Journal of Engineering, Project, and Production Management

2022, 12(3), 224-238

Artificial Intelligence in Construction Projects: A Systematic

Scoping Review
Sofie Bang1 and Nils Olsson2
1
Ph.D. Candidate, Department of Mechanical and Industrial Engineering, Norwegian University of Science and
Technology, Richard Birkelands vei 2B, 7034 Trondheim, Norway, E-mail: [email protected] (corresponding author).
2
Professor, Department of Mechanical and Industrial Engineering, Norwegian University of Science and Technology,
Richard Birkelands vei 2B, 7034 Trondheim, Norway, E-mail: [email protected]

Project Management
Received December 7, 2021; revised March 31, 2022; April 19, 2022; accepted May 14, 2022
Available online June 13, 2022
_________________________________________________________________________________________

Abstract: The use of artificial intelligence (AI) in construction projects has surged in recent years and is believed to
represent a significant potential for increasing productivity and efficiency in the industry. The purpose of this paper is to
present a state-of-the-art view of the field by conducting a review of publications concerning the topic of AI in construction
and comparing the findings to previously conducted reviews. This paper provides an overview of the recent and current
uses of AI in construction projects, through a descriptive analysis of the characteristics and contents of 86 peer-reviewed
articles from 2015 to 2020. Although the application of AI in the industry is not entirely new, construction appears to
currently be behind other industries in terms of adopting and adapting to AI. The results show that a wide range of research
is conducted on AI in construction projects. A limited number of publication channels and authors stand behind a significant
part of the reviewed publications. Most studies are conceptual or use a mixed-methods research design. The research
addresses several areas of application, but there is a predominance of quantitatively based subfields of construction, such
as estimation and cost control, logistics, planning, and scheduling. Future research should focus on developing holistic and
process-oriented frameworks for projects to move from ambition to practice. Findings can inform the future development
and implementation of AI in the construction industry context. For researchers, this study identifies areas in need of further
attention and examines possibilities for future exploration of multidisciplinary approaches that combine construction
engineering, project management, AI, and social science. For practitioners, the study highlights current trends and work
within the field, providing an overview of the potential for pilot studies, tests, and innovations.

Keywords: artificial intelligence, construction projects, literature review, scoping review.

Copyright © Journal of Engineering, Project, and Production Management (EPPM-Journal).

DOI 10.32738/JEPPM-2022-0021
_________________________________________________________________________________________

1. Introduction construction project lifecycle chain, ultimately improving

the sustainability of environmental, economic, and social
The construction industry is complex; conflicting factors (Blanco et al., 2018; Oprach et al., 2019). Benefits
objectives contribute to this complexity, as demands for are expected at the project level, the organisational level,
productivity, resource efficiency, sustainability, and and for the industry as a whole. The construction industry
advances in technology continue to develop rapidly (Wood remains a significant contributor to the gross domestic
and Gidado, 2008; Luo et al., 2017). In the past, the product of many countries. However, it also heavily
construction industry has been considered rather contributes to resource usage, energy consumption, and
traditional and, although it is currently experiencing a waste production, and the sector suffers several
digital shift, it remains behind the curve compared to other occupational fatalities every year (Barker et al., 2007;
sectors in implementing AI-based solutions (McKinsey Becqué et al., 2016; Dong et al., 2019). AI is believed to
Global Institute, 2015). Thus, the practical implementation impact how the industry approaches sustainability, policies
of artificial intelligence (AI) in construction is still on health and safety, risk assessment, planning and
considered a rather unexplored topic. scheduling, strategy, project performance, cost control,
The concept of AI is broad, but it can be defined as a and calculations for operations and lifecycles (Hossain and
system or a structure that has ‘the ability to perform tasks Nadeem, 2019).
in complex environments without constant guidance by a AI is a highly interdisciplinary field, comprising
user’ (University of Helsinki, 2018). AI is believed to elements from computer science, logic, mathematics,
enable an increase in productivity throughout the entire
 Artificial Intelligence in Construction Projects: A Systematic Scoping Review 225

psychology, and neuroscience (Tidemann, 2020; Tørresen, 2017; and Darko et al. (2020) conducted a scientometric
2013). In the construction context, AI systems can be analysis of research activities related to the use of AI in the
grouped into four categories: machine learning techniques, architecture, engineering, and construction (AEC) industry.
knowledge-based techniques, evolutionary algorithms, and
hybrid systems (Akinade, 2017). Machine learning This review examines a range of relevant articles
algorithms have the ability to learn from data (Tidemann, published between 2015 and 2020 to provide a state-of-
2019); in the construction industry, neural networks, the-art perspective of the available technology and its
support vector machines, and fuzzy logic seem to be the current areas of application in construction projects.
most widely used machine learning techniques (Akinade, Reviews conducted by Ilter and Dikbas (2009), Martínez
2017). Knowledge-based systems mimic the problem- and Fernández-Rodríguez (2015), Juszczyk (2017), and
solving expertise of humans to identify solutions for Basaif and Alashwal (2018) considered AI applications in
complex problems (Sowa, 2000). Frequently utilised specific areas. Xiao et al. (2018) conducted a bibliometric
knowledge-based approaches include expert systems, rule- review on publications up to 2017. Darko et al. (2020)
based systems, case-based reasoning, and semantic mapped research interests and themes in the AEC industry,
networks (Akinade, 2017). Evolutionary algorithms are identifying topics such as optimisation, simulation, and
based on biological evolution (Russel and Norvig, 2010); decision-making. This study will contribute to the
evolutionary AI techniques optimise factors and possible research field by examining and assessing the body of
scenarios to find the most suitable outcome (Dasgupta and literature dating from 2015 to 2020, focusing on the
Michalewicz, 1997) – such algorithms can cover broad variety of practical applications of AI in construction
territory, from genetic algorithms to ant colony projects. The study targets use cases and applications as
optimisation, particle swarm optimisation, and artificial well as the research activity itself. Ultimately, this study
bee colonies (Akinade, 2017). Hybrid systems combine provides a state-of-the-art overview for reference to future
two or more AI approaches to maximise the strengths and research endeavours, highlighting relevant resources,
potential collaborators, and areas in need of more work.
overcome the weaknesses of individual approaches
(Russel and Norvig, 2010). For practitioners who wish to implement AI-powered
tools in their projects, it provides a sense of direction for
This study investigates the current and potential use of AI-powered innovation, a resource for identifying
AI in construction projects, through a scoping review of 86 potential AI solutions for their problems, and an
articles from peer-reviewed journals. Providing an opportunity to benchmark their work against previous
overview of the available research will indicate which undertakings in the field.
knowledge exists in the field, and where further research is
required. Specifically, the study addresses the following The remainder of the study is organised as follows: the
research questions (RQs): next section explains the methodology of the review
process; the Results section presents and discusses the
 RQ1: What research has been carried out on AI in main findings of the review; the Conclusion section
construction projects? answers the research questions as defined and summarises
the qualitative characteristics of the body of publications,
 RQ2: What research approaches have been used in the research approaches used, and the gaps identified
studies on AI in construction projects? within the field. The last section reflects upon the
 RQ3: What gaps exist in the research? possibilities this study provides for future research, as well
as the limitations of the conducted review.
The first research question will be answered through a
descriptive analysis of the selected publications. For this 2. Method
purpose, the following data will be collected: title; 2.1. Unstructured Literature Search
author(s); year of publication; study location; and
keywords. The second research question will be answered The perceived feasibility of the study was measured
through a more extensive analysis of the research design of against the comprehensiveness of the scoping process,
each study, assessing and classifying the chosen following the recommendations by Levac et al. (2010).
methodology as conceptual, qualitative, quantitative, or This provided the main motivation for an initial,
mixed. Last, the third research question will be answered unstructured literature search. Conducting this initial
by assessing the overall purpose of each study, its focus of search in an explorative manner provided a broad
attention, significant results, and conclusions; this stage knowledge of the field, and ultimately created a foundation
also includes assessing the answers to the two previous for the literature review. The purpose of the preliminary
research questions. search was to produce a literary warrant, thereby
establishing a suitable foundation for further definition and
Several literature reviews on the topic of AI in indexing of terms and classes during the review. The
construction projects have previously been conducted. For search provided an overview of the topic and contributed
example, Ilter and Dikbas (2009) reviewed AI applications to an initial understanding of the development of the field
in construction dispute resolution; Martínez and and related key concepts.
Fernández-Rodríguez (2015) reviewed AI as a tool for
estimating project success and identifying critical success 2.2. Systematic Scoping Review
factors; Juszczyk (2017) reviewed the use of AI for cost To answer the research questions, a scoping review was
estimation in construction projects; Basaif and Alashwal conducted according to the preferred reporting items for
(2018) reviewed AI applications for risk analysis in systematic reviews and meta-analyses (PRISMA)
construction projects; Xiao et al. (2018) conducted a framework (Moher et al., 2009) and the scoping
bibliometric review of AI in construction engineering and methodology framework presented by Arksey and
management, providing an overview of the most O’Malley (2005). Reviews within the field of management
influential studies of AI in construction between 2007 and are often considered to be comprised of a process of

