sensors

Review
Artificial Intelligence Methods for the Construction and
Management of Buildings
Svetlana Ivanova 1,2, * , Aleksandr Kuznetsov 3 , Roman Zverev 4 and Artem Rada 4

1 Natural Nutraceutical Biotesting Laboratory, Kemerovo State University, Krasnaya Street 6,

650043 Kemerovo, Russia
2 Department of TNSMD Theory and Methods, Kemerovo State University, Krasnaya Street 6,
650043 Kemerovo, Russia
3 Computer Engineering Center, Digital Institute, Kemerovo State University, Krasnaya Street 6,
650043 Kemerovo, Russia; adkuz@inbox.ru
4 Digital Institute, Kemerovo State University, Krasnaya Street 6, 650043 Kemerovo, Russia;
r.zverev@i-digit.ru (R.Z.); rada.ao@kemsu.ru (A.R.)
* Correspondence: pavvm2000@mail.ru; Tel.: +7-384-239-6832

Abstract: Artificial intelligence covers a variety of methods and disciplines including vision, per-
ception, speech and dialogue, decision making and planning, problem solving, robotics and other
applications in which self-learning is possible. The aim of this work was to study the possibilities of
using AI algorithms at various stages of construction to ensure the safety of the process. The objects
of this research were scientific publications about the use of artificial intelligence in construction
and ways to optimize this process. To search for information, Scopus and Web of Science databases
were used for the period from the early 1990s (the appearance of the first publication on the topic)
until the end of 2022. Generalization was the main method. It has been established that artificial
intelligence is a set of technologies and methods used to complement traditional human qualities,
such as intelligence as well as analytical and other abilities. The use of 3D modeling for the design of
buildings, machine learning for the conceptualization of design in 3D, computer vision, planning for
the effective use of construction equipment, artificial intelligence and artificial superintelligence have
been studied. It is proven that automatic programming for natural language processing, knowledge-
Citation: Ivanova, S.; Kuznetsov, A.; based systems, robots, building maintenance, adaptive strategies, adaptive programming, genetic
Zverev, R.; Rada, A. Artificial
algorithms and the use of unmanned aircraft systems allow an evaluation of the use of artificial
Intelligence Methods for the
intelligence in construction. The prospects of using AI in construction are shown.
Construction and Management of
Buildings. Sensors 2023, 23, 8740.
Keywords: artificial intelligence; construction objects; digital twins
https://doi.org/10.3390/s23218740

Academic Editors: Vincent Gan,

Mingzhu Wang, Weiwei Chen and
Ali Ghahramani 1. Introduction
Received: 3 September 2023 Many unpredictable factors, including the human factor and the unstable environment
Revised: 4 October 2023 in conditions of great uncertainty, make the building sector one of the riskiest. Most
Accepted: 23 October 2023 countries experience high rates of urban and industrial development, and therefore it
Published: 26 October 2023 is difficult to prevent a high level of mutilation and accidents on building sites. Falling
from a height, blows, abrasions and electric shock are among the most common kinds of
accidents [1]. The level of fatal accidents in the building sector is higher than in other areas
and accounts for 30.44–40.06% of deaths worldwide [2]. The rate of non-fatal mutilation
Copyright: © 2023 by the authors.
in the building sector is 71.51% higher than in other economic sectors [3]. Among the
Licensee MDPI, Basel, Switzerland.
This article is an open access article
main factors affecting the level of injuries, the absence or insufficient training of employees,
distributed under the terms and
poor awareness of the dangers by project managers and reckless actions of employees
conditions of the Creative Commons
should be noted [3]. Artificial intelligence (AI) is the capability of a computer to learn,
Attribution (CC BY) license (https:// make decisions and perform actions characteristic of human intelligence [4]. In general,
creativecommons.org/licenses/by/ AI combines mathematics, computer science, biology, psychology, cybernetics and many
4.0/).

Sensors 2023, 23, 8740. https://doi.org/10.3390/s23218740 https://www.mdpi.com/journal/sensors

 Sensors 2023, 23, x FOR PEER REVIEW 2 of 37

Sensors 2023, 23, 8740 2 of 35

and many other sciences. AI technologies allow for the creation of algorithms and pro-
grams that train computers to solve problems independently.
other sciences. AI technologies allow for the creation of algorithms and programs that train
Approximately
computers 7.22% ofindependently.
to solve problems the world’s working population are employed in the con-
struction sector. Despite its economic
Approximately 7.22% of the world’s importance,
workingthe obvious problem
population are employed of thisin area
theiscon-
low
struction sector. Despite its economic importance, the obvious problem of this area isof
labor productivity in the construction process, which negatively leads to the waste
labor,
low material
labor resources
productivity andconstruction
in the financial resources.
process, whichSince negatively
construction activity
leads to themakes
wastea
of labor, material resources and financial resources. Since construction activity makesout
great economic contribution to the development of society, it makes sense to carry a
proper construction management in order to raise productivity.
great economic contribution to the development of society, it makes sense to carry out If construction produc-
tivity increases
proper constructionby 50.11–60.04%
management in or order
higher, to itraise
is estimated that If
productivity. this will bring an
construction additional
productivity
RUB 1.6 trillion in profits to the industry each year and further increase
increases by 50.11–60.04% or higher, it is estimated that this will bring an additional RUB global GDP [2].
One of the restrictions of the industry’s development is
1.6 trillion in profits to the industry each year and further increase global GDP [2].low digitalization and the
predominance of manual labor. This leads to the inefficient use
One of the restrictions of the industry’s development is low digitalization and theof funds and resources,
project delays,ofpoor
predominance quality
manual labor.of This
execution,
leads to lowtheperformance
inefficient useindicators
of fundsand andrisks to the
resources,
healthdelays,
project and safety of employees
poor quality [5]. The
of execution, construction indicators
low performance industry and facesrisks
manyto theproblems
health
(Figure
and safety1)ofthat hinder its
employees [5].growth and lead toindustry
The construction an extremely low level
faces many problemsof productivity
(Figure 1)
compared to the manufacturing industry [6]. Indeed, the
that hinder its growth and lead to an extremely low level of productivity comparedbuilding sector is one of the to
least
the digitized branches
manufacturing industryof production
[6]. Indeed,inthe thebuilding
world [7]. Theislack
sector oneof of digital
the leasttechnologies
digitized
and the of
branches excessively
productionmanual nature[7].
in the world ofThethe lack
branch makestechnologies
of digital project management unneces-
and the excessively
sarily wearisome
manual nature of theand more
branch complex
makes project [8,9]. The lack of
management adequate digital
unnecessarily wearisomeknowledge and
and more
technology
complex implementation
[8,9]. in the building
The lack of adequate branch is associated
digital knowledge with project
and technology delays, cost
implementation
ininefficiency,
the building poor quality
branch of execution,
is associated withuninformed
project delays, decision making andpoor
cost inefficiency, low quality
productiv-of
ity as well as health and safety indicators [4]. Taking into account the existing problems of
execution, uninformed decision making and low productivity as well as health and safety
indicators [4]. Taking
labor shortage, into to
the need account
ensurethe existing problems
sustainable of labor
construction rates,shortage, the needpan-
the COVID-19 to
ensure
demicsustainable
and globalconstruction rates,itthe
financial crises, COVID-19
became pandemic
obvious that inandtheglobal
coming financial
years crises,
the con-it
became obvious
struction thatshould
industry in the coming
rapidly years
develop the itsconstruction
technologicalindustry should
potential andrapidly develop
intensively im-
its technological
plement potential
AI methods and intensively implement AI methods [10–13].
[10–13].

Figure1.1.Key
Figure Keyproblems
problemsofofthe
themodern
modernconstruction
constructionindustry.
industry.

