Content uploaded by Vahid Faghihi
Author content
All content in this area was uploaded by Vahid Faghihi on Apr 29, 2023
Content may be subject to copyright.
Sustainable Cities and Society 84 (2022) 104013
Available online 19 June 2022
2210-6707/© 2022 Elsevier Ltd. All rights reserved.
Optimizing sustainability of infrastructure projects through the integration
of building information modeling and envision rating system at the
design stage
Avin Laali
a
, Seyed Hossein Hosseini Nourzad
a
,
*
, Vahid Faghihi
b
a
Department of Construction and Project Management, School of Architecture, University of Tehran, Tehran, Iran
b
Construction Science Department, School of Architecture, Prairie View A&M University, Prairie View, TX, United States
ARTICLE INFO
Keywords:
Infrastructure
Building information modeling
Sustainable design
Optimization
Envision
Sustainability evaluation systems
ABSTRACT
With infrastructure services demand rising, a signicant contributor to sustainable development goals, infra-
structure systems have to evolve and adapt effectively, efciently, and sustainably. The broad set of decision
variables coupled with the trade-off among the economic, social, and environmental aspects complicate sus-
tainability assessment and optimization during the design stage. Thus, this paper presents a Building Information
Modeling (BIM)-based automated framework for real-time evaluation and optimization of sustainability in
infrastructure projects. The proposed framework benets from infrastructure sustainability rating systems that
provide a systemic and balanced set of indicators for careful consideration of sustainability. On the other hand, to
automate the process, BIM, with its 3D shared environment, is used as an integrated platform for dynamic
sustainability analysis. The proposed framework integrates both novelties in real-time during the early design
stages, which helps designers to select the most sustainable alternative. A prototype and a hypothetical case
study are conducted to validate the framework’s applicability through Infraworks 360 as the operating BIM
platform and Envision as the baseline rating system. The results conrm that adopting the suggested method
simplies the inclusion of sustainability into design decisions while facilitating documentation of compliance
with credit assessment for both project teams and verication agencies.
1. Introduction
Despite the environmental consequences, humankind seems unwill-
ing to give up the services provided by the facilities they have built. As a
result, they seek refuge in the most known denition of the concept of
“Sustainable Development,” which has attracted signicant attention in
recent years as development that serves current needs without jeop-
ardizing future generations’ ability to meet their own (WCED, 1987).
This statement implies that we want to develop such amenities while
minimizing or even reversing their adverse impacts. In addition, the 17
Sustainable Development Goals (SDGs) issued by the United Nations
have challenged every industry worldwide to realign its strategies and
efciently carry out sustainable objectives (Marzouk & Othman, 2020;
UN, 2015). As a great contributor to global warming, the construction
industry is also under extreme pressure to stop such damages and go
beyond measures to repair such losses (Tryggestad, 2013). Infrastruc-
ture projects as a central and impactful part of this industry contribute
directly to SDGs 9 and 11 (Clevenger, Ozbek & Simpson, 2013; UN,
2015). Therefore, sustainable development and planning of in-
frastructures become a must as also issued by.
However, infrastructure design is generally complicated because of
the various factors and considerations involved in their design, which
dene signicant limitations and strictly condition their design (Bon-
giorno, Bosurgi, Carbone, Pellegrino & Sollazzo, 2019). Additionally,
sustainability is also commonly thought of and analyzed in terms of the
balance between different economic, social, and environmental di-
mensions, which is not readily apparent to decision-makers and stake-
holders during the early design process (IDB, 2018; Shahtaheri, Flint &
De LA Garza, 2018). Thus, nding an optimum solution becomes even
more of a complex task. These delicate tasks necessitate the provision of
timely, dependable, and practical solutions that can meet multiple,
potentially contradictory aspects of sustainability. Therefore, many ef-
forts have been made to develop novel tools and algorithms to support
the designers in this intricate task (Rees, 2010). Building Information
* Corresponding author.
E-mail addresses: avin.laali@ut.ac.ir (A. Laali), hnourzad@ut.ac.ir (S.H.H. Nourzad), vafaghihi@pvamu.edu (V. Faghihi).
Contents lists available at ScienceDirect
Sustainable Cities and Society
journal homepage: www.elsevier.com/locate/scs
https://doi.org/10.1016/j.scs.2022.104013
Received 9 March 2022; Received in revised form 20 May 2022; Accepted 18 June 2022
Sustainable Cities and Society 84 (2022) 104013
2
Modeling (BIM) tools are among these proposed solutions and novelties.
BIM can be characterized as an integration of processes and technologies
that run throughout an asset’s development phases to facilitate the
generation, storing, managing, and sharing of physical & functional
data. In short, BIM is a digital representation of the facility rich with data
and parametric rules that allow intelligent modications (AGC, 2005).
BIM stores data through articial intelligence (AI) and by using
specialized “smart objects” that depict the realistic element in the 3D
scene. Each smart-object consists of features that guarantee a full
comprehension of the object’s qualities, role, interactions, and inuence
on the external context, enabling various consistent and integrated el-
ements analyses (Chen, Lok, & Jeng, 2016). Therefore, when changes
are made, smart objects are automatically updated (e.g., any alterations
to the horizontal alignment, vertical prole, or road cross-section of a
road have an instantaneous impact on the other representations) (Lorek,
2018). In general, this innovation can considerably minimize errors and
inaccuracies due to the human mind’s cognitive limitations (Hergunsel,
2011).
Sustainability rating systems can be regarded as the product of the
same objectives, recognized as frameworks that provide comprehensive
coverage of all sustainability aspects and their scientic interrelation-
ships, allow setting of sustainable priorities and goals, and enable
effective stakeholder communication (Ando et al., 2005; Cole, 2003;
Liu, Van Nederveen, Wu, & Hertogh, 2018). At the same time, since the
quantitative analysis of data is critical for digital sustainability assess-
ments, the quantitative approach to viewing the sustainable develop-
ment of these frameworks can serve as the foundation for such efforts.
However, their utilization can be time-consuming due to the interactive
data computation that seeks the best tting design that achieves the
desired degree of sustainability (Carvalho, Bragança & Mateus, 2019).
Despite the infrastructure sector’s proven potential and greater
impact, the building industry is well ahead in developing, applying, and
integrating both novelties (Liu, Van Nederveen, Wu, & Hertogh, 2018;
Mcvoy, Nelson, Krekeler, Kolb & Gritsavage, 2010). For example, Bis-
was, Wang & Sung-Hsien (2008) created a program that uses BIM
technology to evaluate the environmental effects of design decisions.
More recently, Carvalho et al. (2019) examined how BIM might aid in
the optimization of building sustainability assessment methodologies,
concentrating on the SBToolPT-H (Sustainable Building Tool (SBTool) in
the context of Portuguese residential buildings). The ndings of the
SBToolPT-H reveal that BIM may be used to examine 24 of the 25
criteria directly or indirectly. Their research comprised one of the rst
attempts to integrate BIM into green building grading and certication.
Some studies validate the same theory in the infrastructure sector but
rarely present practical answers while struggling with nding the proper
platforms and interoperability issues (Liu, Van Nederveen, Wu, & Her-
togh, 2018). Hence, this paper’s novelty addresses the lack of awareness
of the potentialities of BIM for the effective and automated use of sus-
tainability rating systems as benchmarks to develop a sustainability
optimization method.
Hence, this study aims to bridge the mentioned gaps by proposing a
practical solution and presenting an automated BIM-based framework
for sustainability optimization at the design stage of infrastructure
projects. In this way, infrastructure designers can instantly be informed
of a potential design alternative’s sustainability impact. They are also
presented with optimized sustainability alternatives based on the base-
line design and sustainability evaluation system requirements. The
framework’s applicability will be tested by developing a prototype that
uses Infraworks 360 (Chappell, 2015) as the BIM platform and Envision,
version 3 (ISI, 2018b, 2021) for a sustainability evaluation system for
road alignment projects. This prototype employs a genetic algorithm
(GA) as the operating optimization algorithm to support
decision-makers with sustainably optimized and modiable BIM-based
solutions.
2. Literature review
2.1. Sustainability and infrastructure sustainability assessment tools
Although holistic denitions of sustainability have generated sus-
tainable construction industry policies, they are simply not detailed
enough. As a result, the demand for establishing and evaluating envi-
ronmental performance indicators led to the development of several
sustainable “building” rating systems (Cole, 2003; Diaz-Sarachaga,
Jato-Espino, Alsulami & Castro-Fresno, 2016). Afterward, broadening
these assessment tools’ scope to quantify the entire built environment’s
impact on sustainability pillars resulted in the development of sustain-
ability rating tools for infrastructure projects (Ferrer, Thom´
e & Scav-
arda, 2018; Mcvoy et al., 2010). Although various sustainability
assessment tools and rating systems are available, few have the integral
approach reecting various aspects of sustainable development that
enable assessing a wide range of infrastructure projects (Diaz-Sarachaga
et al., 2016). Envision in the USA (ISI, 2018b, 2021) along with
CEEQUAL in the UK (BreGlobal, 2021), and the Infrastructure Sustain-
ability rating scheme in Australia (ISCA, 2021) is among those few that
have the integral approach that reect various aspects of sustainable
development of a wide range of infrastructure projects regardless of
their type or size.