Journal of Engineering, Project, and Production Management, 2022, 12(3), 224-238

226 Bang, S. and Olsson, N.

exploration, discovery, and development (Tranfield et al., publications were reviewed upon request from scholars
2003); therefore, it is desirable to choose a flexible involved in the study.
approach that can be modified throughout the study. The
scoping review enables such a flexible but systematic A set of inclusion and exclusion criteria were defined
approach and comprises five steps: for filtering, to help ensure the relevance and credibility of
the sources for the review. Decisions regarding inclusion
1. Identifying research questions and exclusion criteria remain relatively subjective
(Tranfield et al., 2003); this strengthens the need for a
2. Identifying relevant studies transparent and verifiable process of inclusion and
3. Selection of relevant studies by formulated criteria exclusion. Thus, one criterion used was that the inspected
studies must deal with technology that could be considered
4. Charting the data AI. For example, studies were excluded that simply
5. Collating, summarising, and reporting results discussed challenges of construction projects, or the
construction industry, without any explicit mention of
To clarify and further evolve the framework, Levac et specific solutions. The field and definitions of AI are
al. (2010) present some specific recommendations for each rapidly changing; the availability and accessibility of data
step. For the methodological approach of this review, the and technology are rapidly increasing, while the cost of
recommendations employed included linking the purpose data processing tools is rapidly decreasing. This enables
of the study to the research questions early in the process, applications that were not possible just a few years ago.
in order to facilitate decision-making regarding the Therefore, in order to ensure and capture a state-of-the-art
inclusion and exclusion of relevant publications as the view of the topic, this review only included literature from
scoping review proceeds. The nature of the scoping review 2015 to 2020. Furthermore, the document type was limited
provides for an emergent and iterative process, meaning to include only peer-reviewed articles. As the scoping
that such criteria might not become fully clear until the methodology itself does not include a formal application
later stages of the review (Gough, 2007a). In this review, of quality assessment criteria, strictly including
the inclusion and exclusion criteria as presented produced publications from peer-reviewed sources contributes to an
the final selection of publications. The criteria were implicit quality in the chosen body of publications.
updated throughout the process to sustain the systematic
manner of the review; a more systematic approach helps to The main targets of this analysis were studies of
provide trustworthiness and accountability for the conceptual or practical cases of AI in construction projects;
literature review (Gough, 2007b). however, studies discussing AI in the construction industry
in a more general fashion were also included, as long as the
The next step was to initiate a manual search of selected technology was not explicitly targeted toward
databases. The databases were chosen as they were known infrastructure or industrial construction – such articles
to include significant topics and authors, as identified were excluded. Studies without mention of any specific
through the preliminary search. Additionally, the selected technologies or techniques were also excluded. If a
databases were deemed especially suitable due to their publication discussed a specific technology with an
interdisciplinary nature, and their position as well- explicit functionality but did not name the technology, it
recognised databases for academic articles and was included. Finally, the search was limited to only
publications. The selected databases were Scopus, include publications written in English; any duplicates
ScienceDirect, and Web of Science, each of which were also removed during this process. Following this,
provides an advanced search function that allows the user manual screening of titles, abstracts, and keywords was
to customise their search preferences. Identification and conducted to assess the relevance of the remaining
selection of relevant studies – steps two and three of the publications in the selection; 481 records were screened,
scoping review framework – were structured according to and 374 were excluded. A full-text assessment of the
the PRISMA framework (Moher et al., 2009), as illustrated remaining 107 records was then conducted, to ensure their
in Fig. 1. eligibility and to evaluate the contribution of each study
Tranfield et al. (2003) emphasise the importance of a beyond its title, abstract, and keywords. Twenty-one
well-defined search string in order to create a replicable articles were found to be out of scope, and seven lacked
and transparent search strategy. During the first, sufficient detail to provide an accurate assessment. Eighty-
unstructured search, several search strings were explored. six articles remained to be included in the review.
For example, TITLE-ABS-KEY (construction and 2.3. Classification Framework
artificial intelligence). This search resulted in 60,398 hits
across the three databases. Even after further restrictions, To answer the research questions, several dimensions were
such as year, language, and document types, this search defined along which the selected articles were analysed;
string yielded an unmanageable number of publications. together these constituted the assessment framework and
Moreover, the initial search proved that several terms, provided a foundation for the fourth and fifth steps of the
including expert systems, knowledge engineering, and scoping review framework. The classification framework
even artificial intelligence, seem to lack a single definition was structured to enable a holistic and comprehensive
within the field. Therefore, the final search string needed analysis of the field of AI in the context of construction
to be open enough to include possible variations of such projects and provide a descriptive presentation of the body
words but narrow enough to exclude the most peripheral of publications, according to the recommendations by
subjects. For the scoping search, the string was modified to Arksey and O’Malley (2005). The descriptive features of
TITLE-ABS-KEY (‘construction project*’ AND ‘artificial each publication were collected directly from each
intelligence*’), which resulted in a far more relevant database and included the year of publication, source
selection of publications and 1,608 hits. An additional 21 journal, author(s), location, and keywords. Table 1
describes the classification framework.

Journal of Engineering, Project, and Production Management, 2022, 12(3), 224-238

 Artificial Intelligence in Construction Projects: A Systematic Scoping Review 227

Fig 1. PRISMA flow diagram describing the review process

The publication methodology was classified as either implementation, and sustainability. The contents of the
conceptual, qualitative, quantitative, or mixed-method. publications in each of these categories are further
Some publications did not offer a definitive description of addressed in Section 3.3.
their research methodology; in these cases, the chosen
methodological approach needed to be interpreted from The initial search uncovered countless definitions and
any direct or indirect descriptions provided by the author(s) descriptions of AI-powered technologies and techniques.
themselves. Where the approach of the publication was Thus, the framework defined by Akinade (2017), as
strictly developmental in terms of, for example, a specific described in the introduction, was used for classification
terminology, system, or framework, the methodology was and categorisation: machine learning, knowledge-based
considered to be conceptual. A publication was considered systems, evolutionary algorithms, or hybrid systems. The
to be qualitative if it addressed the subject in a qualitative classification presented in Section 3 was based upon the
manner, such as by discussing certain soft factors description of the techniques provided by the authors
regarding the implementation of AI, its potential or non- themselves and how these compared to the categories in
quantifiable implications, or the effects of its the chosen framework. Where the authors did not provide
implementation. Meanwhile, publications considered a sufficient description of the technique being used or
quantitative addressed the more quantifiable effects of discussed, the technology was labelled N/A.
implementation, or the applications of the tools themselves; 3. Results
use of specific algorithms, for example. Publications were
assessed to be using mixed methods when the research 3.1. Descriptive Analysis
design appeared to use two of the three aforementioned Fig. 2 shows the number of publications in each year for the
methodologies equally. review selection. Although the sample is small, the trend
The categorisation of areas of applications comprised line indicates a steady increase in publications from 2015 to
four steps: 2020; Xiao et al. (2018) noted a gradual increase in
publications on AI in construction up to 2017, and the
1. Identifying common applications increase seems to hold for later years. The years 2016 and
2. Clustering similar applications 2017 appear to show dips in development; during the
review, it was noted that many of the publications from
3. Filtering out rarely mentioned applications 2016 and 2017 were related to infrastructure, roads, and
tunnels. The differences in this sample of publications and
4. Sorting applications by categories the full body of publications from the same years could be
This procedure resulted in nine categories that summarised due to several reasons. One explanation is that the focus
the grouped findings of the literature search: logistics and could have shifted over the years; whereas a certain area of
scheduling, estimation and cost control, health and safety; the industry was more concerned with the use of AI in
project performance and success estimation, strategic earlier years, other areas seem to have experienced an
design, risk management, material properties, reviews, and increased interest in AI as time progressed. Another

Journal of Engineering, Project, and Production Management, 2022, 12(3), 224-238

228 Bang, S. and Olsson, N.

possible explanation lies in the selection of studies provided field, as it provides a suggestion of both where to read and
by the chosen databases – using other databases could where to submit research. There is a clear tendency for the
potentially have yielded additional or different results. conceptual and technically focused studies to be published
in journals such as the Automation in Construction and the
The most frequent publication channels are charted in Journal of Computing in Civil Engineering, whereas
Fig. 3. A significant portion (15%) of the publications were qualitative studies, assessing the potential, barriers, and
published in the Automation in Construction, followed by effects of the implementation of AI are more common in
procedia engineering (8%) and the Journal of Building such journals as the Journal of Civil Engineering and
Engineering (6%). The findings of Xiao et al. (2018) and Management and the Journal of Construction Engineering
Darko et al. (2020) confirm that the Automation in and Management. Certain journals, such as Safety Science
Construction has been the leading publisher in and Energy and Buildings, are more targeted toward
construction-related research in AI for a significant period. specific areas of AI application.
This observation is of interest to anyone involved in the
Table 1. Literature classification framework

Grouping Collected data Purpose

Descriptive features 1.1 Year of publication Describe the characteristics of
1.2 Source the selected articles.

