contrast, RCC Case 7, lacking dowel bars,
registers the highest IRI predictions from Pavement ME, peaking at 2.4 and 2.8 m/km
at 30 and 50 years, respectively. Correspondingly, RCC Case 7 exhibits the highest
GHG emissions among rigid pavements after 18 years, as depicted in Figure 3 (b). All

13th International Conference on Concrete Pavements 10

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

doweled RCCPs (RCC Cases 4 and 5) exhibit similar GHG emissions throughout the
50-year analysis period. The HMA baseline remains the highest life-cycle total
emissions among all cases after 20 years.
Regarding total GHG emission (Figure 3 (c)), except for RCC Case 7, all other
rigid pavements display significantly lower use-phase emissions compared to the
worst-case scenario (Figure 2 (c)). This suggests that DG during the construction
phase can markedly enhance long-term surface smoothness and reduce GHG
emissions during the extended use of RCCPs. Conversely, RCC Case 7 could result in
considerably higher use-phase emissions. However, such a scenario is less likely to
occur with RCC due to its higher joint LTE as compared to undoweled JPCPs.

Figure 3: Best-case scenario: (a) IRI deterioration vs. analysis period, (b) the
corresponding GHG emissions over the same period, and (c) Stacked bar plot expressing
each component contributing to GWP

Low-carbon Concrete Pavement. Lastly, we explored the use of low-carbon mix

designs for both JPCP and RCC by adopting a 50:50 blend of ordinary portland
cement and slag cement, as derived from Table 2. All other mix components remained
consistent with the original JPCP and RCCP mixes, respectively. These choices were
informed by recommendations for low-carbon mixes found in Rangelov et al. (2022),
which have already been implemented in the practice of Fulton Road in Santa Rosa,
CA (Cemex, 2021). This investigation aimed to assess how these innovative low-
carbon mix designs could contribute to a reduced carbon footprint for concrete
pavements. As a result, the material emissions factors computed for these two low-
carbon JPCP and RCCP mixes were 192.0 kg of CO2-eq/m3 and 180.8 kg of CO2-
eq/m3, respectively.

13th International Conference on Concrete Pavements 11

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

In the case of low-carbon JPCP, the design and assumptions mirrored those of
the best-case scenario (Table 4). Similarly, for low-carbon RCCP, all other design and
M&R scheduling parameters aligned with the base-case scenario of RCC Case 6, as
shown in Table 4. Figure 4 represents the LCA results of these low-carbon designs.
Figure 4 (a) reveals that low-carbon mixes for JPCP yield significantly lower initial
embodied emissions compared to those presented in Figure 3 (c). Similar trends were
observed for low-carbon RCC, with a notable reduction of approximately 30% in
embodied emissions. Regarding total emissions, low-carbon JPCP and RCC GHG
emissions were 1.10 and 1.31 million kg of CO2-eq, respectively, while the HMA
baseline reached 4.05 million kg of CO2-eq. When assessing GHG emissions over
time, as shown in Figure 4 (b), the low-carbon JPCP and RCCP exhibit payback
periods of 10 and 9 years, respectively, when compared to the HMA baseline. This
suggests that by using a low-carbon concrete mix, the payback period for concrete
pavements can be nearly halved of those in the best-case scenario (18-20 years, see
Figure 3 (b)) when compared to the competing HMA design. These findings
underscore that the innovative low-carbon cement mix can be an effective approach to
reducing the embodied and total GHG emissions for concrete pavements, offering a
promising avenue for minimizing the carbon footprint associated with such
pavements.

Figure 4: Low-carbon JPCP and RCCP mix design option: (a) Stacked bar plot expressing
each component contributing to GWP; and (b) the GHG emissions over the analysis
period
Discussion
The case studies performed above highlight a few crucial findings. For the worst-case
scenario, the total GHG emissions over the analysis period are closely tied to IRI
performance, as shown in Figure 2. Maintaining a smooth surface significantly
reduces the release of GHG emissions. JPCP’s conservative thickness design, while
introducing more embodied emissions, notably reduces use-phase emissions, resulting
in the lowest total GHG emissions. In contrast, for RCCPs and HMA, implementing
two well-distributed M&R activities proves effective in lowering use-phase
emissions. These results underscore the importance of balancing initial design choices
and ongoing maintenance practices to reduce the environmental impacts of different
pavement types over their life cycles.

For the best-case scenario, according to the analysis presented in Figure 3, it

becomes evident that DG during construction can have a substantially positive impact
on surface smoothness for both JPCP and RCCP. This results in significantly lower
use-phase and overall GHG emissions when compared to the worst-case scenario.