AI in construction refers to the use of computer systems and algorithms to perform

complex construction and construction management tasks [2,4]. It is expected that AI
Sensors 2023, 23, 8740 3 of 35

technologies will determine the development potential of the building branch in the coming
years.
Over the past ten years, studies show that the application of AI technologies is increas-
ingly spreading across the building sector: from all areas of architectural and engineering
design to building services [14], with artificial neural networks (ANN) replacing building
professionals such as architects. The first such developments in the construction indus-
try emerged in the 1960s, for example, an architectural machine [15]. Over the past half
century, many works have been published with solutions to specific construction tasks
using various artificial intelligence technologies. Thus, machine learning (ML) [16] has
replaced knowledge-based systems as a new area of interest for the building branch, while
optimization [13], robotics [17] and 3D printing have become the most interesting topics
of research on the use of AI in construction. In the second decade of the 21st century,
various digital technologies, including augmented reality (AR), blockchain, Internet of
Things (IoT) and quantum computing, were integrated with AI to solve problems in the
building industry [18,19]. In recent years, AI research in the building sector has focused
on the development and implementation of various methods: the probabilistic, statistical
and economic methods, including formal logic, mathematical optimization, artificial neural
networks, received the combined name of architectural artificial intelligence (AAI). The
development of the building branch was affected by the complexity of the technologies
used [20].
Industrialization, digitalization and globalization have significantly influenced the
changes in the building branch. There is an opinion that the next major changes that
the industry will face over the next five years will include consolidation, customer focus,
human resources, internationalization, investment in technology, product approach, spe-
cialization, sustainability value chain and supply chain management [21]. The current
shift from project-based building processes to a product-based approach will change these
innovations. Companies will produce structures outside the construction site, instead of
on-site construction similar to the processes in other types of production activities [7]. It
is expected that in the future, the construction process will be more product-based, so the
design and products will be produced outside the construction site. Moreover, the degree
of internationalization will be increased by consolidating the value chain [22].
Analytics and customer behavior data will be used to optimize future projects. The
degree of development of AI technologies will determine the future of the design, ecosystem
or schedule of building, workplace productivity and occupational safety. Digitalization is
necessary for data-dependent decision making and consolidation of the value chain [23,24].
The effectiveness of artificial intelligence and machine learning in the field of imple-
mentation reaches robotic bricklaying, welding, pouring concrete and even erecting the
entire building using 3D printing technology.
Such software as Allplan (starting from version 2014), ArchiCAD (starting from ver-
sion 2.0), AutoCAD (starting from version 7.1), nanoCAD (starting from version 3.0), Revit
(starting from version 8.0), SCADOffice (starting from version 11.5), Compass (starting
from version 5.0), etc. are particularly popular in construction [25]. The AutoCAD program
allows for the creation of necessary drawing documentation and 2- and 3-dimensional mod-
els of the projected object using the built-in library of graphic elements. There are printing
control capabilities even for three-dimensional objects [25]. The advanced capabilities of
AutoCAD are concentrated in the following applications: Architecture (documents and
drawings); Civil 3D (infrastructure design, land management, landscape, etc.); Inventor 3D
(design of complex sites); and Navisworks (monitoring of architectural projects). Commer-
cial use of AutoCAD requires the purchase of a license [26]. ArchiCAD is considered one of
the best computer applications for architectural and construction design (creation of virtual
models of building construction). It is also possible to create project documentation for the
object being developed [27].
The scientific novelty of the study is presented in a review of modern literature
concerning the issues of using AI. For the first time, published research data on the use of
Sensors 2023, 23, 8740 4 of 35

AI in automating routine but important tasks for construction management, for competent
task planning, to prevent component delays, to resolve various kinds of conflicts and to
meet deadlines for the completion of objects are analyzed. For the first time, arguments are
presented that, by using AI, organizations can identify and eliminate potential problems
faster than with the help of human potential. It has been found that project managers prefer
to use AI to verify and analyze significant amounts of data.
In our opinion, the specifics of AI in construction are as follows: Firstly, the duration
of the assessment of the real effectiveness of innovations associated with the life of the
building, during which certain shortcomings of the design and technological solution can
be identified. In this regard, it is necessary to apply balanced and reasonable approaches
to the selection of innovative building materials and technologies in the construction of
real estate. Secondly, the high level of responsibility of builders for the final result and the
consecutive manifestation of a certain conservatism in the use of AI in order to eliminate
errors in the design or in the choice of technologies, as well as to reduce the level of risk
to human life and health. Thirdly, the formation of certain stereotypes and traditions as
a result of a long history of AI development in construction, and the consecutive use of
tested materials, as well as design and technological solutions that are quite difficult to
modify [23].
Therefore, achievements in the sphere of AI and machine learning have also touched
on the main problems of the building branch. The aim of this work was to study the
possibilities of using AI algorithms at various stages of building, including ensuring the
safety of the process. The main sections of this article are the history of creation, relevance,
problems and prospects of using AI in the construction industry. The collected information
can be useful to both researchers and specialists engaged in the building branch.
The main issues to study in the article are the following. This study describes the
following types of AI: 3D modeling for building design, BIM uses machine learning to
conceptualize design areas in 3D, computer vision, ALICE—construction planning, image
recognition technology for tracking working hours, machine sensors—planning the effec-
tive use of construction equipment, artificial narrow intelligence (ANI), artificial general
intelligence (AGI) and artificial superintelligence (ASI). It also describes automatic program-
ming for natural language processing, knowledge-based systems, robots—design, creation,
operation, maintenance of buildings, adaptive strategies (ES), adaptive programming (EP),
genetic algorithms (GA), differential evolution (DE) and particle swarm optimization (PSO)
mapping using unmanned aerial systems. Our review is not systematic. The authors are
aware that it is impossible to cover the entire volume of materials on the presented topic,
and conclusions were drawn only from the information provided by the reviewed works.
The main hypothesis of this work is the justification of the necessity and acceptability
of the use of AI in construction and other industries. It is planned to conclude that the use
of AI, innovations and nanotechnology improve the quality of the finished product and
reduce construction time and costs. There are a lot of studies in the literature on the use of
AI in science and technology [7,21–26]; however, there is very little full-fledged information
about the use of AI in construction [23].
The scientific novelty of the research lies in the systematization of the existing global
experience in the use of AI at all stages of the construction process as well as in the
development of the conceptual and categorical apparatus of AI in the construction sector.
A diagram that provides a detailed view related to the implementation of a generalizing
literature review is presented in Figure 2.
 Sensors
Sensors 2023,
2023, 23,23, x FOR PEER REVIEW
8740 55 ofof3537

Figure2.2.Types
Figure TypesofofAI
AIand
andtheir
theiruse.
use.

The
Theconstruction
constructionindustry
industryisisnot notwell-receptive
well-receptivetotoinnovative
innovativetechnologies.
technologies.This Thisisis
manifested in the low level of costs for research and development
manifested in the low level of costs for research and development work, the work, the conservatism
ofconservatism
design and contracting
of design and organizations,
contractingasorganizations,
well as the traditionalism
as well as the of controlling
traditionalismstateof
bodies. However, in the last decade, the process of introducing innovations
controlling state bodies. However, in the last decade, the process of introducing into the invest-
ment and construction
innovations into the sector has noticeably
investment accelerated.sector
and construction Computer methods ofaccelerated.
has noticeably modeling
all stages of the production and construction cycle have become widely implemented [22].
Computer methods of modeling all stages of the production and construction cycle have
A serious transformation of raw materials, materials and construction technologies used
become widely implemented [22]. A serious transformation of raw materials, materials
in the industry is predicted, which entails an inevitable organizational revolution in the
and construction technologies used in the industry is predicted, which entails an
construction industry and the integration of interaction between architects, designers,
inevitable organizational revolution in the construction industry and the integration of
construction and maintenance services [23].
interaction between architects, designers, construction and maintenance services [23].
It should be noted that the pace of scientific and technological progress in construction
It should be noted that the pace of scientific and technological progress in
is closely related to the speed of transition to innovative, automated methods of building
construction is closely related to the speed of transition to innovative, automated
construction facilities, robotic complexes and technologies with minimal human labor.
methods of building construction facilities, robotic complexes and technologies with
Today, many experts agree that one of the promising trends in the construction industry is
minimal human labor. Today, many experts agree that one of the promising trends in the
an accelerated transition to the model of modular housing construction, and, consequently,
construction industry is an accelerated transition to the model of modular housing
to the conveyor production of real estate objects using unified panel or modular elements
construction,
and AI algorithmsand,[21].
consequently, to the conveyor production of real estate objects using
unified panel or modular
The unique contribution elements and AI
of this study to algorithms [21].
the use of AI is as follows. For the first time, a
study The
of allunique
types ofcontribution of this
world literature study
on the to the use of
generalization andAIsearch
is as for
follows.
ways For the first
to improve
time, a study of all types of world literature on the generalization and
the quality of construction using innovative technologies and artificial intelligence search for wayswasto
improve
conducted. the quality of construction using innovative technologies and artificial
intelligence was conducted.
Sensors 2023, 23, 8740 6 of 35