1.3 Author(s)
1.4 Location
1.5 Keywords
Method 2.1 Conceptual Classify the chosen
2.2 Qualitative methodology in the field of
study.
2.3 Quantitative
2.4 Mixed methods
Area of application 3.1 Estimation and cost control Explore the area of application
3.2 Health and safety and utilisation of the
technology or technique at
3.3 Logistics and scheduling issue.
3.4 Material properties
3.5 Project performance and success estimation
3.6 Reviews and implementation
3.7 Risk management
3.8 Strategic design
3.9 Sustainability
Technology 4.1 Machine learning Explore which specific
4.2 Knowledge-based systems technology or technique is
being utilised or discussed.
4.3 Evolutionary algorithms
Based on the framework
4.4 Hybrid systems presented by Akinade (2017).
4.5 N/A

25
21
Number of publications

20 17 17
15
15
10
10
6
5

0
2015 2016 2017 2018 2019 2020

Fig. 2. The number of publications per year

Journal of Engineering, Project, and Production Management, 2022, 12(3), 224-238

 Artificial Intelligence in Construction Projects: A Systematic Scoping Review 229

Fig. 4 shows the most prolific researchers within the experience, interests, and knowledge. In total, 441
field. It appears that a limited number of researchers and keywords were defined, out of which 354 were distinct.
authors are involved in a significant amount of the research However, certain keywords were found to be used more
conducted. frequently (Fig. 6). To provide a better understanding of
keyword frequency, interchangeable keywords were
As Fig. 5 shows, the main contribution to the body of grouped. For instance, ‘artificial intelligence’ and ‘AI’
publications comes from the United Kingdom, followed by were simply grouped into ‘artificial intelligence’, as were
China, Taiwan, the United States, and Australia. This ‘construction project’ and ‘construction projects’, and
could be explained by a higher concentration of ‘building information model’ and ‘building information
researchers within the field in these countries, but it seems modelling’. ‘Artificial intelligence’ appearing as the most
reasonable to assume that this could also be due to the fact frequent keyword seems reasonable, as does ‘construction
that this review only included publications written in project(s)’ as the second most frequent keyword.
English. Other countries could be publishing research ‘Construction management’ and ‘decision support systems’
within the field but in their languages. In total, 21 countries as the third and fourth most often used reflect, to some
were represented. A low representation of countries can extent, the focus of the current research. The high
imply that the field is somewhat immature. However, the frequency of ‘machine learning’ and ‘neural networks’
field appears to be evolving, as additional scientific reflects what appears to be the rather predominant position
environments seem to be emerging. of these techniques.
All keywords, meaning not only author keywords, were
assessed, as author keywords are largely reliant on authors’
14 13
12
Number of publications

10
8 7
6 5
4
4 3 3 3
2 2 2 2 2
2
0

Fig. 3. Most frequent publication channels

Main author Co-author
7
Number of publications

6
5
4
3
2
1
0

Fig. 4. Most frequent authors with two or more publications

Journal of Engineering, Project, and Production Management, 2022, 12(3), 224-238

230 Bang, S. and Olsson, N.

14
12
Number of publications 12 11
10
8 8
8
6 6 6
6
4
4 3 3 3 3 3 3
2 1 1 1 1 1 1 1
0

Fig. 5. Most frequent countries of main authors

20 19
18
16
Number of publications

14
12 11

10
8 7
6 6 6
6 5 5
4
4 3 3 3 3 3 3 3 3

2
0

Fig. 6. Most frequent keywords

3.2. Methods The mixed method is the second most frequently used
methodology (28%). Most of the studies classified as
Since not all the studies explicitly described their mixed methods are rooted in a conceptual base, but in
chosen methodology, it is possible that this data may combination with traditionally qualitative or quantitative
contain errors. To elaborate, certain publications do not methods, for example, the observation of specific case
label their own methods within the framework defined for projects or the use of questionnaires. Purely qualitative
this study. Here, the descriptions of the method provided
studies account for a slightly smaller proportion (21%) of
by the authors of the individual studies are used as the the body of publications. These studies are mainly
reference when categorising each study. There seems to be concerned with the prospects surrounding the technology,
a slight tendency towards a conceptual methodology (40%), which include potential future areas of application,
as Fig. 7 illustrates, which can also be seen in previous possibilities, and barriers to the technology itself, related
literature reviews (Juszczyk, 2017). Most of the studies to soft factors, people, and processes. Very few discuss the
based on a conceptual methodology are concerned with use of AI in the context of people and processes, focusing
developing specific AI-powered tools and techniques. on technology awareness and digital maturity with an
More than half of the conceptual studies include some emphasis on AI. However, this discussion seems to be
quantitative testing and validation in the development of lacking in the studies that discuss more specific solutions
the technique; this is still to be considered part of the and tools. This synthesis is supported by previous reviews,
development process itself and thus accounted for as such as the one undertaken by Basaif and Alashwal (2018),
conceptual. Several of the conceptual studies provide which suggests that a gap exists between the potential that
specific solutions or algorithms tailored toward certain the technology constitutes and the evidence of how it is
areas of application. Most were tested on a proof-of- utilised both in practical and academic contexts. Other
concept scale, and the research does not explicitly state studies compare different techniques and tools in a
whether or not it was developed further or implemented on qualitative frame of reference. Purely quantitative studies
a bigger scale. account for only 12% of the body of publications. These

Journal of Engineering, Project, and Production Management, 2022, 12(3), 224-238

 Artificial Intelligence in Construction Projects: A Systematic Scoping Review 231

studies involve the testing of previously developed an increased degree of transparency and communication,
techniques and algorithms and are usually applied to rather both in the research field as a whole and in individual
limited datasets. This could suggest a relatively low degree studies, would be beneficial.
of research-based AI implementation, constituting a great
potential for future implementation and pilots. 3.3. Areas of Application