Due to the relatively recent introduction of these tools, there are
limited project-level studies. Studies have typically concentrated on the
overall benets and drawbacks of rating tools (Ainger & Fenner, 2014;
French, 2012; Hurley, 2009) and evaluations of individual rating tools
(Willetts, Burdon, Glass & Frost, 2010). Grifths, Boyle and Henning
(2017) observed that the tools address comparable impact areas, but the
content, strategy, and opportunities to improve sustainability perfor-
mance vary. The observed that CEEQUAL rewarded incremental actions
more than the other tools, which is likely to suit project teams new with
sustainability principles and techniques. Moreover, Envision was noted
as a valuable reference enabling early project planning and assessment
of larger sustainability and community concerns. It also provides guid-
ance to infrastructure owners who want to push the envelope in terms of
restorative efforts and long-term planning. IS, like Envision, placed a
premium on stakeholder and community participation in sustainability
actions and decisions, which had previously been identied as critical to
achieving sustainable outcomes.
Envision, established in 2012 by the Harvard Graduate School of
Design’s Zofnass Program for Sustainable Infrastructure and the Insti-
tute for Sustainable Infrastructure (ISI), is intended to evaluate and score
the overall contribution of infrastructure projects to sustainability based
on the TBL (ISI, 2018a). This comprehensive approach to infrastructure
development seeks to assess projects in terms of their value to commu-
nities, efcient use of funds, and contribution to long-term sustainabil-
ity. Envision also considers all aspects of the lifecycle, allowing for
better-informed decisions in all stages of planning to deconstruction or
decommissioning (Gaughan, 2012; Saville, Miller & Brumbelow, 2016).
Although Envision is a self-assessment tool, ISI provides an optional
third-party verication mandatory for awards.
The rating is based on 64 criteria, known as credits, organized into
ve categories: Quality of Life, Leadership, Resource Allocation, and a
unique category of Natural World and Climate and Risk (Infrastructure
& Infrastructure, 2018; ISI, 2018a, 2015). Credits are grouped to assist
users in managing the complicated trade-offs and synergies between
them. Each credit is assigned points weighted in line with its estimated
contribution to sustainability. Each of them is rated according to the
following achievement levels, from lowest to highest: Improved,
Enhanced, Superior, Conserving, and Restorative. However, there are
not ve degrees of achievement for every credit as the level degrees are
established by the credit’s type and the capacity to distinguish between
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
3
levels. Credits that do not apply to the specic project are marked as
(N/A
1
), thus removing them from consideration.
2.2. Design stage
Undoubtedly, measures must be taken at the early design stages to
effectively evaluate and enhance an asset’s nearly permanent and
widespread sustainability impacts (Azhar, Carlton, Olsen & Ahmad,
2011). Many studies support this, demonstrating the point of action that
suggests the most potential for sustainability improvements are the early
decision-making stages of an asset (Basbagill, Flager, Lepech & Fischer,
2013; Bueno & Fabricio, 2018). Shahtaheri et al. (2018) proposed a
multi-criteria preference assessment framework for decision-makers to
make more sustainably informed decisions regarding early design al-
ternatives for commercial buildings. Sabatino, Frangopol and Dong
(2015) presented a sustainability-based maintenance optimization de-
cision support for highway bridges on the transportation front. How-
ever, sustainability solutions cannot be formed when there is a lack of
adequate data regarding the sustainability impacts of different strategies
and design alternatives (Shahtaheri et al., 2018). Especially in infra-
structure projects with a more sizeable interaction with their sur-
rounding environment, creating a realistic, responsive 3D model of that
environment becomes ever more obligatory (Bongiorno et al., 2019).
Therefore, BIM as a modeling platform that captures physical and
functional interactions for sustainability optimization becomes evident.
While 3D modeling has become more widely used in building design
and engineering, 2D drawing-based procedures are still commonly used
in infrastructure (Chong, Lopez, Wang, Wang & Zhao, 2016; Kim, Kim,
Ok & Yang, 2015). Bradley, Li, Lark and Dunn (2016) observed that
despite the numerous studies dedicated to BIM-based research at the
design stage of building projects (Hollberg, Genova & Habert, 2020;
Santos, Costa, Silvestre & PYL, 2020; Wong & Zhou, 2015), few solutions
had been presented for infrastructure projects (Minagawa & Kusaya-
nagi, 2015). Nevertheless, recent research aims to remove the barriers to
BIM adoption in the infrastructure domain (Chan, Olawumi & Ho, 2019;
Hartmann, Van Meerveld, Vossebeld & Adriaanse, 2012; Ji, Borrmann,
Beetz & Obergrießer, 2013). By referring to best practices in the building
domain, the aim is to examine and utilize BIM’s potential for improving
the environmental impacts of designs (Liu, Van Nederveen, Wu, &
Hertogh, 2018; van Eldik, Vahdatikhaki, Dos Santos, Visser & Doree,
2020). Several are crucial for improving sustainability evaluation and
optimization among these potentialities, as Bongiorno et al. (2019)
mentioned. These include but are not limited to the “smart objects” in
the BIM environment that reect the project’s key elements in 3D,
representation of the individual pieces and their interactions, and an
extensive relational database of all of the relevant linked information
that can inuence and condition the many project phases.
2.3. BIM and infrastructure sustainability
Infrastructures are supporting systems to sustain human civilization
and activities. The overall scope of infrastructure development includes
transportation systems, which are viewed as the foundation of a pros-
perous economy and social development (Costin, Adibfar, Hu & Chen,
2018). Notably, efcient and innovative technologies are needed to
support the development of new infrastructure services and, often, the
replacement of aging transportation structures (Adshead, Thacker, Ful-
dauer & Hall, 2019). Among these technologies, BIM stands out with its
3D modeling technology, a design process that has changed how the
construction industry operates. The building information model is
interpreted as the product, while the building information modeling is
described as the process (Hooper & Ekholm, 2012).
Although BIM employment in infrastructure projects has recently
received much attention, sustainability evaluation and improvement
through BIM are yet to receive practical solutions. In a study performed
by Liu, Van Nederveen, Wu, & Hertogh (2018), a conceptual framework
based on interviewees’ assessments combined with a critical analysis of
technical requirements guiding the integration is presented, allowing for
what-if scenarios to better support incorporating sustainability into
design decisions. In a literature review of 471 scholarly bibliographies
performed by Liu, LU, Shen and Peh (2020), one of the four main
identied research gaps was the lack of a comprehensive overall
building performance framework to support BIM. Of the few and most
recent studies addressing both of the considerations, van Eldik et al.
(2020) used BIM as a platform for automation of the environmental
impact assessment in the early design stages of infrastructure projects to
provide designers with accurate results of the environmental impact of
all objects associated with the projects’ design.
Smart cities and growing cities need BIM input to help ensure a
stronger adherence to sustainability in our fragile and ecologically un-
stable environment (Marzouk & Othman, 2020). Engaging sustainability
considerations in construction projects’ processes can contribute to
more complexity (Rahmani Asl, Zarrinmehr, Bergin & Yan, 2015). Be-
sides, there is growing attention to the use of BIM as an all-encompassing
source of data to optimize computationally complex problems that
require considerable amounts of data for more accurate analysis. Mar-
zouk and Othman (2020) integrated BIM and the geographical infor-
mation system (GIS) to plan smart cities, starting with the utility
infrastructure for growing cities and those cities still in the formation
stage. They considered wastewater management and freshwater pro-
tection, as well as electrical needs. After dividing the city into different
plots, they analyzed land use, building features, and other information
to help them build the smart city development strategies.
2.4. BIM and optimization
While the assessment and comparison of sustainability impacts of
single alternatives can appear to offer all the right solutions, generating
a globally optimal model through such methods can become highly
complex. Optimization methods can achieve high-performance alter-
natives while overcoming the difculty of balancing conicting and
potentially complex objectives of infrastructure projects, especially
under TBL objectives, for maximum performance achievement
(Nguyen, Reiter & Rigo, 2014; Rahmani Asl et al., 2015; Thangaraj,
Pant, Abraham & Bouvry, 2011). Despite the proven benets of BIM as a
consistent and parametric platform, not many studies that propose
sustainability optimization frameworks at the design stage of infra-
structure projects utilize it as an accurate source of necessary project
information that also provides dynamic access to decision variables
(Bongiorno et al., 2019; Rahmani Asl et al., 2015; van Eldik et al., 2020).
According to Geyer (2009), applying optimization methods at the
design stage of construction projects has a twofold benet: it may lead to
better-performing solutions and a greater understanding of the design
space or the range of viable options. The design of infrastructure projects
can be complex due to the many involved factors and objectives coupled
with sustainability issues (Shahtaheri et al., 2018). Optimization algo-
rithms are a powerful method for supporting this process by comparing
thousands of solutions with low computation efforts (Bongiorno et al.,
2019; Geyer, 2009). While sustainability applications of BIM for build-
ings and especially optimizing their sustainability issues have become
more sophisticated over the past 20 years, knowledge and practical so-
lutions in infrastructure projects are scarce (Liu, Van Nederveen, Wu, &
Hertogh, 2018).
Among studies that prove the benecial synergies of combining
optimization method and BIM for sustainable development of buildings
is a study performed by Liu et al. (2015). This study used Ecotect
Analysis by Autodesk Inc. (Yan et al., 2009) for a Particle Swarm Opti-
mization (PSO)-based optimization of life cycle costs (LCC) and life cycle
carbon emissions (LCCE) of building designs. Also, Rahmani Asl et al.