2. Methodology
The objects of this research were scientific publications and patents of Russian and
foreign authors concerning the use of artificial intelligence in construction and ways to
optimize this process. To search for information, Scopus, Web of Science, and Elibrary
databases were used for the period from the early 1990s (the appearance of the first
publication on the topic) until the end of 2022. The available review and research articles
in English and Russian on 3D modeling for building design, the use of machine learning
to conceptualize design areas in 3D, computer vision, image recognition technology for
tracking working hours, planning the effective use of construction equipment, artificial
narrow intelligence, artificial general intelligence, and individual articles related to the
justification of the relevance of the topic, understanding the properties and mechanisms
of the use of artificial intelligence in construction, and identifying promising areas of
research in this area were selected and analyzed. The main focus was on articles published
in scientific, peer-reviewed journals with a high citation index over the past five years.
Conference materials and book chapters were also used in the analysis. At the same time,
articles available only in the form of abstracts, as well as bibliographies, editorial materials
and articles published not in English were excluded. Generalization was the main method.
Statistical and research data related to the study of the use of automatic programming for
natural language processing, knowledge-based systems, robots, their design, creation and
operation were analyzed. The authors considered arguments based on the hypotheses
of leading scientists about the influence of artificial intelligence in construction and its
optimization and formed their own opinion based on the evidence of these hypotheses.

3. History of Creation and Management of Smart Objects Using Artificial Intelligence

Humanity has been preoccupied with the idea of creating artificial intelligence through-
out its existence, and the active implementation stage was launched with the invention
of computers [28]. The term “Artificial Intelligence” was introduced in 1956 by John Mc-
Carthy at a scientific conference on this topic at Dartmouth College [29], but there is still
no single definition of this concept. Artificial intelligence can be understood as a set of
algorithms and system products capable of performing actions characteristic of human
intelligence. The capabilities of AI and the breadth of its application are limited only by
existing computing power.
It is believed that AI can provide businesses with new economic benefits. At the
moment, innovative technologies are the defining resource and the main factor in the
modernization of the economy of any country. Therefore, the possibilities of using machine
learning, computer vision, and neural networks in various spheres of the economy attract
the attention of researchers and manufacturers around the world [30–49]. The construction
industry is no exception. The history of the creation and management of smart objects
using artificial intelligence technologies in the industry is closely related to robotics [50].
Robots have significant potential for use on the building site. The areas of the building
branch in which robotics can be applied are described in [51]. Based on the performance
requirements, a conceptual description of four types of construction robots has been
prepared. The effectiveness of the use of these robots is noted, indicating the presence of
possible specific problems. The authors [52] believe that most of the construction tasks
can be performed by a robot assembler (moving large building components), a general-
purpose robot (interior finishing work, exterior walls, etc.) and a robot finisher (finishing
large vertical and horizontal surfaces). The study shows the economic efficiency of these
robots due to stable performance and quality of work compared to humans. The study [53]
describes a robotic arc welding system with numerical control, which was developed using
artificial intelligence technologies. The system includes three types of robots and complex
software for preparing and controlling the welding process. The study [50] describes a
method for finding the optimal number of workers with equal functions performing joint
work on construction sites.
Sensors 2023, 23, 8740 7 of 35

The potential of automated data collection In real time, process control and robotics
for remote, large-scale engineering building projects is presented in [51]. Classifications of
automation technologies and robotics in such operations are based on hardware, remote
sensing, analog and digital telecommunications technologies, optical data transmission
methods, monitoring by computer control of devices, automatic control of mobile equip-
ment, stationary manipulators, mobile robots, etc. The addition of selected technologies
with appropriate software increases the efficiency of processes, facilitating data analysis
and design and making management decisions.
The combination of artificial intelligence methods and expert systems expands the
possibilities of reproducing human experience and expert activity. A fundamental limita-
tion of traditional planning methods is their ability not to use the information obtained
when creating a project plan but only the data generated during planning [52]. Artificial
intelligence technologies provide opportunities to create plans, substantiate stored informa-
tion and draw conclusions based on them. The use of AI of a limited amount of subject
domain information for planning, leads to inadequate solutions to complex real-world
project planning tasks. Construction planning includes the choice of construction tech-
nology, the definition of work tasks, the assessment of required resources and duration,
and cost estimates. In the development of planning systems the methodology of receiving
information is used. The 1985 study [53] describes the areas of application of expert sys-
tems based on information in the area of the control and monitoring of building projects.
The developed expert systems are computer programs that can replace highly qualified
specialists. The ability of these systems to solve poorly structured problems and to modify
the solutions found makes their application in this field desirable. Several expert systems
used at different stages of development in construction engineering both for assessing the
safety of buildings during construction and for choosing methods for the construction of
bridge buildings were discussed in [54]. The 1987 review [55] notes that there are many
different expert systems that differ conceptually and by means of implementation, which
have the potential to be used in construction projects. A prototype of the expert system
using the INSIGHT program (2 microcomputer expert-system shell) was developed for the
construction of precast, reinforced concrete constructions [56]. The criteria were collected
from experienced construction experts in this area. The system recommends the numerical
size of the construction team to perform the main work on the construction site, deter-
mines the amount of expected productivity and formulates production problems affecting
productivity.
Saleh [57] presented a computer program that automates the preparation of a project
document and complements the standard AutoCAD system for extracting project data
necessary to perform an automated quantitative statement. The program is written in C and,
if necessary, can interact with other databases that the project designer or contractor works
with. The availability of an accurate and automated method for compiling a statement of
work volumes from such libraries creates a benchmark in professional construction practice.
This study presents a computer program developed by the author. This program is a third-
party achievement that automates the above-mentioned task and complements the standard
AutoCAD system, which is not able to extract project data related to the implementation
of an automated statement of work volumes. The program is written in C and has the
ability to interact with other databases owned by the designer or contractor. AutoCAD was
chosen as the base software because it is a cost-effective CAD software, although Archicad,
Design + available software can perform automatic quantity analysis [57]. In [58], a forecast
of the quality level of concrete constructions is carried out using the QL-CONST1 expert
system, which is able to analyze information generated by human experts in a specific
field of construction engineering and structural safety. The system of automatic design
of wooden frame houses was developed to perform routine operations (accurately and
quickly builds drawings and forms estimates) using AI technologies [59]. In [60], two
planning systems—LESP 2 (a training system for drawing up an inspection plan) and IDA
(a system for finding solutions in construction)—are analyzed.
Sensors 2023, 23, 8740 8 of 35

Deterministic recommendations of professional consultants on the cost of construction

can lead to inadequate investment decisions of developers and investors in the construction
sector. In particular, expert systems [61] based on probabilistic methods and other forms of
AI make it possible to clearly take into account uncertainty and incomplete information,
and the contribution of these methods to risk assessment leads to more realistic invest-
ment decisions in conditions of uncertainty or incomplete information. The prototype
expert system CONSTRUCTION PLANEX [62] is designed for planning modular, high-rise
buildings, including the excavation, foundation and construction of buildings, generates
networks of design work, cost estimates and schedules, including a description of activities,
specification of precedents, selection of suitable technologies and assessment of duration
and costs.
Predicting activity on construction sites using AI is a powerful tool because it allows
for the provision of detailed information about current performance, the need to increase
speed and also helps to reduce the risks of future actions. In parallel, a huge number of
scenarios are simulated, which allows for the quick prediction of probable events and
the ability to “win back” one’s main actions in the real world without taking any real
risks [63,64]. The introduction of AI-based expert systems into construction takes place
in several stages [65]. Information about the object necessary for analysis is entered into
the data processing program. Increasing the amount of initial data collected reduces the
number of incorrect conclusions. Many construction companies already use automated
systems that collect the necessary data, or data from different sources are combined, which
requires more time and effort [65]. Next, a new algorithm is created or an existing one is
being refined [66]. At the last stage, the algorithm is trained and/or self-taught in order to
create a comprehensive strategy for the logistics of construction and all relevant business
processes of the construction company, taking the AI capabilities into account [66].