Another observation is that to some extent, the number In terms of areas of application, the research seems to be
of studies conducted within each methodological approach relatively evenly distributed, as Fig. 8 shows. There
can be observed to change over the years; this suggests a appears to be a predominance of estimation and cost
shift not only in the focal area, as already mentioned, but control (22%) and logistics, planning, and scheduling
possibly also in the methodological stance. Earlier (19%); the two together account for almost half of the body
publications show a tendency towards mixed or purely of publications. As mentioned, the availability of a
quantitative or qualitative studies, whereas later sufficient quantity and quality of data is a challenge in the
publications are more often purely conceptual. This could construction industry. The two predominant areas both
further suggest a field undergoing change. An increasing lean towards the quantitative and more easily measurable
interest in AI within the construction industry becomes area of the industry; time and money are easily
apparent; this is confirmed both by the body of quantifiable.
publications as a whole and individual studies. However,  A third of the studies categorised under estimation
the high concentration of conceptual studies could suggest and cost control examine the application of AI to cost
a gap between theory and practice. prediction and estimation (Shin, 2015; Juszczyk, 2017;
Elmousalami, 2019; Yaqubi and Salhotra, 2019; Juszczyk
Quantitative, 12% et al., 2019; Juszczyk, 2020). Other applications in the
Conceptual, category include tender price evaluations (Zhang et al.,
40% 2015; Bilal and Oyedele, 2020a; Mehrabani et al., 2020),
cash flow prediction and mapping (Cheng et al., 2015;
Cheng et al., 2020a), and cost-effectiveness analysis
(Wang et al., 2019). Furthermore, publications categorised
Qualitative,
as estimation and cost control include assessment of
21%
profitability (Oyedele et al., 2019), profit margin
estimation (Bilal and Oyedele, 2020b), and prediction of
project award price (Chou et al., 2015). Similarly, studies
explore the selection of optimal construction bid price
(Aboelmagd, 2018), the setting of baseline rates
Mixed, (Shahtaheri et al., 2015), and the calculation of the
28% construction site cost index (Juszczyk and Leśniak, 2019).
 The category of logistics, planning, and scheduling
includes publications discussing applications of AI to
Fig. 7. Distribution of chosen methodology
improve construction project schedules (de Soto et al.,
Many studies appear to remain in a development phase 2017), estimation of construction project schedules (Cheng
while very few address the practical adoption of AI-based and Hoang, 2018; Cheng et al., 2020b) progress
technology in the industry and among practitioners at a monitoring (Golparvar-Fard et al., 2015), and prediction of
larger scale. To elaborate, most studies illustrate how risk delay (Yaseen et al., 2020). Other studies discuss the
certain technology can be utilised in different parts of topic of clash relevance prediction (Hu and Castro-
construction projects, for example exploring site layout Lacouture, 2019), resolving design clashes (Hsu et al.,
design (Amiri et al., 2017), or predicting project 2020), and validation of change requests (Dawood et al.,
performance (Mirahadi and Zayed, 2016). However, the 2019). Publications focused on logistics include the
majority of studies lack a larger context for the technology utilisation of AI in resource management (Xing et al., 2016;
– a framework for the technology to operate within. The Podolski, 2016; Camacho et al., 2018), resource-
studies do not discuss organisational or process-oriented constrained scheduling (Li and Womer, 2015; Zheng and
considerations in the adaption and adoption of AI in Wang, 2015), and the resource-levelling optimisation (Iyer
projects. This could, naturally, have many explanations. et al., 2015). For the physical construction site, material
For example, a few studies discuss the lack of access to layout planning (Cheng and Chang, 2019), and site layout
sufficient amounts of quality data. Another possible design (Amiri et al., 2017) are explored.
explanation could lie in the lack of transferability in the
developed models and frameworks, meaning that new  In the category of strategy strategic matters such as
studies are not necessarily able to build on previous project selection (Mousavi et al., 2015; Fallahpour et al.,
research. This could suggest a need for a more standardised 2020), contractor pre-qualification (Kog and Yaman,
framework of technologies and terminology for 2016), and strategic supply chain management & supplier
researchers to operate within when exploring the topic of selection (Taherdoost and Brard, 2019) are examined.
AI in construction. Challenges concerning transferability More specific endeavours are also found, in publications
could ultimately prevent a model built in one environment studying the utilisation of AI in relating organisational
from being useful in another environment, due to characteristics and project delivery methods (Gazder et al.,
differences in requirements and prerequisites; it could also 2018) and enhancing communication between actors
prevent one researcher from effectively building upon the (Khosrowshahi, 2015). Appraisal of decision support
work of another. There is no simple solution to such a systems for modularisation (Sharafi et al., 2018) and
complex problem, but it seems reasonable to assume that prefabrication (Arashpour et al., 2017; Li et al., 2018; Zhou
and Ren 2020) are also seen from a strategic perspective.

Journal of Engineering, Project, and Production Management, 2022, 12(3), 224-238

232 Bang, S. and Olsson, N.

 In the category of health and safety (10%) all studies Notably, even if a lot of the studies address a certain
explore AI utilisation in safety, while two focus specifically area of application conceptually or in general terms,
on the interaction between health and safety (Ayhan and relatively few studies report on actual implementation and
Tokdemir, 2018; Nnaji and Karakhan, 2020). Safety practical use beyond pilots and proofs-of-concept. Most
applications include the identification of factors indicating focus on the potential use or the development of techniques
and influencing safety on the construction site (Poh et al., for future use. No significant links were found in the body
2018; Goh et al., 2018; Xu et al., 2020, Han et al., 2020), of publications between the chosen areas of application
safety planning of temporary structures (Kim et al., 2018), and the chosen methodologies.
planning of safe construction site layouts (Ning et al., 2018)
An overwhelming majority of the studies examine the
and safety assessment (Ayhan and Tokdemir, 2019).
use of AI first and foremost as a decision support tool,
 Publications examining project performance and implying that the human decision-maker is still seen as an
success estimation (10%) are generally targeted toward essential part of the project, the project processes and
project management, the majority focusing on decision activities; this could suggest a low degree of maturity in
support for the project manager, or the discipline and the implementation of AI in the industry.
process of project management itself (Hajdasz, 2015; 3.4. Technology
Gudauskas et al., 2015; Hanna et al., 2018; Mahfouz et al.,
2018; Vickranth et al., 2019). Other studies focus on Fig. 9 shows the distribution of technology discussed
predicting and optimizing project performance, time, and in the publications, based on the authors’ own descriptions,
cost (Mirahadi and Zayed, 2016; Jaber et al., 2019) or categorised by the framework presented by Akinade
project evaluation (Erzaij et al., 2020). (2017); the distribution shows a clear tendency. More than
a third of the publications (38%) do not explicitly state the
 Topics related to risk management (8%) include risk nature or class of the technology in question. Some
analysis (Pruvost and Scherer, 2017; Basaif et al., 2020), risk explanations for this were identified during the search.
assessment (Samantra et al., 2017) and risk prediction (Zou Studies lacking a technical description seem to mainly
et al., 2017). Other publications categorised as risk focus on implications and effects, or potentials and barriers,
management include studies examining the identification of rather than the development or use of specific technologies.
critical risks in projects (Qazi et al., 2016), forecasting of Hybrid systems (26%) and machine learning (26%) were
project status based on threats-opportunities and strength- the main techniques studied in more than half of the
weaknesses (Boughaba and Bouabaz, 2020), and publications. Knowledge-based systems constituted 6% of
construction site accident classification (Cheng et al., 2020c). the reviewed studies, despite case-based reasoning, a type
 One group of articles provides an overview of the of knowledge-based system (Akinade, 2017) being
current situation in the construction industry and maps identified as one of the most frequently used tools in
possibilities, barriers, and implications within the field dispute resolution in projects (Ilter and Dikbas, 2009).
through reviewing the existing body of publications. However, the limited use of case-based reasoning is also
Identified reviews explore the use of relevant technology seen in previous reviews (Xiao et al., 2018). Similarly,
in construction projects: machine learning (Hong et al. evolutionary algorithms only constituted 2% of the studies,
2020), deep learning (Akinosho et al., 2020) and despite previously being identified as one of the most
automation (Faghihi et al., 2015). Eber (2020) investigates frequently used tools in AEC (Darko et al., 2020).
the potential of AI, and Delgado et al. (2019) investigate However, Darko et al. (2020) suggest that genetic
industry-specific challenges in the implementation of AI; algorithms might be more widely utilised as a part of
both in the context of the construction industry. Chen et al. hybrid systems. Akinade (2017) suggests that the strength
(2015) investigate the use of BIM in conjunction with AI. of hybrid systems lies in their capability to overcome
weaknesses related to single AI techniques or algorithms,
 Only one of the publications assessed in the category which makes them a useful option in complex and dynamic
of sustainability (7%) is concerned with social sustainability, construction projects. The majority of the hybrid-classed
specifically dispute resolution (Elziny et al., 2016). The studies describing technology and techniques also utilised
remaining studies mainly explore environmental machine learning, mostly supervised machine learning; a
sustainability, while a few are centred around sustainability notable number were also based on evolutionary
in broader terms. These publications examine design algorithms. Among the publications discussing machine
optimization for sustainability (Liu et al., 2015; Rodriguez- learning, half specifically discussed neural networks. The
Trejo et al., 2017), assessing and classifying sustainability in frequent use of neural networks is also confirmed in
a project (Akbari et al., 2018), or waste reduction previous reviews (Ilter and Dikbas, 2009; Martínez and
(Banihashemi et al., 2017; Bilal et al., 2019). Fernández-Rodríguez, 2015; Juszczyk, 2017; Darko et al.,
2020). The remainder of the publications showed no
 Publications categorised in materials properties
significant trend or preferred technique within the
(5%) are related to the quantitative assessment of
category. There appears to be an increase in the
construction materials, predicting properties of concrete
application of hybrid models in the later years compared to
(Vakhshouri and Nejadi, 2015; Zhang et al., 2020),
the earlier years (Xiao et al., 2018). This could suggest
specific construction elements (Qi et al., 2018) and using
increased use of more compound systems as technology
remote electron microscope technology to monitor the
and industry development because hybrid systems are able
composition of materials (Xu et al., 2020).
to solve more complex tasks than any single system
(Akinade, 2017).