13th International Conference on Concrete Pavements 12

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

Even though simulating faulting in undoweled JPCPs leads to higher IRI predictions
and increased use-phase emissions, practical RCC pavements typically outperform
undoweled JPCPs with much lower faulting and IRI performance due to higher joint
LTE. Considering this, in the best-case scenario, we reasonably equated RCCP’s
faulting performance to that of an 18.75-mm-dowel JPCP, especially in instances
where valid faulting models are unavailable in the Pavement ME. Ultimately, RCCP
and JPCP emerge as promising alternatives to asphalt pavements when assessed from
an LCA perspective.

Conclusions
Roller-Compacted Concrete Pavements (RCCPs) have seen increasing use in low-
volume roads. However, the existing literature lacks comprehensive studies on RCCP
for long-term performance and environmental impact, primarily due to limitations on
data and design models. Furthermore, previous studies have provided limited
documentation regarding the impact of maintenance and rehabilitation (M&R)
treatments, particularly the effect of diamond grinding (DG), when dealing with
RCCP surfaces characterized by relatively high roughness. To bridge this gap, we
conducted a series of case studies to assess the environmental impact and the
sensitivity of RCCPs within different scenarios, employing a full pavement life cycle
assessment (LCA) framework.

These case studies were categorized into two scenarios: worst-case scenario
and best-case scenario, serving to establish upper and lower boundaries for the
pavement performance and greenhouse gas (GHG) emissions associated with RCCPs’
life cycle. The findings highlight the critical role of initial surface treatment through
DG and the importance of adequate M&R to sustain pavement smoothness in
reducing GHG emissions. RCCPs lacking sufficient DG and subjected to delayed
M&R schedules tend to exhibit higher GHG emissions due to Pavement Vehicle
Interactions. On the other hand, with sufficient M&R treatments, both Jointed Plain
Concrete Pavement (JPCP) and diamond-ground RCCP tend to have lower GWP than
the asphalt pavement design selected in this study. We also explored various models
to replicate the realistic degradation of RCCP. The findings suggested that an
optimized RCCP design has the potential to achieve a 50% reduction in carbon
emissions when compared to an asphalt pavement alternative. Lastly, the
incorporation of supplementary cementitious materials into JPCP and RCC mix
designs provides a glimpse into the potential future of reduced carbon footprint for
concrete pavements.

References
AASHTO. (2015). Mechanistic-Empirical Pavement Design Guide: A Manual of Practice.
AASHTO.

AASHTO. (2023). AASHTOWare Pavement ME Design. (ARA, Inc.) Retrieved September 1,

2023, from https://me-design.com/MEDesign/Home.aspx

AzariJafari, H., Guo, F., Gregory, J., & Kirchain, R. (2021). Carbon uptake of concrete in the
US pavement network. Resources, Conservation and Recycling, 167, 105397.

Burwell, B. O., ACPA Arkansas, B. M., Delatte, N., Edwards, J., IHC, I., & Friess, S. (2018).
Roller-compacted concrete pavements as exposed wearing surface. ACPA Guide
Specification-Roller Compacted Concrete Pavements, 1998, 1-129.

13th International Conference on Concrete Pavements 13

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

Cemex. (2021). Statement of Mix Design For Roller Compacted Concrete (RCC) . Cemex.

Ceylan, H. (2010). Guide for Roller-Compacted Concrete Pavements. National Concrete

Pavement Technology Center.

Chhorn, C., Hong, S. J., & Lee, S. W. (2017). A study on performance of roller-compacted
concrete for pavement. Construction and Building Materials, 153, 535-543.
FDOT. (2022). Rigid Pavement Design Manual. Office of Design, Pavement Management
Section Tallahassee, Florida.
Gregory, J., AzariJafari, H., Vahidi, E., Guo, F., Ulm, F. J., & Kirchain, R. (2021). The role
of concrete in life cycle greenhouse gas emissions of US buildings and pavements.
Proceedings of the National Academy of Sciences, 118 (37), e2021936118.

Harrington, D., Abdo, F., Ceylan, H., Adaska, W., Hazaree, C., & Bektas, F. (2010). Guide
for roller-compacted concrete pavements. Portland Cement Association.

Holderbaum, R. E., & Schweiger, P. G. (2007). Guide for Developing RCC Specifications and
Commentary. Portland Cement Association.

Inyim, P., Pereyra, J., Bienvenu, M., & Mostafavi, A. (2016). Environmental assessment of
pavement infrastructure: A systematic review. Journal of Environmental Management, 176,
128-138.

Izevbekhai, B. (2022, 11). Comparative Analysis of Heavy Truck IRI-Induced Excess Fuel
Consumption of Various Network Pavement Interventions. Retrieved from National Road
Research Alliance Research Pays Off: https://www.youtube.com/watch?v=nnbGD1ORpxk

Khazanovich, L., Darter, M. I., & Yu, H. T. (2004). Mechanistic-empirical model to predict
transverse joint faulting. Transportation Research Record, 1896, 34-45.