4. The Role of AI in the Building and Management of Smart Objects

More and more AI is turning into an autonomous, smart assistant, which is able to
emphasize the advantages of the construction business in such matters as budget manage-
ment, financing, drawing, management of automated devices and mechanisms, safety and
accuracy of production, project planning, communications management, etc. [67]. Figure 2
shows the types of AI and their use.
Investors around the world spend more than RUB 10 trillion a year on construction-
related activities, and this figure is expected to grow by another 42.36% by the end of 2023 [4].
The McKinsey report for 2020 [66] noted a lot of work on applied construction solutions
that introduce artificial intelligence into the development of house projects and building.
According to the report [67], as a new technology, artificial intelligence in construction
will approach RUB 4.5 billion by 2026. Artificial intelligence tools perform complex and
expensive tasks more accurately than humans. Over the past decades, AI has proven
to be a fantastic tool in the building sector, allowing for the conservation of money and
time and the creation of incredibly beautiful houses and buildings around the world [68].
In the nearest future, artificial intelligence and incredible computer-generated ideas will
displace building engineers and architects [69]. The drone hovers over the construction site
and almost immediately transmits observations to a tablet or computer (e.g., violations of
safety by the workers, swampiness of the site, insufficient use of construction equipment,
etc.) [70].
The ML construction program [71] predicts possible risks during construction using
the initial data before the building. The developed technology saves the time and work of
the expert team at the design stages and time-consuming planning stages.
In commercial organizations that use sensors for data collection and IoT in the building,
some success has been achieved in studying performance and resource consumption [72].
The neural system is formed from small sensors mounted on the ceiling. After the collected
information is sent to the cloud by platforms such as MindSphere, Siemens or Schneider
Electric EcoStruxure, the data obtained are interpreted and help engineers control the
Sensors 2023, 23, 8740 9 of 35

costs and use of utilities [73]. Companies producing synthetic construction management
software (PlanGrid 2023 or Procore 2023) are constantly improving the technology of using
machine learning in their processes [74]. These technologies help prioritize equipment
maintenance tasks and reduce the risks of injury at work. AI methods in construction
develop management of a developing building in the shortest possible time, reducing the
need for an architect to design, a bank that pays for the development of the project and a
manager who controls all construction functions.
In studies [75,76], AI technologies were used in the construction of railway stations
and bridges. Most of the existing railway stations in the UK are historical buildings and
have been in operation for many years. The digital twin or building information modeling
(BIM) is applied to the King’s Cross Railway Station building in London using Revit-based
building modeling. Implementation and transformation of a 3D model of the station
building into a 6D information model of the building contained characteristics of time,
costs with the calculation of carbon emissions and assumptions about reconstruction using
Revit. The study [75] presents reasonable recommendations for the implementation of BIM
in railway station projects, which can be useful in the design, planning, and operation of
both environmentally friendly and financially profitable construction projects.

5. The Relevance of the Use of AI in the Building Sector

The main AI methods [4] used for the smart construction of buildings and their man-
agement are machine learning, computer vision, robotics, automated planning, knowledge-
based systems, optimization, etc. (Table 1).

Table 1. Distribution of AI technology applications in the building sector.

AI Technologies
Application Areas
A B C D E F
Health/Safety [77–79] [80,81] - - [82,83] -
Project planning and design [83–85] [86,87] [88,89] [4] [90,91] [84,92]
Cost estimation - - [4,93] - [94,95] [96,97]
Contracts, conflict management and tenders [98] [99,100] - - [101] [102]
Supply management, equipment and logistics [2,4,103] - [104] - [13,105] [45]
Site monitoring and performance evaluation [106,107] [108–110] - [111,112] [113] -
Risk management [114,115] - - [4,99] [116,117] -
Stability - - - - [4,86] [118]
A—Machine learning; B—computer vision; C—automated planning; D—robotics; E—knowledge-based systems;
and F—optimization.

Eastman was the first to create the concept of a computerized building design system
with all the functions of modern building information modeling [119]. The basic functions
of building information modeling took a quarter of a century to enter the market, while
some, such as automatic verification of building codes in design, have yet to be fully imple-
mented. Warszawski and Sangrey pointed out in 1985 that “Robotics can be introduced in
construction in several ways: a simple development of robotics and computer technology
within existing procedures or an impressive combination of robotics and computer-aided
design/automated manufacturing, defining the basis for completely new construction
systems (construction of the future) [50].
Various methods of artificial intelligence, the use of which will lead to a more efficient
construction process (automated, reliable and self-modifying, thus, saving time and fi-
nances) are of great importance for the development of the construction industry [2]. Many
artificial intelligence technologies are new, although their ideas in the field of construction
were formed several decades ago but have not found practical application due to the lack of
a reliable digital information base of buildings. The idea of building information modeling
was first mentioned in a 1975 article [119]; the ideas of using AI tools and code verification
appeared in the mid-1980s [120].
 finances) are of great importance for the development of the construction industry [2].
Many artificial intelligence technologies are new, although their ideas in the field of con-
struction were formed several decades ago but have not found practical application due
to the lack of a reliable digital information base of buildings. The idea of building infor-
Sensors 2023, 23, 8740 mation modeling was first mentioned in a 1975 article [119]; the ideas of using AI 10 tools
of 35
and code verification appeared in the mid-1980s [120].
There is a construction tech application. It is based on the building information
modeling
Thereand
is offers the following
a construction techdigitalization
application. tools
It is [120]:
based on the building information
a.modeling
Softwareandtools
offersthat
themanage
followingdesign and construction;
digitalization tools [120]:
b.a. Software and hardware complexes for delivering
Software tools that manage design and construction; information from project to object
b. (BIM-to-field); this group includes tools that through
Software and hardware complexes for delivering information users’ mobilefromdevices
projectdeliver
to object
information about products and processes directly to workplaces
(BIM-to-field); this group includes tools that through users’ mobile devicesin the field; deliver
c. Robotic applications
information about for performing
products construction
and processes work
directly toon the site; in the field;
workplaces
d.c. Software
Roboticand hardwarefor
applications complexes
performingfor collecting information
construction work on thefrom an object and
site;
d. providing
Softwareit and
to control functions
hardware (field-to-BIM).
complexes for collecting information from an object and
The providing it to control
introduction functions
of artificial (field-to-BIM).
intelligence technologies in the construction sector is
currently
Thein its early stages
introduction but has great
of artificial potential
intelligence for the industry’s
technologies in the transformation.
construction sector is
currently in its early stages but has great potential for the industry’s transformation.
6. AI to Ensure the Safety of Construction
6. AI to Ensureoften
Accidents the Safety
occur of Construction
during construction, which shows insufficient attention to
safetyAccidents often
issues, a lack ofoccur during
effective construction,
safety supervisionwhich
and shows insufficient
a lack of attention to
safety knowledge safety
among
issues, a lack of effective safety supervision and a lack of safety knowledge among
construction personnel [121]. Statistics of accidents [122] in construction are presented construc-
tion personnel
(Figure [121].
3). In 2020, Statistics
1008 deathsof accidents
were [122]at
registered inthe
construction are site.
construction presented (Figure 3).
A construction
In 2020, 1008 deaths were registered at the construction site. A construction
worker has a 1/200 risk of getting fatally injured throughout their career [123]. About worker has
a 1/200ofrisk
60.41% of getting accidents
construction fatally injured
occurthroughout
during the their
first career
year of[123].
workAbout
at the 60.41% of con-
construction
site [122]. About 2.3 million working days are missed per year due to construction About
struction accidents occur during the first year of work at the construction site [122]. acci-
2.3 million
dents [124]. working days are missed per year due to construction accidents [124].

Figure
Figure3.3.Statistics
Statisticsofofaccidents
accidentsininthe
thebuilding
buildingsector:
sector:1—falls
1—fallsfrom
froma aheight;
height;2—bruises;
2—bruises;3—electric
3—electric
shock; 4—entrapment; and 5—the
shock; 4—entrapment; and 5—the rest. rest.

ToToimprove
improvebuilding
buildingprocesses
processesand
andthethelevel
levelofofpersonnel
personnelsafety,
safety,new
newtechnologies
technologies
are
areconstantly
constantlybeing
beingintroduced
introducedinto
intothe
theconstruction
constructionsector
sector(Table
(Table2).
2).Zhang
Zhanghas
hascreated
created
a real-time accident detection and early warning platform. As a result, warnings are sent
automatically when a risk occurs. Experimental results show that when using artificial
intelligence technology and computer vision to manage the safety of civil construction,
the safety management of workers in construction is implemented, and the level of the
construction safety management has increased to 97.41% [121]. Bigham et al. have de-
veloped an AI platform that enhances construction safety by identifying hazards at the
pre-construction stage and offers applicable Occupational Safety Management codes for the
identified hazards. The authors pay special attention to the main cause of building accidents
(the risk of falling from elevations) [125]. In recent years, the development of technologies
and appropriate software for machine learning has been aimed at providing improved
solutions for potential security threats and risks in the construction environment [126].
Sensors 2023, 23, 8740 11 of 35

Table 2. AI approaches used to ensure safety in construction.