Journal of Engineering, Project, and Production Management, 2022, 12(3), 224-238

 Artificial Intelligence in Construction Projects: A Systematic Scoping Review 233

Materials properties; 5%

Sustainability , 7%
Estimation and cost
Review and overview , 7% control , 22%

Risk management , 8%

Logistics, planning and

Project performance and scheduling , 19%
success estimation , 10%

Health and safety , 10%

Strategy , 12%

Fig. 8. Distribution of areas of application

Hybrid; 26 % Knowledge based systems; 6 %

Evolutionary algorithms; 2 % Others; 50 %

Machine
learning; 26 %

N/A; 38 % Neural networks; 50 %

Fig. 9. Distribution of discussed technology

As part of the screening process, studies using the term The findings of this study indicate a versatile body of
AI without addressing specific techniques or technology literature, with a few characteristics that stand out. There
were discarded. This was warranted for a significant seems to be a steady increase in the number of publications
number of studies, which implies that many authors use from the years 2015 to 2020. Three journals, Automation
‘AI’ somewhat loosely; the same can be said for machine in Construction, Procedia Engineering, and the Journal of
learning. One explanation could be a lack of Building Engineering amount account for a quarter of the
unambiguously defined terminology and vocabulary in the publications, with the rest of the articles being distributed
academic field, especially in the context of the construction evenly among the remaining 48 publication channels. A
industry. Another explanation could lie in the fact that limited number of authors produced the majority of the
these are ‘buzz words’, popularised by the media; this can publications; correspondingly, a limited number of
contribute to the confusion of definitions. This observation countries are also far more prominent than others.
is to some extent validated by the high number of
exclusions required during the screening process (Fig. 1); The preferred approaches in the field have changed
a majority of the exclusions were caused by the high during the last few years, indicating a rapidly developing
number of papers discussing technology not explicitly field. Studies are often descriptive in nature, due to the lack
defined as AI.4. conclusion of empirical evidence. Purely conceptual studies constitute
almost half of the reviewed publications, suggesting a
This paper contributes to the current state of research theoretical foundation, but a lack of practical
on AI in construction projects by presenting a state-of-the- implementation beyond small-scale testing and proofs-of-
art view of the field of AI in construction. For researchers, concept. This can be taken as a sign that the field itself
it provides an overview of the most influential publication remains at an emergent stage, but at the same time, this
channels, authors, methodologies used, and areas of provides an understanding of the great potential the field
application, ultimately providing a direction for future demonstrates. Existing case-based research can and should
research. For practitioners, it illustrates possible areas of be used as a foundation for larger-scale studies.
innovation and application of AI-powered techniques and
serves as a tool for benchmarking. The field is rapidly evolving, together with new
technologies, techniques, and tools being developed both

Journal of Engineering, Project, and Production Management, 2022, 12(3), 224-238

234 Bang, S. and Olsson, N.

in and out of the construction context. A visible change in necessarily conclusively representative of the field of AI in
preferred methods, as well as a change in keywords over construction projects. Another possible limitation is the
time, implies that the field is indeed developing. The organisation of the search as a manual search of chosen
conceptual methodology seems to be the preferred databases. This may have led to some relevant studies
approach in the field of study. The extensive use of being missed, thereby possibly under-estimating or
conceptual methodology suggests that this method works wrongly assessing the extent of research regarding AI in
in a research context but could at the same time suggest a construction projects.
need for other, more practically focused methods to further
develop the field. The wide thematic range of previous Furthermore, limitations are associated specifically
studies has provided a valuable foundation for future with the scoping review methodology itself: the scoping
research, but the field is assumed to benefit from a shift review does not formally evaluate the quality of the
toward more interdisciplinary based studies. Among the publications reviewed and relies on the implicit quality of
barriers to practical implementation is the lack of sufficient the publication sources. The descriptive nature of the
quantity and quality of data, as well as transferability methodology can result in broader, less defined searches;
among developed models and frameworks. This could be however, it also ensures flexibility and resilience in the
due to the immaturity of certain technologies within the study and allows for more rapid mapping which is
industry context, posing problems in practical beneficial for an expanding research field.
implementation, testing and surveying. This is supported References
by the findings, specifically the limited extent of big-scale
experimental and practical implementations. It has been Aboelmagd, Y. M. R. (2018). Decision support system for
shown that AI has been applied to several areas, and the selecting optimal construction bid price. Alexandria
body of evidence is relatively evenly distributed Engineering Journal, 57(4), 4189-4205. doi:
thematically, with a slight predominance of more 10.1016/j.aej.2018.11.007
quantitatively focused areas of application. This highlights Akbari, S., Khanzadi, M., and Gholamian, M. (2018).
the need for a degree of standardisation and structure in the Building a rough sets-based prediction model for
field, allowing researchers to assess and compute both the classifying large-scale construction projects based on
qualitative and quantitative areas of the industry. sustainable success index. Engineering, Construction
Standardisation of collected data, process-oriented and Architectural Management, 25(4), 00-00. doi:
frameworks, industry wide definitions of terminology and 10.1108/ECAM-05-2016-0110
technology are believed to enable a greater degree of Akinade, O. O., (2017). BIM-based software for
transparency and interdisciplinary collaboration in the construction waste analytics using artificial
field, ultimately contributing to the research field evolving. intelligence hybrid models. Doctoral dissertation.
The identified research appears to have focused mostly on University of the West of England.
the technology itself, and less on the context the Akinosho, T. D., Oyedele, L. O., Bilal, M., Ajayi, A. O.,
technology would be operating within; this suggests that Delgado, M. D., Akinade, O. O., and Ahmed, A. A.
the field could benefit from an increased focus on (2020). Deep learning in the construction industry: A
organisational and process-oriented research in the context review of present status and future innovations. Journal
of AI and construction of Building Engineering. 32, 101827. doi:
10.1016/j.jobe.2020.101827
5. Limitations and Future Research Amiri, R., Sardroud, J. M., and de Soto, B. G. (2017). BIM-
For future research, this study provides a sense of direction based Applications of Metaheuristic Algorithms to
and highlights where current gaps in the research are to be Support the Decision-making Process: Uses in the
found. It becomes apparent that AI holds significant Planning of Construction Site Layout. Procedia
potential for increased productivity and sustainability in Engineering. 196, 558-564. doi:
construction projects, but the construction industry seems 10.1016/j.proeng.2017.08.030
to lack the progress seen in other industries. For the future Arashpour M., Bai Y., Aranda-mena G., Bab-Hadiashar A.,
of the field, transparency and explicit definitions across all Hosseini R., and Kalutara P. (2017). Optimizing
sub-fields will be of particular importance for the field as decisions in advanced manufacturing of prefabricated
a whole to mature and develop – and to a greater extent to products: Theorizing supply chain configurations in
ensure comparability and transferability among studies and off-site construction. Automation in Construction,84,
findings. The research currently lacks empirical data and 146-153. doi: 10.1016/j.autcon.2017.08.032
research on implementation and performance beyond Arksey, H. and O’Malley, L. (2005) Scoping studies:
small-scale testing and proof-of-concepts. Research towards a methodological framework. International
Journal of Social Research Methodology, 8(1), 19-32.
mapping the effects of the increased use of AI also seems
to be lacking. Pilots and testing are important first steps in doi: 10.1080/1364557032000119616
a developing field; however, in order to truly change Ayhan, B. U. and Tokdemir, O. (2018). Predicting the
deliveries and deliverables through the use of AI future outcome of construction incidents. Safety Science, 113,
research must focus on developing holistic frameworks for 91-104. doi: 10.1016/j.ssci.2018.11.001
projects to move from ambition to practice. Ayhan, B. U. and Tokdemir, O. (2019). Safety assessment
in megaprojects using artificial intelligence. Safety
A few limitations can be associated with this study. Science, 118, 273-287. doi: 10.1016/j.ssci.2019.05.027
First, the research may be limited by deficiencies in data Banihashemi, S., Tabadkani, A., and Hosseini, M. R.
collection and analysis, as a limited number of sources (2017). Modular Coordination-based Generative
were reviewed. Second, limitations could be associated Algorithm to Optimize Construction Waste. Procedia
with the chosen framework for the review. For example, Engineering, 180, 631-639. doi:
only articles written in English were included in the final 10.1016/j.proeng.2017.04.222
sample; therefore, the chosen publications are not

Journal of Engineering, Project, and Production Management, 2022, 12(3), 224-238

 Artificial Intelligence in Construction Projects: A Systematic Scoping Review 235