1
Not Applicable.
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
4
(2015) presented a building performance optimization based on per-
formance simulation on top of Revit, a widely used BIM platform. This
combination could generate and assess building model designs and
search for the best sustainably accomplished options to aid designers in
performing sustainability performance design analysis. As stated, very
few studies addressed the potentialities of BIM as an enabler to access
full range of information for sustainability performance optimization of
infrastructure projects. As such, Bongiorno et al. (2019) investigated the
possibility and advantages of such integration in the case of 3D highway
alignment optimization. A study performed by Bongiorno et al. (2019) is
among the few studies that address the possible advantages of inte-
grating optimization of highway alignments with BIM serving as a
comprehensive source of data and enabler of model visualization and
modication.
3. Method
This research was carried out to develop a BIM-based framework for
the sustainability optimization of infrastructures. A prototype was
developed and tested on a hypothetical case study to validate the
framework’s applicability.
3.1. Framework for BIM-based sustainability optimization
The framework’s broad applicability is provided through its modular
design, allowing methods, tools, and types of each of the four compo-
nents to be altered: sustainability rating tool, infrastructure project,
optimization algorithm, and the BIM platform, as shown in Fig. 1.
Fig. 1 also represents the inter-relationship among these primary
elements and their transactions. The infrastructure type indicates the
applicable criteria from the baseline rating system and design con-
straints and parameters for the optimization algorithm. In this study,
road alignments were selected to implement the framework. Trans-
portation infrastructures are a vital part of any civilization’s economy,
safety, and well-being. There is a clear need to advance the existing and
future transportation infrastructures to adjust and go beyond the
growing concerns revolving around sustainability issues (Costin et al.,
2018). Additionally, this choice was made due to the congested presence
of this infrastructure type, with over four million miles of road network
in the United States and an annual average increase of 0.4 percent,
making it one of the most prevalent and vital forms of infrastructure
(USDOT, 2018).
In addition to providing a comprehensive set of sustainability
criteria, the sustainability rating system (Envision) determines the rules
and formulas for assessing the infrastructure’s sustainability perfor-
mance, hence, the sustainability constraints and tness function of the
optimization algorithm. In this study, Envision system was selected as
the baseline rating system. The main reasons for choosing Envision in
this study are as follows:
a) This rating system can ensure and evaluate the sustainability of all
types of infrastructure projects of different sizes.
b) It can also be exible since “the nal Envision score is presented as a
percentage of the total applicable points.” (Envision, 2020).
c) The nal reason for selecting Envision is the inter-relation of its
credits which enlightens the effects of gaining points from one credit
based on other credits.
Therefore, in this study, one of the mainstream sustainability
assessment tools, Envision, was selected as the baseline evaluation tool
for the comprehensiveness and scalability of the framework.
Additionally, the I-BIM relational database can serve as a compre-
hensive data source for the optimization algorithm. This is done through
the smart-object information included in the BIM relational database.
Smart objects are not only a realistic representation of any element in a
3D scenario but rather a collection of features that ensure an in-depth
representation of each object’s characteristics, roles, interactions, and
inuence on the external context. This study will use Autodesk Infra-
works 360 as the BIM platform for implementing the proposed frame-
works as the most compatible BIM tool concerning its capabilities of
modeling the selected infrastructure type (roads) and its decision vari-
ables through road alignment smart objects in a realistic manner. For
instance, for any road alignment smart object, a design speed can be
determined manually and automatically that species the minimum and
maximum radius, spiral length, and tangent length based on predened
road design standards (e.g., AASHTO standard). Infraworks 360 was
used as the authoring platform to evaluate the integrated BIM-based
framework’s applicability and develop the prototype in JavaScript
format using the scripting console feature to access and manipulate data
through the API. This platform can also incorporate project and common
data through modeling or importing features.
Lastly, once the objective function, which maximizes achievable
points from Envision, and the design and environmental constraints are
set, the optimization algorithm could be run using the scripting console
of the BIM tool to load the developed optimization algorithm’s coding.
Outputs are then produced, being the optimal solution modeled and
Fig. 1. The basic components of the framework.
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
5
displayed in the BIM platform as a road alignment 3D smart object and
its detailed points achieved from the sustainability evaluation rating
system. As mentioned, the BIM platform provides the input for calcu-
lating the required value by the rating system to create an optimal
solution.
The developed prototype is based on a road alignment infrastructure
project design in this study. Among the various mathematical models
and optimization methods, genetic algorithm (GA) approaches and the
swarm intelligence (SI) methods are both iterative and evolutionary
techniques. They provide the benets of efciency and easy model
formulation and have proven reliable. The rst study for a GA-based
search model for road alignment optimization was performed by Jong
(1998). That research considered alignments as a series of 3D points of
intersection (PIs) distributed in equally distanced vertical cutting
planes. Many other studies followed to perfect the model’s efciency
and results (Jong et al., 2000; Kang et al., 2007; Kang et al., 2009; Kim
et al., 2007). The model formulation in this study is inspired by Al-
Hadad’s work (Al-Hadad, 2011). This study considers any road align-
ment as a series of station points that are often spaced evenly and dene
the nal alignment. Unlike former methods that determine the place-
ment of these points secondary to the alignment form when completing
the nal details, we looked at the series of station points that make up
the alignment as a primary to the alignment form.
3.2. Model implementation
The proposed framework is implemented in two main phases: (1)
developing a BIM-based sustainability assessment algorithm and (2)
developing a BIM-based sustainability optimization algorithm. The
development of the sustainability assessment algorithm is required to
evaluate the produced solutions (i.e., chromosomes) in the optimization
algorithm. The procedure for implementing the proposed framework is
depicted in detail in Fig. 2.
3.3. Case study
A potential highway alignment was considered to test the designed
prototype’s efciency and applicability in sustainability assessment and
optimization of the alignments and the proposed method’s accuracy.
The study area is located between Alachua and Bradford counties in the
southeastern region of Florida, USA. The start- and end-points of the
potential highway connect CR-18 and CR-1471 in Florida. Although the
potential road alignment is hypothetical, real-world geographic data is
used for this case study. The study site area is about 78
2
miles as the
model builder feature of the Infraworks platform allows a maximum of
such area for creating BIM models of potential projects. The Euclidean
distance between the start- and end-points is about 8 miles. The reason
for this choice of the case study area was threefold. (1) Although Envi-
sion can be used worldwide, it is a US-based sustainability rating system
used extensively in the USA and Canada. Therefore, the study area was
considered in the USA for this study’s purposes and the selected credits’
applicability. (2) This study’s scope only examines a 2D highway model
optimization. Therefore, the suggested alternatives were given fewer
variations in terms of earthwork; to be pragmatic, the state of Florida
was selected as it has low height variations. (3) Access to accurate data
concerning the geographical locations of wetlands and conservation
areas from this area in the formats of shapeles was possible.
3.4. The BIM-based sustainability assessment algorithm
The rst phase in achieving the study’s objective is creating a sus-
tainability assessment algorithm based on the selected sustainability
rating system. This module will provide the means for calculating the
tness of the produced chromosomes in the genetic algorithm. As the
rst step, applicable credits from Envision to the selected infrastructure
should be identied. For that purpose, former studies that have
proposed road alignment optimization models as minimization of their
cost components were studied to determine the Envision’s credits
equivalent to the cost components for validating their applicability to
road alignments as presented in Figs. 3 and 4. As some decision variables
are not evaluated under credits from Resource Allocation and Climate
and Resilience categories, the horizontal curvature, vertical curvature,
and grade violation cost components do not have an immediate link to
these credits (Fig. 4).
These cost components include but are not limited to location-
dependent, length-dependent, and earthwork costs. In this approach,
the best solution costs the least as the optimization model is objective
function is formulated to minimize the overall cost of the potential so-
lution (Kang, Jha & Schonfeld, 2012). The design constraints in these
algorithms are imposed as penalizing costs, meaning that violating these
design constraints leads to cost increment of the generated solutions
and, therefore, decreases their tness (Al-Hadad & Mawdesley, 2010).
The applicability of these credits to the project type was based on the
indicated cost components and sustainability design items of highway
projects proposed by Tsai and Chang (2012). Since quantitative criteria
are more compatible for assessment and optimization automation to
eliminate the possibility of cognitive bias, the sustainability assessment
tool was studied to select quantitative criteria in the second step. Table 1
presents an example of analyzing two credits from Envision to distin-
guish quantitively measurable credits from quantitative credits and
identify the necessary data for the evaluation and the BIM method that
may be utilized to acquire the relevant data.
Former studies that proposed road alignment optimization models
mainly consider the objective function as a minimization function of
conicting cost components (i.e., a decrease in one cost component may
increase another). Hence, the lower a solution’s total cost, the tter it is.
The denition of the optimization problem in this paper is based on
conicting criteria as well; however, instead of cost components, sus-
tainability credits form the objective function. (Al-Hadad & Mawdesley,
2010; Jha & Schonfeld, 2000; Tsai & Chang, 2012). Thus, the objective
function is a maximization of the achievable scores from Envision
credits, which are inherently conicting. Hence, the higher the score,
the tter the solution.