Prediction
The Proposed Safety Solution AI Method Sources
Accuracy, %
Early detection and accident prevention platform Computer vision and AI 97.40 [121]
Predicting accidents at a construction site Artificial neural network and case-based reasoning 86.00 [127]
Detection of abnormal situations and transfer of
Machine learning and decision tree - [128]
information to the operator
AI and UAVs with deep learning algorithms, including
Road construction safety monitoring platform - [128]
YOLOv4 and DeepSORT
Practical guide to the preliminary prevention of
Artificial neural networks 80.00 [129]
accidents on construction sites
Detection of hazards at construction sites Convolutional neural networks based on regions - [130]
Assessment of safe and unsafe conditions on
Building information modeling and computer vision - [128]
construction sites
Model for determining potential risks on construction Machine learning methods: (AdaBoost analysis,
91.98 [131]
sites (fatal accidents) decision tree and logistic regression, random forest)
Helmet wearing recognition of construction workers Convolutional neural network 83.89 [132]
Automatic computer vision, which uses a
Proximity warning alert system 90.00 [133]
convolutional neural network model based on regions
Safety belt detection system for workers to reduce the
Computer sight and convolutional neural networks 99.0 [134]
risk of falling from elevations
A system that ensures the safety of builders by creating
advance warnings and alarms at underground Internet of Things - [4]
construction sites

Huang et al. [135] presented new concepts for the use of AI in building fire safety for
predicting dangerous situations: AI and Big Data for predicting limiting design conditions
to reduce the cost of fire protection systems; digital twins for fire extinguishing in a building,
combining a network of IoT sensors to visualize smoke in buildings in real time; and Digital
twins and a real-time AI-based fire prediction mechanism.
In megaprojects, accounting systems are becoming increasingly important when
ensuring security. The collected data is transferred to large repositories of information, as
there is a need to investigate problems to the smallest detail in order to solve them correctly.
The collected data are characterized by a significant level of heterogeneity due to the large
number of characteristics and types of information. For the virtual presentation of the
behavior of objects in construction, the digital twin technology appears promising. This
makes it possible for the operator to control the worker and equipment actions to prevent
accidents [136].
To predict construction accidents, Ayhan and Tokdemir [127] used two technologies:
an artificial neural network and case-based reasoning. The artificial neural network’s
performance was worse than the case-based reasoning’s. In test cases with a fatal outcome,
the prediction accuracy of case-based reasoning was 86.31%, and the prediction error
did not exceed 18.23%. The researchers presented preventive actions to eliminate safety
violations that should be implemented before and during the construction phase. To study
the normal functioning of the system Gheraibia et al. [128] proposed a new approach
using machine learning in combination with real-time operational data. Subsequently,
if there is any abnormal situation regarding the normal functioning model, the method
allows for the ability to find an explanation for the anomaly in the failure tree and then
share this information with the operator. If the failure tree cannot find an explanation
for the situation, a number of different recommendations including potential adjustments
to the failure tree are provided depending on the nature of the situation. When creating
a road construction safety monitoring platform, Zhu et al. [137] combined UAVs and
artificial intelligence technologies. To compensate for the lack of surveillance data from
the road construction site, the proposed technology made it possible to control the safety
factors of road construction in general: for builders, for construction equipment and
safety signs, and fences. For automatic detection and tracking of security factors, deep
learning algorithms included YOLOv4 and DeepSORT. Despite the fact that the proposed
platform demonstrated good results in field trials, this study had some limitations. Firstly,
Sensors 2023, 23, 8740 12 of 35

the characteristics of the monitoring platform applying to an unmanned aerial vehicle

(UAV) were determined by the stability, durability and other parameters of the UAV itself.
Secondly, there was a need to increase the efficiency of the UAV object detection algorithm
due to the uniqueness of the survey. In order to use the proposed monitoring platform
in different regions, it is necessary to ensure that all the requirements and laws of UAV
operation are met for different methods of controlling its flights.
Yeon-Chul et al. [129] applied artificial neural networks to predict accidents during
construction and to assess its applicability. As a result, the ANN model showed higher
prediction accuracy (80.16%) than discriminant analysis models (70.11%). It is expected
that the presented model can be used as a practical guide for the prevention of accidents
on construction sites. Li et al. [130] described the method of rapid object detection by
regional convolutional neural networks to detect the danger of construction on objects and
use mixed reality so that artificial intelligence could detect danger. The rapid detection
of objects by a regional convolutional neural network allows for the acquisition of expert
knowledge for the identification of objects in the image. Unlike image classification, the
complexity of object detection always implies an increase in complexity, which requires
solutions in terms of speed, accuracy and simplicity.
The process of ensuring safety at the construction site should begin from the very
planning stage. BIM can also help improve safety planning at construction sites. BIM
helps in verifying the detection of incidents that may occur during construction, and many
security threats can be avoided by planning in advance. Since manual inspection can lead
to some errors, detecting the behavior of workers in real time can help reduce the number
of accidents on construction sites. With the help of AI, one can easily monitor the safety of
construction sites. Computer vision was used to develop a safety model. By training the
model using a quantum number of images, the proposed model helps to assess safe and
unsafe conditions on construction sites and, thus, to a certain extent, indirectly reduce the
number of accidents at construction sites [138].
Based on the data (2011–2016) on industrial accidents from the Ministry of Employ-
ment and Labor Relations of the Republic of Korea, Choi et al. [131] developed a model
using machine learning to predict the potential risk of fatal accidents at construction sites.
The data presented contain information on 137,323 injuries and 2846 deaths (date of the
accident, employer’s scale, age, gender and work experience of each victim, and type of
construction). AI methods (machine learning, logistic regression, decision tree, random
forest and AdaBoost analysis) were used to identify the distribution of a dataset with the
allocation of significant variables affecting the classification in each algorithm. The predic-
tion accuracy of the presented model was quite high (91.98%). In this study, random forest
analysis identified significant factors (in descending order of significance): month/season,
employment level, age, day of the week and work experience.
Helmets play an important role in protecting builders from accidents. However, the
wearing of helmets among workers is not strictly observed for various reasons. In order to
increase the safety of construction sites, in most ongoing works, the availability and proper
use of protective helmets are monitored by multi-stage data processing with restrictions
on generalization and refinement. A convolutional neural network is the basis of a one-
step system that automatically tracks the presence of protective helmets on construction
personnel, determining the appropriate colors. The accuracy of the evaluation by the
developed system is 83.89% [132].
The study [20] presents quantitative models for predicting types of injuries (ML
algorithms, including fine trees, ensemble of reinforced trees, xgboost, random forest, two
types of machines with support vectors and logistic regression). A total of 16,878 records
of accidents at construction sites in Australia were used as the initial data for testing the
algorithms. The quality of the simulation was checked by six performance indicators.
The best values for these indicators were demonstrated by the random forest algorithm.
The nature and mechanism of the accident had a significative effect impact on the critical
characteristics of the types of injuries.
Sensors 2023, 23, 8740 13 of 35