Barker, T., Bashmakov I., Alharthi, A., Amann, M., Cheng, M. Y., Kusoemo, D., and Gosno, R. (2020a). Text
Cifuentes, L., Drexhage, J., Duan, M., Edenhofer, O., mining-based construction site accident classification
Flannery, B., Grubb, M., Hoogwijk, M., Ibitoye, F. I., using hybrid supervised machine learning. Automation
Jepma, C. J., Pizer, W. A., and Yamaji, K. (2007). in Construction, 118(1), 103265.
Mitigation from a cross-sectoral perspective. 10.1016/j.autcon.2020.103265
Proceedings of the Climate Change 2007: Mitigation. Cheng, M. Y., Cao, M. T., and Herianto, J. G. (2020b).
Contribution of Working Group III to the Fourth Symbiotic organisms search-optimized deep learning
Assessment Report of the Intergovernmental Panel on technique for mapping construction cash flow
Climate Change, Cambridge University Press, considering complexity of project. Chaos, Solitons &
Cambridge, United Kingdom and New York, NY, Fractals, 138, 109869. doi:
USA. 10.1016/j.chaos.2020.109869
Basaif, A. A., Al-Ashwal, A. M., Rahim, F. A., Karim, S. Cheng, M. Y., Chang, Y.-H., and Korir, D. (2020c). Novel
B., and Loo, S. C. (2020). Technology awareness of Approach to Estimating Schedule to Completion in
artificial intelligence (AI) application for risk analysis Construction Projects Using Sequence and
in construction projects. Malaysian Construction Nonsequence Learning. Journal of Construction
Research Journal, 1, 182-195. Engineering and Management, 145(11), 04019072. doi:
Basaif, A. A. and Alashwal, A. M. (2018). A review of the 10.1061/(ASCE)CO.1943-7862.0001697
application of artificial intelligence for risk analysis in Cheng, M. Y. and Hoang, N. D. (2018). Estimating
construction projects. Book of Abstracts: ASEAN Post Construction Duration of Diaphragm Wall Using
Graduate Conference: Inclusive Built Environment Firefly-Tuned Least Squares Support Vector Machine.
Towards Realising New Urban Agenda, Kuala Lumpur, Neural Computing and Applications, 30, 2489-2497.
Malaysia, 9-9. doi: 10.1007/s00521-017-2840-z
Becqué, R., Mackres, E., Layke, J., Aden, N., Liu, S., Cheng, M. Y., Hoang, N.-D., and Wu, Y.-W. (2015). Cash
Managan, K., Nesler, C., Mazur-Stommen, S., Flow Prediction for Construction Project Using a
Petrichenko, K., and Graham, P. (2016). Accelerating Novel Adaptive Time-Dependent Least Squares
Building Efficiency. Eight Actions for Urban Leaders. Support Vector Machine Inference Model. Journal of
World Resources Institute. WRI Ross Center for Civil Engineering and Management, 21(6), 679-688.
Sustainable Cities. doi: 10.3846/13923730.2014.893906.
Bilal, M. and Oyedele, L. O. (2020a). Big Data with deep Chou, J., Lin, C. W., Pham, A.-D., and Shao, J.-Y. (2015).
learning for benchmarking profitability performance in Optimized artificial intelligence models for predicting
project tendering. Expert Systems with Applications, project award price. Automation in Construction, 54,
147, 113194. doi: 10.1016/j.eswa.2020.113194 106-115. doi: 10.1016/j.autcon.2015.02.006
Bilal, M. and Oyedele, L. O. (2020b). Guidelines for Darko, A., Chan, A. P. C., Adabre, M. A., Edwards, D. J.,
applied machine learning in construction industry—A Hosseini, M. R., and Ameyaw, E. E. (2020). Artificial
case of profit margins estimation. Advanced intelligence in the AEC industry: Scientometric
Engineering Informatics, 43(C), 101013. doi: analysis and visualization of research activities.
10.1016/j.aei.2019.101013 Automation in Construction, 122, 103081. doi:
Bilal, M., Oyedele, L. O., Akinade, O. O., Delgado, J. M. 10.1016/j.autcon.2020.103081
D., Akanbi, L. A., Ajayi, A. O., and Younis, M. S. Dasgupta, D. and Michalewicz, Z. (1997). Evolutionary
(2019). Design optimisation using convex Algorithms in Engineering Applications. Springer,
programming: Towards waste-efficient building New York.
designs. Journal of Building Engineering, 23, 231-240. Dawood, H., Siddle, J., and Dawood, N. (2019).
doi: 10.1016/j.jobe.2019.01.022 Integrating IFC and NLP for automating change
Blanco, J. L., Fuchs, S., Parsons, M., and Ribeirinho, M. J. request validations. Journal of Information Technology
(2018). Artificial intelligence: Construction in Construction. 24, special issue, 540-552. doi:
technology’s next frontier. The Building Economist, 7- 10.36680/j.itcon.2019.030
13. doi: 10.3316/informit.048712291685521 de Soto, B. G., Rosarious, A., Rieger, J., Chen, Q., and
Boughaba, A. and Bouabaz, M. (2020). Identification and Adey, B. T. (2017). Using a Tabu-search Algorithm
risk management related to construction projects. and 4D Models to Improve Construction Project
Advances in Computational Design, 5(4), 445-465. doi: Schedules. Procedia Engineering, 196, 698-705. doi:
10.12989/acd.2020.5.4.445 10.1016/j.proeng.2017.07.236
Camacho, A., Canizares, P.C., Estevez, S., and Nunez, M. Delgado, J. M. D., Oyedele, L., Ajayi, A., Akanbi, L.,
(2018). A tool-supported framework for work planning Akinade, O., Bilal, M., and Owolabi, H. (2019).
on construction sites based on constraint programming. Robotics and automated systems in construction:
Automation in Construction, 2(86), 190-198. doi: Understanding industry-specific challenges for
10.1016/j.autcon.2017.11.008 adoption. Journal of Building Engineering, 26, 100868.
Chen K., Lu W., Peng Y., Rowlinson S., and Huang G. Q. doi: 10.1016/j.jobe.2019.100868
(2015). Bridging BIM and building: From a literature Dong, X. S., Jackson, R., Varda, D., Betit, E., and Bunting,
review to an integrated conceptual framework. J. (2019). Trends of Fall Injuries and Prevention in the
International Journal of Project Management, 33(6), Construction Industry. The Center for Construction
1405-1416. doi: 10.1016/j.ijproman.2015.03.006 Research and Training (CPWR). Data report.
Cheng, M. Y. and Chang, N. W. (2019). Dynamic Eber, W. (2020). Potentials of artificial intelligence in
construction material layout planning optimization construction management. Organization, Technology
model by integrating 4D BIM. Engineering with and Management in Construction: An International
Computers, 35(3), 703-720. doi: 10.1007/s00366-018- Journal, 12, 2053-2063. doi: 10.2478/otmcj-2020-
0628-0 0002