The selected sustainability assessment criteria, dened and decoded
as mathematical rules and indicators, are then translated into sustain-
ability score achievement scenarios based on the infrastructure’s design
parameters and decision variables. These evaluation scenarios are in-
tegrated with the BIM tool to form an automated sustainability evalu-
ation functionality that can be applied to any proposed infrastructure
BIM-based model and return the achieved sustainability score, as
shown in Fig. 5. Lastly, the optimization problem formulation is per-
formed based on the BIM tool’s data formats and the integrated sus-
tainability assessment algorithm, translating into the objective function.
3.5. The BIM-based sustainability optimization algorithm
The development of the optimization algorithm in this study can be
summarized in ve main steps:
1 Formation of chromosome structure appropriate to the problem;
2 Determination of the evaluation conditions (objective function);
3 Creation of a random initial population of chromosomes within the
boundary of the study area (initial answer);
4 Selection of an appropriate mechanism for producing offspring (se-
lection, crossover, mutation);
5 Termination criteria must also be set to stop the algorithm when
those conditions are met.
3.5.1. Chromosome representation
The rst step of formulating a GA optimization is setting the proper
chromosome conguration structure, which will be based on the
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
6
infrastructure evaluation parameters and the data format of the opti-
mization problem’s search area. This paper addresses the assessment
and optimization of horizontal road alignments; as such, the model
formulation is based upon the fact that any road alignment created with
whatever method can be dened by the coordinates of its PIs. Therefore,
the suggested optimization model considers the X and Y coordinates of
the PIs as decision variables for horizontal alignment optimization, and
the arrangement of its PI series can determine the placement of the
highway alignment. This model uses the GA to nd the best path axis
candidate. In this model, the road alignments are represented by chro-
mosomes combined of genes dened by the intersection points’ 2D co-
ordinates (x, y). The search area is a rectangle determined by the start-
and end-points of the proposed baseline alternative.
The data format of the study area is a key point in designing the
algorithm’s formulation. The development of both algorithms must be
structured to suit the accessible data format from the BIM platform for
each element of the study area and decision variables for the intended
credit assessment. The framework’s primary foundation is based on the
infrastructure’s BIM model due to the parametric modeling advantages
of BIM that facilitate the setup of an asset’s decision variables. The smart
objects in this study (i.e., roads and coverage areas) are represented as
GeoJSON geometry types developed by the International Organization
for Standardization (ISO) and the Open Geospatial Consortium (OGC)
(Schmid, Galicz & Reinhardt, 2015). These are the primary data format
used in the BIM-based model, meaning that their connecting points’
coordinates dene each object, as represented in Fig. 6. Thus, the model
consists of objects dened by their coordinates and additional properties
and attributes, as shown in Fig. 6. The BIM-based road smart objects are
LineStrings, a sequence of 2D points considered as “PIs
2
” connected by
line segments (with xed slopes). Hence, the design of the road align-
ments is translated into the proper placement of their X, Y, and Z
coordinates.
Since the nal proposed alignment inherits the attributes of the input
“smart object” (e.g., road width, assembly components, materials,
number of lanes, design speed, and slope), it is important to adjust the
input smart objects’ attributes accordingly. In this study, “planning
roads” were selected as the input “smart object,” representing a 2-lane
road with 40 ft width. The planning road design speed, a customizable
feature, is set as 70 mph, which determines the overall geometry of the
road (e.g., degree of horizontal and vertical curvatures). Other attributes
like the road’s width determine the overlapping area with environ-
mentally protected areas, hence, entirely changing the assessment of a
potential solution. Therefore, changing the input smart-object type and
other adjustable properties of the BIM-based road can affect the poten-
tial output optimum solutions.
3.5.2. Objective function
The second step is dening the objective function and optimization
constraints. Since this study aims to present a framework for sustain-
ability optimization of infrastructure projects by integrating BIM and
sustainability assessment systems, the objective function can be dened
Fig. 2. Model formulation and framework implementation procedures and the expected outputs.
2
Points of Intersection.
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
7
as maximizing the total achievable points from the rating system. Thus,
the objective function can be dened as Eq. (1):
Max Credits =∑
nIP
i=1
aiCrediti(1)
As stated, a horizontal road alignment optimization often focuses on
potentially opposing cost factors that deal with the location and length
of the alignment (Al-Hadad & Mawdesley, 2010; Kang et al., 2012; Maji
& Jha, 2009). Therefore, four of Envision’s quantiable credits were
selected that are, on the one hand, related to the horizontal alignment
decision variables and are also potentially conicting decision objec-
tives. These credits include NW1.2, NW1.3, CR1.1, and CR1.2. The rst
two credits are location-dependent sustainability criteria (NW1.2 and
NW1.3). Suppose the generated alignments avoid affecting these envi-
ronmentally sensitive areas. In that case, they have a better sustain-
ability performance. Up to a certain point dened in the sustainability
assessment system, the further they get from these environmentally
sensitive lands, the higher their tness score. The latter two credits
(CR1.1 and CR1.2) are length-dependent sustainability criteria related
to road alignment design. Thus, the shorter the length of the generated
alignments, the more their tness improves. These selected criteria are
potentially conicting credits in the design of road alignments since
avoiding the environmentally sensitive areas might add to the length of
the alignment and vice versa.
The rules dened in Envision regarding the evaluation of the NW1.2
criterion require the project to provide specic amounts of vegetated or
natural buffers from wetlands, shorelines, and waterbodies. In other
words, the farther the project gets from these sensitive areas, the higher
the project score. The project team is required to provide evidence and
documentation to prove intentional avoidance of these areas. These
rules also apply for NW1.3, which considers the percentage of farmland
avoided or preserved during infrastructure development as the metric
for evaluation. The highest possible level of achievement from these
credits does not fall within the scope of this study. The detailed scenarios
with their scores for determining the level of achievement are presented
in Table 2, which translate into Eqs. (2) and (3). In Infraworks 360, these
environmentally sensitive areas are modeled as coverage areas, a closed
polygonal chain consisting of a nite number of straight-line segments,
shown in Fig. 7.
CrNW1.2= {
2,∑
n
k=1
lk×W<∑
n
i=1
0.9Ai,minD≥50
6,50 <minD≤100
12,100 <minD≤150
16,∑
n
k=1
lk×W<∑
n
i=1
0.9Ai,minD≥150
,D={dj}n
j(2)
Fig. 3. Relevant credits from Envision system to cost components of road alignments (i.e., Natural World and Quality of Life categories).
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
8
CrNW1.3= {
2,∑
n
k=1
lk×W<∑
n
i=1
0.1An
8,∑
n
k=1
lk×W<∑
n
i=1
0.5An
12,∑
n
k=1
lk×W<∑
n
i=1
An
(3)
Where lk is the length of the alignment overlapping with the environ-
mentally sensitive area (k), W is the width of the area covered by the
alignment, Ai is the area of the environmentally sensitive area (i), dj is
the distance of the (jth) segment of the alignment.
The CR1.1 credit intends to reduce material extraction, renement/
manufacture, and transport impacts over the project’s life cycle. The
calculations for determining the achievement level in this credit are
based on the percentage of reduction in the net embodied carbon of
materials over the project’s life compared to the baseline. This can also
be viewed as designing projects to use fewer materials. In the case of
road alignments, this can be interpreted as a reduction in road length
that reduces used materials and, ultimately, reduces overall project
impact. CR1.2 credit intends to reduce the project’s share of contribu-
tion to climate change by reducing greenhouse gas emissions during the
project’s operation stage. When it comes to road alignments, this can be
viewed as a reduction in road length because a longer road means an
increase in vehicle travel distance and thus higher greenhouse gas
(GHG) emission levels. Table 3 presents the detailed performance
evaluation scenarios. Eqs. (4) and (5) demonstrate how the percentage
of length reduction translates into the achievable credit scores.
CrCR1.1= {
5,ln≤0.95 lb
10,0.95 lb<ln≤0.85 lb
15,0.85 lb<ln≤0.8lb
(4)
CrCR1.2= {
8,ln≤0.9lb
13,0.9lb<ln≤0.75 lb
18,0.75 lb<ln≤0.5lb
(5)
Where ln is the length of potential alternative alignment, lb is the length
of the baseline alignment.
Fig. 4. Relevant credits from Envision to cost components of road alignments (i.e., Resource Allocation and Climate and Resilience categories).
Table 1
Analyzed credits from Envision.
metrics intent extent BIM use for evaluation
Natural
World
Siting NW1.1 Preserve
Sites of High
Ecological Value
Avoidance of high
ecological value sites and
establishment of protective
buffer zones.
Avoid placing the project and
temporary works on a site that
has been identied as being of
high ecological value.
Habitat
Expansion
In BIM, tools such as Infraworks avoidance
areas can be set in the way of the development
of the projects. The ability to import data from
GIS enables designers to import maps of sites
with high ecological value and dene needed
buffers to avoid these areas. The modules to
calculate the possible existing overlap enable
partial assessment of this credit. Creating a
database of different types of high ecological
value sites and the associated buffers and
protection plans is recommended.
Climate and
Resilience
Emissions CR1.1 Reduce
Net Embodied
Carbon
Percentage of reduction in
net embodied carbon of
materials.
Reduce the impacts of material
extraction, renement/
manufacture, and transport
over the project life.