Thanks to advances in computer vision based on 2D digital images/videos, it has

become possible to capture and identify automatically unsafe behavior and risks of per-
sonnel in real time. The study of the potential practical application of computer vision in
the sphere of the building sector has stimulated a lot of research in this area. The inabil-
ity of computer vision to accurately detect objects was the main obstacle to its spread in
construction practice. Developments in the sphere of deep learning of computer vision,
which increase the accuracy, reliability and ability to generalize object detection, have
made it possible to eliminate this disadvantage and expand the possibilities of its use on
construction sites [137].
Load-bearing buildings (concrete or steel) ensure the stability of engineering construc-
tions due to load redistribution. When performing construction work on the site, people
often tend to take a shortcut, crossing the supports to perform their daily tasks and save
time. With such unsafe behavior, the probability that a person will be injured or fatally
injured increases significantly. To solve this problem of detecting people crossing struc-
tural supports during the building of an object, Fang et al. [133] developed an approach
to automatic computer vision using a mask region-based convolutional neural network
(CNN). The developed algorithms (1) automatically detect the presence of people and
(2) confirm their presence, identify the relationship between people and concrete/steel
supports. The detection accuracy in the first and second cases was 90.35% and 75.31%. The
developed Mask R-CNN can only detect people crossing supports during construction
work [134–136,139].
Solving safety problems in the building branch determines the increased interest in the
improvement and application of computer vision methods. Most of the studies are focused
on the recognition of objects and actions, a small part of which studied the recognition
of unsafe behavior. Studies developing computer vision-based systems for assessing and
predicting safety at construction sites have practically not been conducted [140].
Computer vision techniques can be used to detect and recognize faces at various
points to ensure that only qualified people can enter certain areas to perform the necessary
work, including working on certain equipment. A motion detection mechanism in certain
areas can be implemented by motion detection cameras. The interaction of workers and
machines can be recorded in real time, and the emotions of workers can be monitored
to check their well-being. The safety of employees in the workplace is always the main
concern of companies. Drowsiness detection devices can be installed on cranes to monitor
the concentration of crane operators during operation [141]. To minimize the risk of falling
from elevations, the presence of safety belts on construction workers was detected using
computer vision technologies and convolutional neural networks. The accuracy of such
models can reach 99.35% to determine workers who do not have safety equipment to work
at heights [126].
The Internet of Things is a concept of network data transmission between physical
objects equipped with built-in means and technologies of interaction with each other or
with the external environment. This AI technology operating on a building information
modeling platform is able to ensure the safety of builders by creating advance warnings
and alarms at underground construction sites [4].
Ensuring safety at construction sites with the help of artificial intelligence requires the
development and training of algorithms, taking into account the specific tasks expected to
be solved on the construction site. The effectiveness of such a tool for ensuring safety at
construction sites, is affected both by the reliability and accuracy of AI forecasts and by the
amount of funds invested in the development of these technologies.

7. Application of AI for Cost Estimation Contracts and Management of Applications,

Tenders and Conflicts
In the management of building facilities, AI technologies are an important tool for
forecasting the company’s strategy, making bidding decisions, forecasting construction
deadlines and computing the cost of the construction process [142]. Digital cost estima-
Sensors 2023, 23, 8740 14 of 35

tion and contract management products are widely used to improve the efficiency of
construction project processes at the planning stage (Table 3).

Table 3. Artificial intelligence approaches used to estimate costs and manage contracts and conflicts,
bids and tenders.

The Proposed Safety Solution AI Method Sources

BIM-IoT-blockchain technology;
Estimation and planning BIM for estimation of the time and cost; [143–145]
Deep learning for estimation of the time and cost.
Blockchain technologies;
Supply chain management Risk monitoring system; [2,146–148]
Mobile supply chain.
Machine learning algorithms;
Construction contract management [103,149,150]
Blockchain technologies.
Cost management Blockchain technologies and encryption. [4,151]
Statistical and cluster analysis;
Audit systems of financial statements of building
Blockchain technologies for audit of the BIM [4,88,151,152]
company
modification.

Accurate cost estimation at the preliminary stages of the development of a construction

project is crucial for making informed planning decisions. To process the actual data,
Dimitriou et al. [153] applied the intelligence of feed forward artificial neural networks
(FFANN) to accurately assess the list of volumes of work on the construction of bridges.
The FFANN model captures complex relationships in a dataset and provides accurate cost
estimates.
Nonparametric estimation of the cost of construction projects using artificial neural net-
works seems promising for preliminary calculations. The cost predictors characterizing the
project/object formed the basis of the estimation concept [154]. The prediction model based
on the neural network of back propagation (BP) and improved particle swarm optimization
(PSO) is an effective system for managing water consumption costs in construction. The
researchers identified four main factors affecting water consumption during construction
(the volume of concrete pouring during the day, the weather during the day, the number of
workers and the amount of wood used during the day). The forecast data mostly coincided
with the actual data, and the relative average error of such a model was 2.66% [155].
The assessment of the consumption of basic materials in civil construction is very
important at the initial project stages. Its value is expressed in the influence of the amount
of materials on the formation of prices of individual items, and hence on the formation
of the total construction cost. Among other things, construction companies use the esti-
mation of the quantity of materials as a basis for offering on the market. The accuracy of
the offer taking into account the general conditions of business implementation directly
affect the profit that the company can receive from a particular project. Due to the low
performance of the initial mathematical model of the pre-cost budget of the installation and
construction facility, Lin and Lu proposed a mathematical model of the pre-cost budget of
the installation and construction facility based on an improved neural network algorithm.
The authors reported that the initial cost of an assembly construction project includes the
cost of manufacturing components, the cost of transportation components and the cost
of installation components. Based on the cost estimation result, a mathematical model of
the pre-cost budget of the object of pre-erected construction is built based on the project
parameters [156]. There is not enough available data in the project’s early stages, notably
when it comes to data needed to estimate the consumption of materials, and therefore the
accuracy of estimating the consumption of materials at the project early stages is lower. The
article presents research on the application of AI to estimate the consumption of concrete
and reinforcement and the selection of optimal evaluation models. The evaluation model
was developed using artificial neural networks. The best model of an artificial neural
Sensors 2023, 23, 8740 15 of 35

network showed high accuracy in estimating material consumption expressed as an aver-

age absolute error in percentages; 8.56% for estimating concrete consumption and 17.31%
for estimating reinforcement consumption [157]. The complexity and interdependence of
sources of the risk of delays in the implementation of projects are the main problems of
the construction sector. Gondia et al. [158] have developed a machine learning model that
is able to accurately predict the risk of delay of construction projects using objective data
sources. The prediction was performed on the basis of a Bayesian model.
Since construction contracts usually have an extremely high cost and are financed
from the state budget, bidding procedures should not allow inconsistent behavior of
bidders. The bidding offer for construction projects includes a number of factors. Minli
and Shanshan [159] applied the neural network method to the bidding offer, which not
only developed the offer accuracy but also made it more convenient and easier to use. To
conduct a transparent bidding procedure, Anysz et al. [142] developed a database that
includes hundreds of bidding procedures in the Polish road construction industry based
on artificial neural network machine learning methods. As the main factors, the authors
indicated the bidding participants, the location of road sections, the cost of applications,
the winners and the types of roads. Each procedure was evaluated and assigned to a
set with a given level of probability of collusion. Then, two machine learning methods
were applied. The first method consisted in training an artificial neural network (ANN)
to classify procedures according to the above-mentioned sets. Another method was to
use the predictive capabilities of artificial neural networks enriched with fuzzy set theory.
The created tool can be used for future bidding procedures as a preliminary check of the
appearance of collusion [153].
Minli and Shanshan [159] used a back propagation model in an artificial neural
network to predict the bidding offer. The authors identified various factors influencing
the bidding offer and applied these factors as input nodes of the network for performing
repetitive operations in the network. As a software support tool, the MATLAB application
software package was used to reduce the load when writing code and expand the use of a
neural network.
To reduce risks and the possibility of managing them, Kim et al. [160] developed a
model for predicting financial losses due to accidents, which allows for the organization of
effective and sustainable management of construction works. This model was based on the
algorithms of deep learning using data on financial losses at construction sites. As a result
of validation, the model has increased the reliability of forecasting the cost of financial
losses, as well as improved the forecasting method, which makes it suitable for effective
management of construction projects and reducing the risk of financial losses.
AI in the field of cost estimation, contract management, conflicts, bids and tenders in
construction can significantly improve the efficiency and accuracy of the process, save time
and reduce the risk of conflicts. But the lack of a sufficient basis for analysis, for example,
emerging conflict situations, leads to distortion of the true picture, incorrect decisions and
a devaluation of the obvious benefits of existing technologies.