Lemieux, G., & Thibodeau, C. (2017). Comparison of environmental impacts on the life cycle
of roller compacted concrete to asphalt pavement as part of the reconstruction of Boulevard
Industriel in Chateauguay. Montreal, Canada.

Li, H., & Khazanovich, L. (2022). Multi-gene genetic programming extension of AASHTO
ME for design of low-volume concrete pavements. Journal of Road Engineering, 252-266.

Louhghalam, A., Akbarian, M., & Ulm, F. J. (2014). Scaling relationships of dissipation-
induced pavement–vehicle interactions. Transportation Research Record, 2457(1), 95-104.

Louhghalam, A., Akbarian, M., & Ulm, F. J. (2015). Roughness-induced pavement–vehicle

interactions: key parameters and impact on vehicle fuel consumption. Transportation
Research Record, 2525(1), 62-70.

Louhghalam, A., Akbarian, M., & Ulm, F. J. (2017). Carbon management of infrastructure
performance: Integrated big data analytics and pavement-vehicle-interactions. Journal of
Cleaner Production, 142, 956-964.

Montgomery, R., Schirmer Jr, H., & Hirsch, A. (2015). Improving environmental
sustainability in road projects. WORLD BANK GROUP REPORT NUMBER 93903-LAC.
NAPA EPDs. (2023, 10 20). Emerald Eco-Label EPD Tool. Retrieved from
https://asphaltepd.org/

13th International Conference on Concrete Pavements 14

Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

NASA. (2022, 07 22). National Aeronautics and Space Administration. Retrieved from The
POWER project: https://power.larc.nasa.gov/data-access-viewer/

NRMCA. (2019). Industry-Wide Environmental Product Declarations. ilver Spring, MD,

National Ready Mix Concrete.

Piggott, R. W. (1999). Roller-compacted concrete pavements: A study of long term

performance. Portland Cement Association.

Rangelov, M., Dylla, H., Dobling, B., Gudimettla, J., Sivaneswaran, N., & Praul, M. (2022).
Readily implementable strategies for reducing embodied environmental impacts of concrete
pavements in the United States. Transportation Research Record, 2676(9), 436-450.

Robbins, M. M., & Tran, N. (2015). Literature Review: The Impact of Pavement Roughness
on Vehicle Operating Costs. Retrieved from http://www.ncat.us/files/reports/2015/15-02.pdf

Santero, N. J., Masanet, E., & Horvath, A. (2011). Life-cycle assessment of pavements. Part I:
Critical review. Resources, Conservation and Recycling, 55(9-10), 801-809.

Santero, N., Masanet, E., & Horvath, A. (2010). Life-Cycle Assessment of Pavements: A
Critical Review of Existing Literature and Research, SN3119a,. Portland Cement Association,
Skokie, Illinois, USA.

Skolnik, J., Brooks, M., & Oman, J. (2013). Fuel usage factors in highway and bridge
construction (Vol. 744). Transportation Research Board.

Swei, O., Xu, X. N., Wildnauer, M., Gregory, J., & Kirchain, R. (2014). Supplementary
information for comparative pavement life cycle assessment and life cycle cost analysis.
Massachusetts Institute of Technology.

Townsend, T. G., Ingwersen, W. W., Niblick, B., Jain, P., & Wally, J. (2019). CDDPath: A
method for quantifying the loss and recovery of construction and demolition debris in the
United States. Waste Management, 84, 302-309.

Umer, A., Hewage, K., Haider, H., & Sadiq, R. (2017). Sustainability evaluation framework
for pavement technologies: An integrated life cycle economic and environmental trade-off
analysis. Transportation Research Part D: Transport and Environment, 53, 88-101.

US EPA. (2020). Construction and demolition debris management in the United States, 2015.
U.S. Environmental Protection Agency Office of Resource Conservation and Recovery .

World Bank. (2005). Transport—Greenhouse Gas Emissions Mitigation in Road

Construction and Rehabilitation: A Toolkit for Developing Countries. World Bank.

Wu, Z., Mahdi, M., & Rupnow, T. (2016). Performance of Thin RCC Pavements Under
Accelerated Loading. Springer.

Xu, X., Akbarian, M., Gregory, J., & Kirchain., R. (2019). Role of the use phase and
pavement-vehicle interaction in comparative pavement life cycle assessment as a function of
context. Journal of Cleaner Production, 230, 1156-1164.

13th International Conference on Concrete Pavements 15

 Li, Levin, Azarijafari, Gregory, Kirchain, Mack, Zollinger and Mahdi

Xu, X., Swei, O., Xu, L., Schlosser, C. A., Gregory, J., & Kirchain, R. (2020). Quantifying
location-specific impacts of pavement albedo on radiative forcing using an analytical
approach. Environmental Science & Technology, 54(4), 2411-2421.

Zollinger, C. (2015). Recent advances and uses of roller compacted concrete pavements in
the United States. Paving Solutions.

13th International Conference on Concrete Pavements 16

View publication stats