Journal of Engineering, Project, and Production Management, 2022, 12(3), 224-238

236 Bang, S. and Olsson, N.

Elmousalami, H. (2019). Intelligent methodology for Canadian Journal of Civil Engineering, 45, 1-8. doi:
project conceptual cost prediction. Heliyon, 5(5), 10.1139/cjce-2017-0237
e01625. doi: 10.1016/j.heliyon.2019.e01625 Hong, T., Wang, Z., Luo, X., and Zhang, W. (2020). State-
Elziny, A. A., Mohamadien, M. A., Ibrahim, H., and Fattah, of-the-Art on Research and Applications of Machine
M. K. (2016). An expert system to manage dispute Learning in the Building Life Cycle. Energy and
resolutions in construction projects in Egypt. Ain Buildings212(3). doi: 10.1016/j.enbuild.2020.109831
Shams Engineering Journal, 7(1), 57-71. doi: Hossain, M. A. and Nadeem, A. (2019). Towards
10.1016/j.asej.2015.05.002 digitizing the construction industry: State of the art of
Erzaij, K. R., Rashid, H. A., Naji, H. I., and Ali, R. H. Construction 4.0. Proceedings of the International
(2020). Projects evaluation in construction industry. Structural Engineering and Construction, 6(1). doi:
Periodicals of Engineering and Natural Sciences, 8(3), 10.14455/ISEC.res.2019.184
1808-1816. Hsu, H. C., Chang, S., Chen, C. C., and Wu, I C. (2020).
Faghihi, V., Nejat, A., Reinschmidt, K., and Kang, J. Knowledge-based system for resolving design clashes
(2015). Automation in construction scheduling: a in building information models. Automation in
review of the literature. The International Journal of Construction, 110. doi: 10.1016/j.autcon.2019.103001
Advanced Manufacturing Technology, 81(9), 1845- Hu, Y. and Castro-Lacouture, D. (2019). Clash Relevance
1856. doi: 10.1007/s00170-015-7339-0 Prediction Based on Machine Learning. Journal of
Fallahpour, A., Wong, K., Rajoo, S., Olugu, E., Nilashi, M., Computing in Civil Engineering, 33(2), 04018060. doi:
and Turskis, Z. (2020). A fuzzy decision support 10.1061/(ASCE)CP.1943-5487.0000810
system for sustainable construction project selection: Ilter, D. and Dikbas, A. (2009). A review of the artificial
an integrated FPP-FIS model. Journal of Civil intelligence applications in construction dispute
Engineering and Management, 26(3), 247-258. doi: resolution. Proceedings of the Managing IT in
10.3846/jcem.2020.12183. Construction. 26th International Conference, 41-50.
Gazder, U., Shakshuki, E., Adnan, M., and Yasar, A.-U.- doi: 10.1201/9781482266665-63
H. (2018). Artificial Neural Network Model to relate Iyer, P., Liu, Y., Sadeghpour, F., and Brennan, R. W.
Organization Characteristics and Construction Project (2015). A Fuzzy-Logic based Resource Levelling
Delivery Methods. Procedia Computer Science, 134, Optimisation Tool. IFAC-Papers OnLine, 48(3), 1942-
59-66. doi: 10.1016/j.procs.2018.07.144 1947. doi: 10.1016/J.IFACOL.2015.06.371
Goh, Y. M., Ubeynarayana C.U., Wong K. L. X., and Guo Jaber, F., Al-Zwainy, F. M., and Hachem, S. (2019).
B. H. W. (2018). Factors influencing unsafe behaviors: Optimizing of predictive performance for construction
A supervised learning approach. Accident Analysis & projects utilizing support vector machine technique.
Prevention, 188, 77-85. doi: Cogent Engineering, 6(1), 1685860. doi:
10.1016/j.aap.2018.06.002 10.1080/23311916.2019.1685860
Golparvar-Fard, M., Peña-Mora, F., and Savarese, S. Juszczyk, M. (2017). The Challenges of Nonparametric
(2015). Automated Progress Monitoring Using Cost Estimation of Construction Works with the use of
Unordered Daily Construction Photographs and IFC- Artificial Intelligence Tools. Procedia Engineering,
Based Building Information Models. Journal of 196 (1), 415-422. doi: 10.1016/j.proeng.2017.07.218
Computing in Civil Engineering, 29(1), 04014025. doi: Juszczyk, M. (2020). Development of Cost Estimation
10.1061/(ASCE)CP.1943-5487.0000205 Models Based on ANN Ensembles and the SVM
Gough, D. (2007a). Weight of Evidence: a framework for Method. Civil and Environmental Engineering
the appraisal of the quality and relevance of evidence. Reports,30(3), 48-67. doi: 10.2478/ceer-2020-0033
Research Papers in Education, 22(2), 213-228. doi: Juszczyk, M. and Lesniak, A. (2019). Modelling
10.1080/02671520701296189 Construction Site Cost Index Based on Neural Network
Gough, D. (2007b). Giving voice: Evidence-informed Ensembles. Symmetry, 11(3), 411. doi:
policy and practice as a democratizing process. 10.3390/sym11030411.
Marginality and difference in education and beyond. Juszczyk, M., Zima, K., and Lelek, W. (2019). Forecasting
London: Trentham Books, 31-45. of Sports Fields Construction Costs Aided by
Gudauskas, R., Jokubauskiene, S., Zavadskas, E., Ensembles of Neural Networks. Journal of Civil
Kaklauskas, A., Binkyte-Veliene, A., Peciure, L., Engineering and Management, 25(7), 715-729. doi:
Budryte, L., and Prialgauskas, D. (2015). Intelligent 10.3846/jcem.2019.10534
Decision Support System for Leadership Analysis. Khosrowshahi, F. (2015). Enhanced project brief:
Procedia Engineering, 122, 172-180. doi: Structured approach to client-designer interface.
10.1016/j.proeng.2015.10.022 Engineering, Construction and Architectural
Hajdasz, M. (2015). Managing repetitive construction in a Management, 22(5), 474-492. doi: 10.1108/ECAM-10-
dynamically changing project environment: 2014-0128
Conceptualizing the system–model–simulator nexus. Kim, K., Cho, Y., and Kim, K. (2018). A BIM-Driven
Automation in Construction, 57, 132-145. doi: Automated Decision Support System for Safety
10.1016/j.autcon.2015.05.005 Planning of Temporary Structures. Journal of
Han, Y., Yin, Z. Z., Zhang, J., Jin, R., and Yang, T. (2020). Construction Engineering and Management, 144(8).
Eye-Tracking Experimental Study to Investigating the doi: 10.1061/(asce)co.1943-7862.0001519
Influence Factors of Construction Safety Hazard Kog, F. and Yaman, H. (2016). A multi-agent systems-
Recognition. Journal of Construction Engineering and based contractor pre-qualification model. Engineering,
Management, 146(8), 04020091. doi: Construction and Architectural Management, 23(6),
10.1061/(ASCE)CO.1943-7862.0001884 709-726. doi: 10.1108/ECAM-01-2016-0013
Hanna, A., Iskandar, K., Lotfallah, W., Ibrahim, M., and Levac, D., Colquhoun, H., and O’Brien, K. K. (2010).
Russell, J. (2018). A Data-driven Approach for Scoping studies: advancing the methodology.
Identifying Project Manager Competency Weights.

Journal of Engineering, Project, and Production Management, 2022, 12(3), 224-238

 Artificial Intelligence in Construction Projects: A Systematic Scoping Review 237

Implementation Science, 5(69). doi: 10.1186/1748- Construction Industry through Artificial Intelligence
5908-5-69 and Platform Thinking. Digitale Welt, 3, 40-44. doi:
Li, H. and Womer, N. K. (2015). Solving stochastic 10.1007/s42354-019-0211-x
resource-constrained project scheduling problems by Oyedele, L. O., Kusimo, H. O., Owolabi, H. A., Akanbi, L.
closed-loop approximate dynamic programming. A., Ajayi, A. O., Akinade, O. O., and Delgado, J. M. D.
European Journal of Operational Research, 246(1), (2019). Investigating profitability performance of
20-33. doi: 10.1016/j.ejor.2015.04.015 construction projects using big data: A project analytics
Li, Y. S., Hwang, B. G., Shan, M., and Looi, K. Y. (2019). approach. Journal of Building Engineering, 26, 100850.
Developing a Knowledge-based Decision Support doi: 10.1016/j.jobe.2019.100850
System for Prefabricated Prefinished Volumetric Podolski, M. (2016). Management of resources in
Construction. IOP Conference Series: Earth and multiunit construction projects with the use of a tabu
Environmental Science, 385, 012002. doi: search algorithm. Journal of Civil Engineering and
10.1088/1755-1315/385/1/012002 Management. 23(2), 1-10. doi:
Liu, S., Meng, X., and Tam, C. (2015). Building 10.3846/13923730.2015.1073616
Information Modeling Based Building Design Poh, C. Q. X., Ubeynarayana, C. U., and Goh, Y. M.
Optimization for Sustainability. Energy and Buildings, (2018). Safety leading indicators for construction sites:
105, 139-153. doi: 105. 10.1016/j.enbuild.2015.06.037 A machine learning approach. Automation in
Luo, L., He, Q., Jaselskis, E. J., and Xie, J. (2017). Construction, 93. doi: 10.1016/j.autcon.2018.03.022
Construction Project Complexity: Research Trends and Pruvost, H. and Scherer, R. (2017). Analysis of risk in
Implications. Journal of Construction Engineering and building life cycle coupling BIM-based energy
Management, 143(7), 04017019. doi: simulation and semantic modelling. Procedia
10.1061/(ASCE)CO.1943-7862.0001306 Engineering, 196, 1106-1113. doi:
Mahfouz, T., Kandil, A., and Davlyatov, S. (2018). 10.1016/j.proeng.2017.08.068
Identification of latent legal knowledge in differing site Qazi, A., Quigley, J., Dickson, A., and Kirytopoulos, K.
condition litigations. Automation in Construction, (2016). Project Complexity and Risk Management:
94(C), 104-111. doi: 10.1016/j.autcon.2018.06.011 Towards modelling project complexity driven risk
Martínez, D. M. and Fernández-Rodríguez, J. C. (2015). paths in construction projects. International Journal of
Artificial Intelligence Applied to Project Success: A Project Management, 34, 1183-1198. doi:
Literature Review. International Journal of Interactive 10.1016/j.ijproman.2016.05.008
Multimedia and Artificial Intelligence, 3(5), 77-84. doi: Qi, C., Fourie, A., Ma, G., and Tang, X. (2018). A hybrid
10.9781/ijimai.2015.3510 method for improved stability prediction in
McKinsey Global Institute (2015). Digital America. construction projects: A case study of stope
McKinsey & Company. MGI Industry Digitization hangingwall stability. Applied Soft Computing, 71,
Index. Retrieved from https://www.mckinsey.com/ on 649-658. doi: 10.1016/j.asoc.2018.07.035
February 28, 2020 Rodriguez-Trejo S., Ahmad A.M., Hafeez M.A., Dawood
Mehrabani, M. N., Golafshani, E. M., and Ravanshadnia, H., Vukovic V., Kassem M., Naji K.K., and Dawood N.
M. (2020). Scoring of Tenders in Construction Projects (2017). Hierarchy based information requirements for
Using Group Method of Data Handling. KSCE Journal sustainable operations of buildings in Qatar.
of Civil Engineering, 24(3), 1996–2008. doi: Sustainable Cities and Society, 32, 435-448. doi:
10.1007/s12205-020-1537-5 10.1016/j.scs.2017.03.005
Mirahadi, F. and Zayed, T. (2016). Simulation-based Russell, S. J. and Norvig, P. (2010). Artificial Intelligence.
construction productivity forecast using Neural- A Modern Approach. Third edition. Upper Saddle
Network-Driven Fuzzy Reasoning. Automation in River, New Jersey: Pearson Education.
Construction, 65, 102-115. doi: Samantra, C., Datta, S., and Mahapatra, S.S. (2017). Fuzzy
10.1016/j.autcon.2015.12.021 based risk assessment module for metropolitan
Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., and construction project: An empirical study. Engineering
The PRISMA Group (2009). Preferred Reporting Items Applications of Artificial Intelligence, 65(C), 449-464.
for Systematic Reviews and Meta-Analyses: The doi: 10.1016/j.engappai.2017.04.019
PRISMA Statement. PLoS Medicine, 6(7), e1000097. Shahtaheri M., Nasir H., and Haas C. T. (2015). Setting
doi: 10.1371/journal.pmed.1000097 baseline rates for on-site work categories in the
Mousavi, S., Vahdani, B., Hashemi, H., and Ebrahimnejad, construction industry. Journal of Construction
S. (2015). An Artificial Intelligence Model-Based Engineering and Management, 141(5), doi:
Locally Linear Neuro-Fuzzy for Construction Project 10.1061/(ASCE)CO.1943-7862.0000959
Selection. Journal of multiple-valued logic and soft Sharafi, P., Rashidi, M., Samali, B., Ronagh, H., and
computing, 25(6), 589-604. Mortazavi, M. (2018). Identification of Factors and
Ning, X., Qi, J., Wu, C., and Wang, W. (2018). A tri- Decision Analysis of the Level of Modularization in
objective ant colony optimization-based model for Building Construction. Journal of Architectural
planning safe construction site layout. Automation in Engineering, 24(2). doi: 10.1061/(ASCE)AE.1943-
Construction, 89, 1-12. doi: 5568.0000313
10.1016/j.autcon.2018.01.007 Shin, Y. (2015). Application of Boosting Regression Trees
Nnaji, C. and Karakhan, A. A. (2020). Technologies for to Preliminary Cost Estimation in Building
safety and health management in construction: Current Construction Projects. Computational intelligence and
use, implementation benefits and limitations, and neuroscience, 2015(4), 149702. doi:
adoption barriers. Journal of Building Engineering, 29, 10.1155/2015/149702
101212. doi: 10.1016/j.jobe.2020.101212 Sowa, J. F. (2000). Knowledge representation. First
Oprach, S., Bolduan, T., Steuer, D., Vössing, M., and edition. Pacific Grove: Brooks.
Haghsheno, S. (2019). Building the Future of the