8. AI at the Design Stage

Construction planning is a complex process. Traditional planning methods are usually
based on intuition and analysis, which often do not take into account all significant charac-
teristics and are accompanied by errors. As a result, there are project delays, cost overruns
and low project productivity [45]. At the design stage, the possibility of using AI technolo-
gies is important for predicting the shear strength of the soil, the preliminary cost and the
duration of construction. At the initial stage, the volume of such information is limited,
but at least the preliminary cost and duration of construction tasks are important for both
bidders and customers [161,162]. Similarly, before starting construction, it is important to
know the shear strength of the soil in order to have an idea of the ability of the building to
withstand the harsh effects of flooding and/or earthquakes.
Sensors 2023, 23, 8740 16 of 35

Artificial neural networks, regression and the support vector machine are the most
popular artificial intelligence technologies in the building branch, in descending order.
The popularity of these methods is explained by high accuracy, minimal time and easy
implementation of the generalized optimal solution [163]. Improving the accuracy of
construction planning can be achieved by including additional initial data (historical and
site-specific) in the AI analysis, which traditional planning methods often do not take into
account. To achieve optimal performance when performing construction projects, Obianyo
et al. [45] applied soft computing methods to monitor project activities and evaluate the
construction schedule. An ANN and neuro-fuzzy model (ANFIS) was built based on the
implementation documents and data from the construction schedule of a residential two-
story reinforced concrete frame building. Microsoft Project, MATLAB, and the Levenberg–
Marquardt learning algorithm (Trainlm) were used to implement the analysis. MAE, RMSE
and R-values (parameters of the loss function) were taken as criteria for evaluating the
efficiency of the developed models. The efficiency of the ANFIS model exceeded the ANN
model with satisfactory results for solving complex relationships between model variables
to obtain an accurate target response.
With the rapid development of Industry 4.0, AI methods made it possible to introduce
computer-aided design into the development of building materials [164]. The digital twin
contributes to the digitization of building materials. Intelligent application and operation
of buildings using the IoT provides the basis for optimal placement of building materials.
At the design stage, innovative materials are used in the construction of modern buildings
(fiber, carbon fiber, peat blocks, magnesium oxide board, micro cement, infrared heating
panels, eco-wool, nanocrete, etc.), the advantages of which are described in [13].
The processes of design and project management in construction are changing sig-
nificantly with the use of computer vision technologies. These technologies allow for the
management of the process of making managerial decisions, and to obtain information, they
provide the collection, processing, analysis of digital images, extracting multidimensional
data about a real phenomenon [165].
The transition to information modeling began in the last century (USA since the 1970s,
the Nordic countries since 2002, China since 2016 and Russia since 2022). In Singapore,
more than 80.60% of the building branch has been transferred to BIM technologies, in the
USA—at least 50.10%, in the UK—at least 70.00%, in France—at least 60.46%, and in Russia
12.27% of developers use BIM technologies [166].
Object detection using an algorithm of deep learning and computer vision can be
used to test components of building materials. This may include contour detection, object
counting and object sizing [141]. Cao et al. [167] considered the use of artificial intelligence
algorithms in various civil engineering applications, such as forecasting and evaluating
various parameters of composite beams and connectors for shear, as well as determining
the compressive strength of concrete.
Elhegazy et al. [168] have developed a simple computer model of a multi-storey
building ceiling system for optimal planning at the design stage. The proposed AI model
is able to predict costs very quickly and accurately when a significant amount of data is
available for training the model. In addition, after training the model, deep knowledge of
the internal mechanism of AI is not required to apply it in real projects. In such cases, the
estimator can simply enter the basic structural properties and the corresponding specific
cost of materials in order to obtain accurate output data on the cost of the structural
assembly.
Recently, various technologies and systems have been implemented on construction
sites to improve communication, coordination, planning and monitoring of the project,
including web technologies, cloud computing, building information modeling and track-
ing technologies. These new applications are commonly used in various technological
combinations to improve construction monitoring and allow models to be compared at
the project and actual construction stages. Building information modeling is widely used
Sensors 2023, 23, 8740 17 of 35

in construction projects to improve communication between different parties at different

stages of project design and implementation [169].
Image-based object detection is an important step in monitoring the building progress.
High and reliable detection speed can be provided by neural network and machine learning
technologies. Since the marking of training data is performed manually, this is a very
tedious and time-consuming process.
Braun and Borrmann [170] described a new approach based on 4D information models
of buildings in combination with the method of reverse photogrammetry for automatic
marking of images of building constructions. A large number of object snapshots are
applied to monitor the progress of point cloud processing. Any element of the building
model can be projected onto the resulting images by comparing the information model and
the resulting point cloud of the building. This approach allows for the automated marking
of objects and buildings. For training based on neural network images, these marked-up
data are used.
Jiang and Bai [171] determined the location of static vegetation using high-resolution
orthoimages obtained using drones. The proposed method includes an ultra-precise neural
network encoder based on an orthoimage, a CNN elevation map decoder and an algorithm
for disassembling an orthoimage with overlap and assembling an elevation map. The ex-
perimental datasets are eight pairs of orthoimages and elevation maps (1536 × 1536 pixels),
which are cropped to 64 pairs. This research project expanded the use of unmanned aerial
vehicles in construction, made it possible to evaluate the effectiveness of the CNN when
shooting sites and strengthened the CNN to work with large-scale images of a construction
site. This makes it possible to plan the location of objects in hard-to-reach places at the
design stage [163].
When using neural networks in the design of construction projects, it is necessary to
take into account the existing risks (insufficient accuracy of predictions due to incorrect
training of the model or insufficient initial data; difficulty in evaluating and verifying the
design due to the unpredictability of neural network forecasts; information bias or errors;
and the limited time required for a complete calculation) when making final conclusions.

9. AI in Construction Logistics
Advances in technology, a revolution in business procedures and the need to change
warehouse operations as a result of the accumulation of orders, as well as the associated
difficulties and lack of managerial skills, paved the way for the emergence of products.
Moreover, since warehousing plays a vital role in the supply chain and is a key element of
logistics, intelligent product placement is very necessary to improve the management of
the organization and achieve success. The use of AI in warehouse operations expands the
possibilities of warehouse functioning in logistics, management and coordination [172,173].
The growing fields of application and the depth of autonomous systems in logistics
create a new level of tasks for the analysis and development of concepts of human–machine
interaction. Due to the shortage of highly qualified personnel in some regions and the goals
of increasing efficiency and sustainability, logistics operators must by all means achieve
technical progress, including automation of processes [174].
The authors [175] recognize a significant gap in the low interest in logistics 4.0 and AI
shown in the research. The concept of logistics 4.0 can be described as big data analytics in
real time; for example, the optimization of routing and reduced storage requirements due
to new production technologies and autonomous robots with tracking and decision-making
systems, leads to optimization of inventory management and information exchange in real
time [176]. With the growing popularity of AI technologies, many modern logistics enter-
prises are trying to use these approaches to optimize logistics communication and improve
logistics efficiency [177]. Technological concepts of cyber-physical systems, the Internet
of Things (IoT) and the industrial IoT, etc., will allow for the organization of intellectual
logistics [178]. Pandian proposed to use AI to optimize work in the warehouse; to automate
logistics chains. The author used the algorithms of the IoT, AI and cloud computing to
Sensors 2023, 23, 8740 18 of 35

access the resources in the warehouse at any time. The proposed method of automated
warehouse logistics uses sensor networks to collect information about the number of goods
entering and leaving the warehouse, as well as artificial intelligence to properly handle
them in the warehouse, for example, to place them on the right rack, to collect goods from
the rack in accordance with the placed order, etc. The information collected by the sensor is
transmitted via the Internet to the ThinkSpeak cloud so that customers from anywhere in
the world can know about the availability of goods in stock [173].
Arabi and co-authors have developed a way to detect vehicles based on a deep learn-
ing algorithm. Taking into account the necessarily small-scale nature of the image datasets
of construction machines, the authors presented an improved version of the MobileNet
single detector, which is applicable to embedded devices, as a detection model. The results,
including a constant average accuracy above 90.24%, confirm the excellent performance of
solutions in real time [179]. The Smart Logistics program is aimed at the successful imple-
mentation of intelligent and cost-effective supply chains based on flexible and cooperative
networks and interconnected organizations. In addition, information exchange is carried
out through the use of modern technologies of information and communication, data trans-
mission networks, subjects and sensors, as well as technologies for tracking materials and
automatic identification. Automated transport, transition and storage systems supported by
autonomous vehicles should provide the possibility of partial and/or complete self-control
of systems [178]. Rashid and Louis [180] have shown that automated, reliable, real-time
recognition of equipment operation on construction sites can help improve work efficiency,
minimize downtime and reduce emissions. Previous efforts to recognize the activity of
construction equipment have explored various classification algorithms, accelerometers,
and gyroscopes. Such methods require segmentation of continuous operational data with
fixed or dynamic windows to extract statistical features. Recent developments in deep
neural networks, especially recurrent neural networks (RNNs), open up new possibili-
ties for classifying sequential time series data with repetitive lateral connections. RNNs
can automatically learn high-level representative functions through the network, rather
than creating them manually; this makes it more suitable for recognizing complex actions.
However, the application of RNNs requires a large set of training data, and obtaining
this data on real construction sites is a practical problem. The proposed data extension
methodology for a deep learning RNN that recognizes the actions of equipment is tested on
synthetic data from sample sets that were collected during two real earthmoving operations.
The presented deep learning structure surpassed the traditionally used machine learning
classification algorithms for recognizing actions in terms of accuracy and generalization of
the model [180].
The use of blockchain technology (a continuous sequential chain of blocks containing
information built according to certain rules) in supply chain management is also gradually
expanding. Supply chain management has long sought to reduce costs and improve
efficiency; it is also trying to optimize resources and reduce sector fragmentation. Trust has
always been an important factor in relationship management and in the effectiveness of
supply chain operations. Blockchain technology (Figure 4) helps in solving various tasks of
the construction industry (data tracking, contracting and resource transfer in supply chain
management). These applications help build trust between partners by making supply
chains and other joint project management processes transparent [181].
 improve efficiency; it is also trying to optimize resources and reduce sector fragmenta-
tion. Trust has always been an important factor in relationship management and in the
effectiveness of supply chain operations. Blockchain technology (Figure 4) helps in solv-
ing various tasks of the construction industry (data tracking, contracting and resource
transfer in supply chain management). These applications help build trust between
Sensors 2023, 23, 8740 partners by making supply chains and other joint project management processes trans- 19 of 35
parent [181].