50% Quantity take-off features enable designers to
evaluate the materials needed for different
design alternatives, and from there, it is a
straightforward process of calculating baseline
and alternative comparison.
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
9
Additionally, an alternate approach was devised for constructing
performance evaluation scenarios for C.R.1.1 and C.R.1.2. The length
reduction scenarios (C.R.1.1 and C.R.1.2) were converted into length
increment evaluation scenarios, i.e., length increment concerning the
Euclidean distance between the baseline design’s start- and end-points
percentage. This conversion was done due to two reasons: 1. Suppose
a proper baseline alignment is not provided for the assessment to occur,
2. This scenario design has a better restricting effect since the compar-
ison is made to the shortest possible alignment. The alternative scenarios
for length reduction are described in Table 4. Eqs. (6) and (7) are
Fig. 5. The BIM-based Sustainability Assessment Module Output (total tness (score) is the sum of all the achieved scores from four Envision credits represented as
na12, na13, cr11, and cr12).
Fig. 6. A wired representation of the selected search area, representing coordinated data style of objects.
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
10
another way of illustrating how these scenarios translate into achievable
credit scores.
CrCR1.1= {
5,ln≥0.5lu
10,0.5lu<ln≤0.3lu
15,0.3lu<ln≤0.15 lu
20,0.15 lu<ln≤0.05 lu
(6)
CrCR1.2= {
8,ln≥2lu
13,2lu<ln≤0.5lu
18,0.5lu<ln≤0.25 lu
22,0.25 lu<ln≤0.1lu
(7)
where lu is the Euclidean distance between the start- and end-points of
the baseline alignment.
Therefore, the objective function consists of the selected credits from
Envision. It means that the generated candidates are evaluated based on
Table 2
Detailed level of achievement scenarios of credits NW1.2 and NW1.3.
Improved Enhanced Superior Conserving Restorative
Natural
World
Siting NW1.2 Provide
Wetland & Surface
Water Buffers
2 (avoids 90% of
wetland with a 50 ft
buffer)
6 (avoids all wetland
with a 100 ft buffer)
12 (avoids all wetland
with a 150 ft buffer)
16 (avoids 90% of
wetland with a 200 ft
buffer)
20 (restoring
previously disturbed
areas)
NW1.3 Preserve Prime
Farmland
— 2 (less than 10%
disturbance)
8 (less than 5%
disturbance)
12 (100% avoidance) 16 (protection against
future disturbance)
Fig. 7. A coverage area is dened by its connecting points (magenta colored), as is its prefered style through the “select style/color” window.
Table 3
Detailed level of achievement scenarios of credits CR1.1 and CR1.2.
Improved Enhanced Superior Conserving Restorative
Climate and
Resilience
Emissions CR1.1 Reduce Net
Embodied Carbon
5 (At Least 5%
Reduction)
10 (At Least 15%
Reduction)
15 (At Least 30%
Reduction)
20 (At Least 50%
Reduction)
—
CR1.2 Reduce Greenhouse
Gas Emissions
8 (At Least 10%
Reduction)
13 (At Least 25%
Reduction)
18 (At least 50%
Reduction)
22 (100%
Reduction)
26 (Carbon
Negative)
Table 4
Detailed Type 2 length reduction level of achievement scenarios of credits CR1.1 and CR1.2.
Improved Enhanced Superior Conserving Restorative
Climate and
Resilience
Emissions CR1.1 Reduce Net
Embodied Carbon
5 (more than 50%
increment)
10 (less than 50%
increment)
15 (less than 30%
increment)
20 (less than 15%
increment)
—
CR1.2 Reduce Greenhouse
Gas Emissions
8 (more than 100%
increment)
13 (less than 100%
increment)
18 (less than 50%
increment)
22 (less than 25%
increment)
26 (Carbon
Negative)
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
11
their tness (compliance) regarding the credits’ constraints. Thus, the
objective function is a total credit maximization (Crtotal) function
comprising the four credits, NW1.2 (CrNW1.2), NW1.3 (CrNW1.3), CR1.1
(CrCR1.1), and CR1.2 (CrCR1.2) presented in Eq. (9):
Maximize Crtotal =CrNW1.2+CrNW1.3+CrCR1.1+CrCR1.2(8)
3.5.3. Initial population generation
A random initial population of size is generated such that:
•The start- and end-PIs dene the boundary of the search area, which
is determined by the BIM operator (design team) by using the input
road smart object for creating the start- and end-points of the
alignment (Fig. 8);
•All PIs are within the search area’s boundaries so that each gene of a
chromosome is assigned random X and Y coordinate values, as
expressed in Eqs. (9) and (10):
Xmin ≤Xi≤Xmax (9)
Ymin ≤Yi≤Ymax (10)
The rst generation of the alignments is generated based on their X
values to produce candidates with fewer loops, such that (Eq. (11)):
Xi≤Xi+1(11)
3.5.4. Selection and offspring generation
3.6.4.1. Selection. This study chooses GA’s parent chromosomes based
on their tness score for breeding the next generation. Crossover and
mutation are then carried out on the selected parents for producing
offspring. The offspring is then evaluated to nd the ttest individuals
and eliminate the more imperfect solutions for breeding the next gen-
eration by merging them with genetic operators.
3.6.4.2. Crossover. A single-point crossover method was used in this
paper. This approach swaps genes from both parents in a single point
randomly picked to exchange genetic information between them
(Ahmad Al-Hadad, 2011). This procedure forms two offspring, merging
two separate segments from the parents represented in Fig. 9.
3.6.4.3. Mutation. Two mutation methods were used to help with
evolving the performance of the solutions over successive generations.
The mutation operator works on one individual simultaneously, main-
taining diversity at a bit position while ensuring each individual’s
mutability. A standard GA mutation operator is used for randomly
changing the position of a randomly picked PI of an individual within
the boundaries of the search space and the boundaries of the former and
the subsequent PIs’ X coordinates. This will prevent the creation of
unwanted loops in the produced alignments. If Xi is the coordinate value
at the i position before the mutation, the operator assigns the new X
value as (Eq. (12)):
Xp=Xi−1+rnd(Xi+1−Xi−1)(12)
Furthermore, Grouped Point Mutation (GPM) dealt with a group of
sequentially linked PIs (Al-Hadad, 2011), taking a bigger step in the
mutation process while improving the alignments’ smoothness by
replacing the sequential PIs with straight segments.
Fig. 8. Determining the boundary of the search area using planning road smart objects in the BIM platform.
Fig. 9. Crossover.
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
12
3.5.5. Stopping condition
Two elimination criteria are used to stop the search for optimum
solutions. The rst one is reaching up to the total generation number.
The end-user species this criterion, which is supposed to be big enough
for the optimization algorithm to reach a meaningful and acceptable
answer. The second criterion is if the best individual’s tness remains
unimproved for several successive generations. The user also denes the
number of unimproved tness scores before termination.
The BIM-based sustainability algorithm enables designers and
decision-makers to provide optimized sustainability alternatives with a
detailed breakdown of the nal presented solution’s performance con-
cerning the dened credits from the sustainability assessment tool. Also,
a 3D representation of the road alignment is generated. As mentioned,
another advantage of the developed BIM-based optimization prototype
is that the 3D model can automatically inherit certain properties of the
utilized smart object for the optimization problem. This trait eliminates
the need for the operator to redene these properties, including the
horizontal curve radius, minimum transition curve length, materials,
and feature styles. This study selected planning roads to implement the
developed prototype. Other input variables that can be adjusted based
on the users’ preferences include population size, PI number, and total
generation number.
4. Model application
This section tests the proposed framework’s applicability by testing
the developed prototype on a hypothetical case study. The prototype
was scripted using the latest Autodesk InfraWorks - JavaScript API
Documentation 2014 to access the software’s internal data protocols.
The Application Programming Interface (API) documentation is a
technical source that guides how an API should be used and successfully
integrated with the new application. This exhaustive manual (API
documentation) covers all the required information to work with the API
and provides information on functions, classes, forms of returns, argu-
ments, and more. The prototype’s rst development step was integrating
the sustainability assessment criteria rules into the BIM platform to
utilize the assessment algorithm as both the constraints and the opti-
mization algorithm’s objective function. The sustainability assessment
and optimization algorithm scripts are loaded into the scripting console.
For the script to run, the user must select the baseline alignment and the
environmentally sensitive areas imported as coverage areas into the
model. The scripts use the selected objects (i.e., roads and environ-
mentally sensitive areas) as inputs and then analyze the data to produce
outputs in terms of a brief report on the achieved points with a 3D
optimized and modiable BIM road alignment model.
However, sustainability involves various aspects and issues that
entangle the design with critical economic, environmental, social, and
geometric constraints. These constraints bind the road alignment
completely, making the design process much more intricate. The
achievable scores are dependent on the asset’s interaction with its sur-
rounding environment and its improvements compared to the baseline
design alternative. Accordingly, a model consisting of the surrounding
environment and baseline model was created using the Model Builder
feature in the BIM platform. The remaining data that cannot be sourced
originally from the Model Builder feature can be imported into the
model. In this study, the externally sourced data for evaluating credits
NW1.2 and NW1.3 were the environmentally sensitive areas imported
using shapele formats. The dened rules and assessment criteria were
then turned into scripted scenarios that calculate the level of achieve-
ment. Using JavaScript programming language, the scripted scenarios
treat each object as a GeoJSON object and then calculate the location-
dependent credits’ performance level. The achievement calculation
level is based on the distance and overlaps of the road and the coverage
areas. In contrast, in the case of length-dependent credits, the calcula-
tions are based on the suggested model’s percentage of improvement
compared to the baseline model.