Journal of Engineering, Project, and Production Management, 2022, 12(3), 224-238

238 Bang, S. and Olsson, N.

Taherdoost, H. and Brard, A. (2019). Analyzing the Yaqubi, M. K. and Salhotra S. (2019) The automated cost
Process of Supplier Selection Criteria and Methods. estimation in construction. International Journal of
Procedia Manufacturing, 32, 1024-1034. doi: Innovative Technology and Exploring Engineering.
10.1016/j.promfg.2019.02.317 8(7), 845-849.
Tidemann, A. (2020). Kunstig intelligens (Artificial Zhang, J., Huang, Y., Ma, G., Sun, J., and Nener, B. (2020).
Intelligence). Store Norske Leksikon (SNL). Great A metaheuristic-optimized multi-output model for
Norwegian Encyclopedia. Retrieved from predicting multiple properties of pervious concrete.
https://snl.no/kunstig_intelligens on February 28, 2020. Construction and Building Materials, 249, 118803. doi:
Tidemann, A. (2019). Maskinlæring (Machine learning). 10.1016/j.conbuildmat.2020.118803
Store Norske Leksikon (SNL). Great Norwegian Zhang, Y. and Luo, H. and He, Y. (2015). A System for
Encyclopedia. Retrieved from Tender Price Evaluation of Construction Project Based
https://snl.no/kunstig_intelligens on February 28, 2020. on Big Data. Procedia Engineering, 123, 606-614. doi:
Tranfield, D., Denyer, D., and Smart, P. (2003). Towards 10.1016/j.proeng.2015.10.114
a Methodology for Developing Evidence-Informed Zheng, X. L. and Wang, L. (2015). A multi-agent
Management Knowledge by Means of Systematic optimization algorithm for resource constrained project
Review. British Journal of Management, 14(3), 207- scheduling problem. Expert Systems with Applications.
222. doi: 10.1111/1467-8551.00375 42(15-16). doi: 10.1016/j.eswa.2015.04.009
Tørresen, J. (2013). Hva er kunstig intelligens? (What is Zhou, J. and Ren, D. (2020). A hybrid model of external
Artificial Intelligence?) Oslo: Universitetsforlaget. environmental benefits compensation to practitioners
University of Helsinki (2018). Introduction to AI. How for the application of prefabricated construction.
should we define AI? Elements of AI. Retrieved from Environmental Impact Assessment Review, 81, 106358.
https://www.elementsofai.com/ on March 24, 2021 doi: 10.1016/j.eiar.2019.106358
Vakhshouri, B. and Nejadi, S. (2015). Predicition of Zou, Y., Kiviniemi, A., and Jones, S. (2017). Retrieving
Compressive Strength in Light-Weight Self- similar cases for construction project risk management
Compacting Concrete by ANFIS Analytical Model. using Natural Language Processing techniques.
Archives of Civil Engineering, LXI(2), 53-72. doi: Automation in Construction, 80. doi:
10.1515/ace-2015-0014 10.1016/j.autcon.2017.04.003
Vickranth V., Bommareddy S. S. R., and Premalatha V.
(2019). Application of lean techniques, enterprise Sofie Bang is a Ph.D. Candidate at the
resource planning and artificial intelligence in Norwegian University of Science and
construction project management. International Technology. The Ph.D. is conducted
Journal of Recent Technology and Engineering, 7, 6C2. in collaboration with Construction
Wang D., Liu J., Wang X., and Chen Y. (2019). Cost- City Cluster, and explores and maps
effectiveness analysis and evaluation of a ‘three-old’ the opportunities, challenges, and
reconstruction project based on smart system. Cluster future outlooks associated with the
Computing, 22, 7895-7905. doi: 10.1007/s10586-017- increased use of artificial intelligence
1490-3 in construction projects. Bang holds
Wood, H. and Gidado, K. (2008). An Overview of an MSc in Mechanical Engineering,
Complexity Theory and its Application to the with a specialization in Project and Quality management.
Construction Industry. Proceedings of 24th Annual Her research interests include artificial intelligence,
Conference of the Association of Researchers in construction, project management and sustainability.
Construction Management, 677-787.
Xiao, C., Liu, Y., and Akhnoukh, A. K. (2018). With a Ph.D. from the Norwegian
Bibliometric Review of Artificial Intelligence (AI) in University of Science and Technology
Construction Engineering and Management. (NTNU) and an MSc from Chalmers
International Conference on Construction and Real in Sweden, Olsson is a full professor
Estate Management 2018, 32-41. doi: in project management at the
10.1061/9780784481721.004 department of production and quality
Xing Y., Song Z., and Deng X. (2016). Optimizing the engineering at NTNU in Trondheim,
schedule of dispatching construction machines through Norway. He has served as research
artificial intelligence. Chemical Engineering coordinator for the Concept research
Transactions,51, 493-498. doi: 10.3303/CET1651083 programme on large governmental
Xu, J., Zheng, J., Zhao, B., and Gong, D. (2020). projects and has had a professorship in facilities
Application of remote electron microscope technology management. Olsson has extensive experience as a
in construction management of building engineering. consultant, research scientist and manager. His consulting
Acta Microscopica, 29, 1. 328-337 experience includes Ernst & Young and DNV (Det Norske
Yaseen, Z., Ali, Z., Salih, S., and Al-Ansari, N. (2020). Veritas). Current research is focused on emerging aspects
Prediction of Risk Delay in Construction Projects of project management and railway traffic management.
Using a Hybrid Artificial Intelligence Model.
Sustainability, 12(4), 1-14. doi: 10.3390/su12041514

Journal of Engineering, Project, and Production Management, 2022, 12(3), 224-238