4.1. Optimized alignments
Fig. 10 represents the BIM model of the depicted study area with the
dened boundary of the search area by the start- and end-points of the
baseline alignment. The environmentally sensitive areas were imported
in two steps. Two of these areas represent the same type of control area,
i.e., conserved wetlands and one preserved prime farmland. The
coverage area type properties are used for calculations in the script.
The rst obtained optimized alignment with a run time of 15 min was
conducted with 8 PIs and a population size of 300 and 300 generations.
The optimized alignment successfully avoids two environmentally sen-
sitive areas. It inevitably crosses less than 515 m
2
s of one of the wetland
conservation areas, gaining only 2 points from credit NW1.2. The total
achieved score is 54, representing the best possible trade-offs among the
four intended credits that provide an acceptable length reduction while
keeping the required distance from control areas in a smaller window of
opportunity than the former obtained optimized alignment.
The second optimized alignment was conducted through 500 gen-
erations and a population size of 200 with 8 PIs, which took only about
24 min to obtain. This alignment’s tness value is 65, which is achieved
due to avoiding all of the environmentally valuable areas while limiting
the road length to a minimum. Table 5 shows a detailed report of ach-
ieved points from each credit. The BIM operator can modify the output
alignment and reevaluate the modied alignment using the sustain-
ability assessment algorithm to produce another report on the align-
ment’s sustainability performance. The rest of the proposed alignments
and their related models in Infraworks are presented in Table 5 and
Fig. 11.
4.2. Sensitivity analysis
To illustrate the model capabilities and further validation of the
developed prototype, an extensive analysis of sensitivity to key model
parameters is performed. The performed sensitivity analyses of the
proposed model are based on the study conducted by Kang et al. (2012)
and done by excluding or changing the conditioning scenarios and
factors.
4.2.1. Sensitivity to objective function components
The sensitivity analysis to objective function components is intended
to demonstrate the model formulation’s sensitivity to the credits
selected from Envision. It shows these credits’ applicability to road
Fig. 10. Case study area and baseline inputs.
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
13
alignment design optimization and their relevance for constructing a
highway optimization model. Three different scenarios, as presented in
Eqs. (13), (14), and (15) were designed to show each evaluation crite-
ria’s impact on the resulting road alignments. All input parameters used
in the scenarios are identical except for the credits selected to compose
the objective function as follows:
Scenario A
1
Maximize Crtotal =CrCR1.1+CrCR1.2+(13)
Scenario A
2
Maximize Crtotal =CrNW1.2+CrNW1.3(14)
Scenario A
3
Maximize Crtotal =CrNW1.2+CrNW1.3+CrCR1.1+CrCR1.2(15)
The alignment optimized with scenario A
1
, while disregarding the
location-dependent credits in the formulation of the objective function,
as shown in Fig. 12 Case 2-B, has a length of about 8.82 miles that affects
high environmental value areas.
Fig. 12 Case 2-B shows the optimized alignment with location-
dependent credits as the objective function. As shown in Table 6, Case
2-B, with a length of 11 miles, has not affected any environmentally
sensitive areas or buffers. However, the route’s length is about 1.2 miles
longer than in the third case (Fig. 12 Case 2-C), which includes both the
length reduction and high environmental valued site avoidance credits
as the optimization problem’s objective, as shown in Table 6.
4.2.2. Sensitivity to length-dependent scenario design
Due to this study’s effort to integrate Envision with Infraworks to
facilitate sustainability assessment and optimization, the scenarios were
dened precisely based on the evaluation systems instructions for
evaluation. However, each criterion in Envision consists of only 4 or 5
scenarios. Therefore, the alignment length reduction scenarios have
large intervals that undermine the optimization algorithm’s effort to
reduce the path length. If these scenarios are broken down to a greater
extent and placed in more precise and rigorous intervals, such as
Table 7, the path length reduction will improve. It should be noted that
breaking the nal scenarios into smaller intervals had an improved ef-
fect on reducing the length.
Since Envision was considered the base sustainability assessment
tool for implementing the proposed framework, the evaluation scenarios
were formulated strictly based on the system’s dened levels of
achievement evaluation criteria. Because these scenarios’ measurement
sensitivity is regulated over a wide range, they impose a relatively
smaller conditioning effect on the optimization problem. For instance,
the difference in reduction percentage between the superior and
conserving levels of achievement in credit CR1.1 is 20%, which leaves a
wide range for the optimization algorithm in length reduction; thus, the
length reduction was between 30% and 50%, which is a very wide range
and can acquire a high score. Therefore, if the scenarios were broken
down into smaller, more strict intervals, as presented in Table 7. The
optimization algorithm’s effort for length reduction improved based on
the Type 2 length reduction scenario design. Comparing the result
alignments in Fig. 13, Case 3-A (optimized with scenarios originally
dened by Envision) and Fig. 13, Case 3-A (optimized through more
restrictive length reduction scenarios) makes this matter evident. The
results are presented in Table 8.
4.2.3. Sensitivity to PI number
Determining the optimal number of PIs is a key input for the model
implementation in individual projects. As such, this parameter may
affect the accuracy of the solutions. In this section, different scenarios
are designed to demonstrate the proposed model’s sensitivity to the
number of PIs. Also, the model’s runtime may vary by changing this
parameter. To discover the optimal number of PIs in road alignment
Table 5
Detailed report of the BIM-based optimized alignments achieved points from each credit.
Optimized
alignment
PI
number
Length reduction
scenario type
NW1.2 NW1.3 CR1.1 CR1.2 Total
score
Alignment
length
Disturbance
(m2)
Length increment
percentage
runtime
(Seconds)
generations Population
size
Case 1 8 1 2 12 20 22 54 15,202 515.3 13.81% 366.479 300 200
Case 2 8 1 16 12 15 22 65 16,377.8 0 22.62% 1440.74 500 200
Case 3 15 2 16 12 15 22 65 15,750 0 17.92% 4444.458 500 200
Case 4 15 1 16 12 10 18 56 17,604 0 31.80% 3000 300 200
Case 5 8 2 16 12 15 22 65 15,880 0 18.89 1386.213 500 200
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
14
optimization, we ran the model several times with different numbers of
PIs. Fig. 14 shows the optimized alignments with varying sizes of pop-
ulation and PI numbers through different generation numbers (5, 10,
and 15). The detailed results of each case are presented in Table 9.
Since the search area is relatively small and the objective function
consists of only two categories of credit components, no more than 15
PIs were tested. As the results indicate, increasing the number of PIs can
improve the nal obtained alignments with increased population size
and generation size because of the increased search opportunity and the
added precision. It is also observed that a greater number can over-
burden the computation process and increase the runtime while causing
dysfunction.
5. Discussion
This BIM-based end-user application intends to simplify and speed
up the process of nding desirable sustainable infrastructure alterna-
tives. At the same time, it will encourage the implementation of sus-
tainable measures and assist designers in making project-related
Fig. 11. Optimized alignments.
Fig. 12. Sensitivity of optimized alignments to objective function components.
Table 6
Sensitivity analysis results based on objective function components.
Scenario Length
increment
Alignment
length
Score from
CR1.2
Score from
CR1.1
Total
score
A
1
6.63% 14,242 22 20 42
A
2
29.76% 17,331.3 18 15 33
A
3
18.89% 15,880 22 15 37
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
15
decisions. The results from the model application to a hypothetical
project in ve cases (Table 9) indicate that rst, with the increase of PI
numbers and generations, the runtime also increases. Simultaneously,
the optimized alignments also signicantly satisfy higher performance
achievement levels. The sensitivity objective function components
analysis conrms that sustainability evaluation criteria can and should
be jointly optimized for effective optimization. Many trade-off oppor-
tunities exist depending on the exibility and preferences specied with
the input parameters. Furthermore, although it may seem obvious, a rise
in PI numbers can benet from an elevated generation size to yield more
accurate and rened results.
The developed prototype allows for repetitive real-time sustainabil-
ity assessment of infrastructure projects at various design stages without
the use of additional software. Furthermore, it has been shown that the
model can not only effectively optimize highway alignments that satisfy
the chosen sustainability evaluation criteria and provide information
about the resulting alignments to designers but also satisfy basic high-
way design standards. This is a result of the BIM-based optimized model
with its parametric capabilities (e.g., predened in compliance with
road standards and real-time adaptation of territorial conditions).
Additionally, the optimum alignments were obtained without difculty
and within reasonable calculation timeframes. It’s worth noting that
these results resonate with what used to be merely a theory. The model is
expected to apply to many real-world projects (other types of projects)
and perform well in nding the most desirable alternative alignments
during the early stages of road development.
While the framework and prototype are generalizable and custom-
izable, the breadth of the pilot implementation was limited, and several
considerations could be extensively explored in future studies. There-
fore, despite the established strengths, the subsequent model enhance-
ments are desirable:
•Although this model does not cover all of Envision’s credits in the
prototype, it provides a foundation for further development and an
expansion of BIM processes towards sustainability issues. The opti-
mization problem’s objective function comprises only four sustain-
ability criteria from the baseline sustainability evaluation system:
NW1.2, NW1.3, CR1.1, and CR1.2. However, all Envision credits
must be considered to build comprehensively balanced sustainable
alternatives. For instance, adding credit “RA1.5 Balance Earthwork
on Site” to the evaluation criteria can signicantly improve the nal
outcome as it responds to both sustainability considerations and
general design objectives. This is because this credit intends to
minimize transportation and environmental consequences by mini-
mizing the transportation of soils and other excavated items off-site.
Table 7
Design with more restrictive evaluation scenarios for length reduction credits.
Credits Scenario 1 Scenario 2 Scenario 3 Scenario 4 Scenario 5 Scenario 6 Scenario 7 Scenario 8
CLIMATE AND
RESILIENCE
EMISSIONS CR1.1 Reduce Net
Embodied Carbon
2 (more than 50%
length increase)
5 (up to 50%
length increase)
8 (up to 35%
length increase)
11 (up to 25%
length increase)
14 (up to 15%
length increase)
17 (up to 10%
length increase)
20 (up to 5%
length increase)
–
CR1.2 Reduce
Greenhouse Gas
Emissions
8 (more than 100%
increase)
10 (up to 100%
length increase)
12 (up to 65%
length increase)
14 (up to 40%
length increase)
16 (up to 25%
length increase)
18 (up to 20%
length increase)
20 (up to 15%
length increase)
22 (up to 5%
length increase)
Fig. 13. Sensitivity to more restricted scenario design.
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
16
At the same time, the BIM tool can reciprocate through the produc-
tion of realistic and computable terrain representations.
•In parallel to this, to ensure not only sustainability but also cost-
effectiveness, it is desirable to improve the model’s evaluation
criteria for yielding more accurate and realistic outcomes. Thus, in
an upgraded model version, cost-effectiveness should be
simultaneously evaluated and optimized for all the generated high-
way alignments through a trade-off analysis.
•The uniqueness of diverse infrastructure projects necessitates
specialization of the framework’s elements down to the last details of
each of them. This means that specic decision variables of each
infrastructure type determine how each framework element is
employed. For instance, although Envision is available to all forms of
Table 8
Sensitivity analysis results based on scenario design of length reduction credits.
Optimized alignment Alignment length Length reduction type Length increment percentage Runtime (seconds) Generations Population size
Case 3-A 14,244 1 6.64% 24.247 500 200
Case 3-B 13,592 2 1.76% 36.468 500 200
Fig. 14. Sensitivity to PI number.
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
17
infrastructure, the performance evaluation criteria of its credits may
differ for each infrastructure type, or they may not apply at all. While
each element or the detailed process of their use might be altered, the
overall data ow can still follow the proposed framework. For
example, if PSO were selected instead of GA, the optimization pro-
cedure would have changed, but the overall information ow would
still follow Fig. 1.
•The results are limited to Envision as the baseline for performing the
optimization and assessment algorithms. Envision determines the
interlinkage between the sustainability credits and their relative
weight in the current version of the model, which is a key input
parameter affecting the generated highway alignments. It affects the
trade-off between the corresponding objective function components
and, consequently, the outcome. Hence, the balance between the
applicable criteria needs to be in line with the context of the potential
project.
Hence, further extensions of this research might include modeling
other sustainability criteria, cost components, or decision objectives to
increase the optimization’s reliability and improve its nal presentable
solutions.
6. Conclusions
This paper proposes a BIM-based framework for automating sus-
tainability assessment and optimizing infrastructure projects. This
framework was tested through a prototype development used in a case
study to prove its reliability, effectiveness, and efciency. A sensitivity
analysis of the key model parameters was then performed to test the
model’s reliability and Envision’s selected credits concerning an align-
ment optimization problem. The proposed automated framework makes
it possible for designers to investigate and understand the potential
sustainability impact of an infrastructure project in real time during the
design stage. Notably, these projects have wide-ranged sustainability
impacts at a time of expected expansion by almost two and a half times
due to population growth and the increasing importance of sustain-
ability issues.
Additionally, the developed model uses a BIM platform for auto-
mating both the assessment and the optimization algorithm; therefore,
project teams and sustainability assessment agencies can access the BIM
model in a single platform. Hence, team projects can provide the
required sustainability evaluation documentation within a shareable
environment. Also, sustainability assessment agencies can use the BIM-
based assessment algorithm to qualify the BIM-based documentation
provided by the project team and certify infrastructure projects in an
efcient and timely manner.
Moreover, the developed prototype avoids interoperability issues by
creating entirely BIM-based algorithms. The integrated BIM process
classies the BIM object’s attributes and facilitates the extraction of
values demanded by the rating system and thus the sustainability
evaluation of any given infrastructure alternative. Another key contri-
bution is the reduced cognitive load of balancing the complicated trade-
off between conicting infrastructure sustainability performance
criteria through the automated optimization of positive sustainability
impact. Also, the proposed framework can overcome two of the most
critical hindrances to BIM-based infrastructure sustainability improve-
ment—initial and additional costs due to changes in traditional work
processes and sustainability analyses.
In conclusion, this paper demonstrated the plausibility of BIM and
Envision integration for automated sustainability assessment and opti-
mization under certain conditions. The developed prototype uses a BIM
platform with smart objects designed for roads. However, BIM platforms
might not be capable of modeling other types of infrastructure as smart
objects. In that case, the capabilities of BIM should be extended in future
works to include different types of impactful infrastructure and possibly,
enable a systemic and interrelated analysis of interconnected infra-
structure systems.
Declaration of Competing Interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Acknowledgments
This research did not receive any specic grant from funding
agencies in the public, commercial, or not-for-prot sectors.
Supplementary materials
Supplementary material associated with this article can be found, in
the online version, at doi:10.1016/j.scs.2022.104013.
References
Adshead, D., Thacker, S., Fuldauer, L. I., & Hall, J. W. (2019). Delivering on the
sustainable development goals through long-term infrastructure planning. Global
Environmental Change, 59, Article 101975.
AGC, A. (2005). The contractors guide to BIM. URL: http://iweb.agc.org/iweb/Purchase
/ProductDetail.aspx.
Ahmad Al-Hadad, B. (2011). An approach to the highway alignment development process
using genetic algorithm based optimisation. University of Nottingham.
Ainger, C. M., & Fenner, R. A. (2014). Sustainable infrastructure: Principles into practice.
London, UK: ICE publishing.
Al-Hadad, B. A. (2011). An approach to the highway alignment development process using
genetic algorithm based optimisation. University of Nottingham.
Al-Hadad, B. M. A., & Mawdesley, M. (2010). A genetic algorithm approach to a 3D
highway alignment development. IJCCI (ICEC), 129–136.
Ando, S., Arima, T., Bogaki, K., Hasegawa, H., Hoyano, A., & Ikaga, T. (2005).
Architecture for a sustainable future. Tokyo: Architectural Institute of Japan.
Azhar, S., Carlton, W. A., Olsen, D., & Ahmad, I. (2011). Building information modeling
for sustainable design and LEED® rating analysis. Automation in Construction, 20,
217–224.
Basbagill, J., Flager, F., Lepech, M., & Fischer, M. (2013). Application of life-cycle
assessment to early stage building design for reduced embodied environmental
impacts. Building and Environment, 60, 81–92.
Biswas, T., Wang, T., & Sung-Hsien, R., K.. (2008). Integrating sustainable building
rating systems with building information models.
Bongiorno, N., Bosurgi, G., Carbone, F., Pellegrino, O., & Sollazzo, G. (2019).
Potentialities of a highway alignment optimization method in an I-BIM environment.
Periodica Polytechnica Civil Engineering, 63, 352–361.
Bradley, A., Li, H., Lark, R., & Dunn, S. (2016). BIM for infrastructure: An overall review
and constructor perspective. Automation in Construction, 71, 139–152.
Bueno, C., & Fabricio, M. M. (2018). Comparative analysis between a complete LCA
study and results from a BIM-LCA plug-in. Automation in Construction, 90, 188–200.
Table 9
PI sensitivity analysis results.
Optimized alignment PI number Alignment length Total score Runtime (seconds) Generations Population size
Case 1-A 5 17,593 56 2743.360 200 300
Case 1-B 5 15,302.1 56 2555.214 300 500
Case 1-C 5 15,145 56 1596.850 200 500
Case 1-D 10 14,687 46 1807.147 200 300
Case 1-E 10 1609.5 70 2743.360 300 500
Case 1-F 15 16,385 65 3154.568 200 300
Case 1-G 15 15,697.7 65 4345.724 300 500
A. Laali et al.
Sustainable Cities and Society 84 (2022) 104013
18
Carvalho, J. P., Bragança, L., & Mateus, R. (2019). Optimising building sustainability
assessment using BIM. Automation in Construction, 102, 170–182.
Chan, D. W., Olawumi, T. O., & Ho, A. M. (2019). Perceived benets of and barriers to
Building Information Modelling (BIM) implementation in construction: The case of
Hong Kong. Journal of Building Engineering, 25, Article 100764.
Chong, H. Y., Lopez, R., Wang, J., Wang, X., & Zhao, Z. (2016). Comparative analysis on
the adoption and use of BIM in road infrastructure projects. Journal of Management in
Engineering, 32, Article 05016021.
Clevenger, C. M., Ozbek, M. E., & Simpson, S. (2013). Review of sustainability rating
systems used for infrastructure projects. In 49th ASC annual international conference
proceedings (pp. 10–13). Asscociated Schools for Construction.
Cole, J. (2003). Building environmental assessment methods-a measure of success.
Construction. Citeseer.
Costin, A., Adibfar, A., Hu, H., & Chen, S. S. (2018). Building Information Modeling
(BIM) for transportation infrastructure–Literature review, applications, challenges,
and recommendations. Automation in Construction, 94, 257–281.
Diaz-Sarachaga, J. M., Jato-Espino, D., Alsulami, B., & Castro-Fresno, D. (2016).
Evaluation of existing sustainable infrastructure rating systems for their application
in developing countries. Ecological Indicators, 71, 491–502.
Chen, S.Y., Lok, K., & Jeng, T. (2016). Smart BIM objects for design intelligence.
Envision, (2020). Envision [Online]. Available: https://sustainableinfrastructure.org/
[Accessed].
Ferrer, A. L. C., Thom´
e, A. M. T., & Scavarda, A. J. (2018). Sustainable urban
infrastructure: A review. Resources, Conservation and Recycling, 128, 360–372.
French, G. (2012). State of the world report 2012: Sustainable infrastructure. International
Federation of Consulting Engineers (FIDIC).
Gaughan, M. (2012). Assessing sustainability of pipeline projects using envision rating
system. Pipelines 2012: Innovations in design, construction, operations, and maintenance,
doing more with less.
Geyer, P. (2009). Component-oriented decomposition for multidisciplinary design
optimization in building design. Advanced Engineering Informatics, 23, 12–31.
BreGlobal. (2021). Ceequal. In B. Global (Ed.), BRE global launches ceequal version 6 to
support infrastructure projectsteams sustainability goals. ed. Manual available for
purchase at website entitled.
Grifths, K. A., Boyle, C., & Henning, T. F. (2017). In Comparative assessment of
infrastructure sustainability rating tools. Conference: Transportation Research Board 96th
Annual Meeting-Transportation Research Board (p. 01265).
Hartmann, T., Van Meerveld, H., Vossebeld, N., & Adriaanse, A. (2012). Aligning
building information model tools and construction management methods.
Automation in Construction, 22, 605–613.
Hergunsel, M.F. (.2011). Benets of building information modeling for construction
managers and BIM based scheduling.
Hollberg, A., Genova, G., & Habert, G. (2020). Evaluation of BIM-based LCA results for
building design. Automation in Construction, 109, Article 102972.
Hooper, M., & Ekholm, A. (2012). A BIM-Info delivery protocol. Construction Economics
and Building, 12, 39–52.
Hurley, J. (2009). Sustainable or status-quo: Investigating sustainability assessment of
residential estate development. In 4th State of Australian Cities National Conference,
24-27 November 2009 ….
IDB. (2018). What Is Sustainable Infrastructure?. A framework to guide sustainability across
the project cycle. Washington, DC: Inter-American Development Bank. IDB Invest.
Infrastructure, I. F. S., & Infrastructure, Z. P. F. S. (2018). Envision sustainable
infrastructure framework, version 3. Washington, DC: Institute for Sustainable
Infrastructure.
ISI. (2018b). In I. F. S. Infrastructure (Ed.), Envision: Sustainable infrastructure framework
guidance manual (3rd ed.). Washington, DC: Institute for Sustainable Infrastructure.
ISCA. (2021). IS Rating Scheme [Online]. Available: Available at https://www.cer
tiedenergy.com.au/is-rating [Accessed 01/06 2021].
ISI. (2015). Envision rating system for sustainable infrastructure. Washington, DC, USA:
Institute for Sustainable Infrastructure.
ISI. (2018a). Envision. Washington, DC: Institute for Sustainable Infrastructure.
ISI. (2021). Envision [Online]. Available: Https://sustainableinfrastructure.org/resource
/website-tutorials/ [Accessed 06 /01 2021].
Jha, M. K., & Schonfeld, P. (2000). Geographic information system–based analysis of
right-of-way cost for highway optimization. Transportation Research Record, 1719,
241–249.
Ji, Y., Borrmann, A., Beetz, J., & Obergrießer, M. (2013). Exchange of parametric bridge
models using a neutral data format. Journal of Computing in Civil Engineering, 27,
593–606.
Kang, M.-W., Jha, M. K., & Schonfeld, P. (2012). Applicability of highway alignment
optimization models. Transportation Research Part C: Emerging Technologies, 21,
257–286.
Kim, J.-U., Kim, Y.-J., Ok, H., & Yang, S.-H. (2015). A study on the status of infrastructure
BIM and BIM library development. In 2015 International Conference on Computational
Science and Computational Intelligence (CSCI) (pp. 857–858). IEEE.
Liu, Y., Van Nederveen, S., Wu, C., & Hertogh, M. (2018). Sustainable infrastructure
design framework through integration of rating systems and building information
modeling. Advances in Civil Engineering, 2018.
Liu, Z., LU, Y., Shen, M., & Peh, L. C. (2020). Transition from building information
modeling (BIM) to integrated digital delivery (IDD) in sustainable building
management: A knowledge discovery approach based review. Journal of Cleaner
Production, Article 125223.
Lorek, S. (2018). What is bim (building information modeling).
Maji, A., & Jha, M. K. (2009). Multi-objective highway alignment optimization using a
genetic algorithm. Journal of Advanced Transportation, 43, 481–504.
Marzouk, M., & Othman, A. (2020). Planning utility infrastructure requirements for
smart cities using the integration between BIM and GIS. Sustainable Cities and Society,
57, Article 102120.
Mcvoy, G.R., Nelson, D.A., Krekeler, P., Kolb, E., & Gritsavage, J.S. (.2010).Moving
towards Sustainability: New York State Department of Transportation’s GreenLITES
Story. Green streets and highways 2010: An interactive conference on the state of the
art and how to achieve sustainable outcomes,461–479.
Minagawa, M., & Kusayanagi, S. (2015). Study on BIM utilization for design
improvement of infrastructure project. Procedia engineering, 125, 431–437.
Nguyen, A.-T., Reiter, S., & Rigo, P. (2014). A review on simulation-based optimization
methods applied to building performance analysis. Applied Energy, 113, 1043–1058.
Rahmani Asl, M., Zarrinmehr, S., Bergin, M., & Yan, W. (2015). BPOpt: A framework for
BIM-based performance optimization. Energy and Buildings, 108, 401–412.
Rees, W. (2010). What’s blocking sustainability? Human nature, cognition, and denial.
Sustainability: Science, Practice and Policy, 6, 13–25.
Sabatino, S., Frangopol, D. M., & Dong, Y. (2015). Sustainability-informed maintenance
optimization of highway bridges considering multi-attribute utility and risk attitude.
Engineering Structures, 102, 310–321.
Santos, R., Costa, A. A., Silvestre, J. D., & PYL, L. (2020). Development of a BIM-based
environmental and economic life cycle assessment tool. Journal of Cleaner Production,
265, Article 121705.
Saville, C. R., Miller, G. R., & Brumbelow, K. (2016). Using envision to assess the
sustainability of groundwater infrastructure: A case study of the twin oaks aquifer
storage and recovery project. Sustainability, 8, 501.
Schmid, S., Galicz, E., & Reinhardt, W. (2015). WMS performance of selected SQL and
NoSQL databases. In International Conference on Military Technologies (ICMT) 2015
(pp. 1–6). IEEE.
Shahtaheri, Y., Flint, M. M., & De LA Garza, J. M. (2018). Sustainable infrastructure
multi-criteria preference assessment of alternatives for early design. Automation in
Construction, 96, 16–28.
Thangaraj, R., Pant, M., Abraham, A., & Bouvry, P. (2011). Particle swarm optimization:
Hybridization perspectives and experimental illustrations. Applied Mathematics and
Computation, 217, 5208–5226.
Tryggestad, K. (2013). Sustainable buildings and infrastructure: Paths to the future. Taylor &
Francis.
Tsai, C. Y., & Chang, A. S. (2012). Framework for developing construction sustainability
items: The example of highway design. Journal of Cleaner Production, 20, 127–136.
UN. (2015). Transforming our world: The 2030 agenda for sustainable development. New
York: United Nations, Department of Economic and Social Affairs.
USDOT. (2018). Highway Statistics [Online]. Available: Https://www.fhwa.dot.gov/po
licyinformation/statistics/2017/hm260.cfm [Accessed 2021].
van Eldik, M. A., Vahdatikhaki, F., Dos Santos, J. M. O., Visser, M., & Doree, A. (2020).
BIM-based environmental impact assessment for infrastructure design projects.
Automation in Construction, 120, Article 103379.
WCED, (1987). Report of the World Commission on Environment and Development: Our
common future. Accessed Feb, 10.
Willetts, R., Burdon, J., Glass, J., & Frost, M. (2010). Environmental and sustainability
impact assessment of infrastructure in the United Kingdom. Transportation research
record, 2158, 143–150.
Wong, J. K. W., & Zhou, J. (2015). Enhancing environmental sustainability over building
life cycles through green BIM: A review. Automation in Construction, 57, 156–165.
A. Laali et al.