Supplier of building
materials

Customer
(construction Blockchain: Warehouse
company) - Customer's order
- Logistics order
- Shipping invoices
-History of
movements

Carrier company Logistics

company

Figure 4. Application
Figure of blockchain
4. Application technology
of blockchain in the
technology building
in the sector
building (management
sector (managementofofsupply
supply chains).
chains).
10. AI in the Process of Building Construction
10. AI inComputer
the Process of Building
vision methodsConstruction
are promising in the sectors of design, architecture, build-
ing and management
Computer of facilities
vision methods to facilitate
are promising decision-making
in the processes
sectors of design, at thebuild-
architecture, building
ing stage. Manual monitoring
and management on to
of facilities construction sites is extremely
facilitate decision-making tediousatand
processes thedifficult
buildingdue
to the
stage. clutter
Manual of the sites.
monitoring Recently, scientists
on construction have beentedious
sites is extremely interested in the possibilities
and difficult due to the of
using
clutter of computer vision to scientists
the sites. Recently, solve management
have beentasks on construction
interested sites [182].
in the possibilities With the
of using
help of computer vision technologies and other AI methods, Ercan and
computer vision to solve management tasks on construction sites [182]. With the help of Wang discovered
objectsvision
computer of an technologies
automated system for AI
and other determining the time
methods, Ercan andspent
Wangon training construction
discovered objects of
an automated system for determining the time spent on training construction must
workers. To obtain a work permit in the construction industry, employees be trained
workers. To
in technical skills.
obtain a work permit in the construction industry, employees must be trained in technical
skills. To qualify, they must go through a training and evaluation process. During the
evaluation,
To qualify,according
they musttogo thethrough
provided technicaland
a training drawings, students
evaluation perform
process. Duringassembly
the
(electrical wiring, wall cable duct, etc.). It is a time-consuming process
evaluation, according to the provided technical drawings, students perform assembly for experts to check
(manually
(electrical and/or
wiring, wallvisually) the quality
cable duct, etc.). It and
is acorrectness of the process
time-consuming installation
for performed
experts to by
check (manually and/or visually) the quality and correctness of the installation per- it
interns. In order to reduce a significant number of man-hours during the assessment,
is necessary
formed to automate
by interns. In order tothereduce
assessment process,number
a significant for example, by usingduring
of man-hours computer vision
the as-
technology [183].
Augmented reality applications allow for a significant increase in the productivity
of building work, reducing the number of alterations and improving the transmission
of design ideas. Augmented reality can help significantly save time when assembling
a pipe coil compared to traditional approaches for both untrained engineers and well-
trained professional pipe installers. Those whose cognitive abilities are assessed as low,
get more benefit from the augmented reality application than other subjects. Kwiatek and
co-authors [184] proposed a manual augmented reality application that was developed
to help in the assembly and inspection of pipe assemblies using 3D scanning, scanning
and building information modeling, flexible workflow and a touch user interface. The
experiments were conducted with 21 professional pipe fitters and 40 engineering students
whose spatial cognitive abilities were measured before they assembled a complex pipe coil
using traditional and augmented reality tools. The architecture, design, construction and
facility management sector is considered one of the most intensive areas where vision-based
Sensors 2023, 23, 8740 20 of 35

systems/methods are used to facilitate decision-making processes during the construction

phase. Extensive research has been conducted to explore the possibilities of using computer
vision to solve management tasks in the field. This article provides an overview of the last
decade of computer vision research with an emphasis on modern methods in a typical
vision-based scheme and discusses the problems associated with their application.
Statistical analysis of the experimental results for these groups of subjects confirmed
several observations: augmented reality can help significantly save time when assembling
a pipe coil compared to traditional methods, both for untrained engineers and for well-
trained professional pipe installers. Additional studies of augmented reality applications
reveal the potential of their impact on the construction sector and expand the range of
possible applications [184].
Using images from UAVs Jiang and Bai [163] used a deep learning model to estimate
the heights of building constructions. The image-based convolutional neural network
encoder, a CNN elevation map decoder and an algorithm for disassembling the image
with overlap and assembling the elevation map have compiled the presented methodology.
This model allows for the accurate estimation of the height of construction sites using
unmanned drones. Li et al. [172] proposed a method based on a deep convolutional neural
network algorithm for detecting and classifying building defects during video surveillance
inspections. Inspection images collected from sewer lines (24.7 km) were used to train
and test the neural network. The hierarchical approach to classification has increased the
accuracy of detecting high-level defects by almost 5.00% (from 78.40% to 83.22%).
Zhang et al. [136] used digital twin technology to assess the movement of workers
and construction equipment on the construction site. To do this, the authors used an
open-source video dataset of 2064 video clips with 5 types of actions for excavators and
dump trucks, including the corresponding optical flow dataset for predicting movements.
To support intelligent construction, a digital twin is used, which has become a universally
recognized concept of the virtual representation of a physical object. The study says that
recognition of the movement of construction equipment and the actions of people on virtual
construction sites is no less important. A similar system is being developed in Kuzbass
(Russia) within the framework of the project “Geoinformation system of digital regional
management” [185]. Comparing a significant number of studies on the recognition of
human actions, the results of which can be extended to builders and other construction
workers, the number of studies on the recognition of the actions of construction equip-
ment is limited due to the insufficiency of video datasets with the actions of construction
equipment for training.
However, such a dataset is still relatively insufficient compared to the datasets available
in other areas of applications using deep learning solutions. It is expected that a higher
level of accuracy will be achieved after a much larger dataset is developed. It is important
to develop more complete sets of video data for different types of equipment in different
conditions, for example, when the camera is moving or in adverse weather conditions
(for example, in strong wind or rain). In addition, it is important to study the perception
of actions in which several types of construction equipment move simultaneously. In
many cases, a surveillance camera captures a construction site with several pieces of
equipment; this becomes a problem when different types of equipment require different
action recognition algorithms in one frame to achieve better recognition performance.
Creating a digital twin of a construction object based on huge amounts of data about a
building or an engineering object provides an idea of the performance and behavior on the
construction site in real time, providing intelligent information exchange from a real object
to its digital twin and optimizing the performance of the control tool. Advanced analytics
in real time is carried out in a digital double, and adjustments to the construction object are
carried out by a human operator, software or robots [186].
Artificial intelligence systems will potentially be able to study very accurate and de-
tailed models of individual objects and make a competent choice of management. The
industry is already using digital twins to optimize equipment maintenance, identify possi-
Sensors 2023, 23, 8740 21 of 35

ble failure options for parts and automate more processes. The capabilities of digital twins
will only expand thanks to more advanced sensors, faster cloud and advanced computing,
and constantly improving AI [187].
Despite the historically slow introduction of new technologies, construction companies
are beginning to successfully use AI technologies for everything (from task automation
to data mining to obtain useful information). Options for using AI technologies when
performing work on real construction sites are given in Table 4. Here are a few ways AI is
being exploited in the building branch today:

Table 4. Construction objects using AI technologies.

Object Technologies Used Country Sources

Dam construction AI technologies and 3D printer China [188]
Safety monitoring at construction sites Internet of Things and remote sensing China, USA, Russia, Canada, etc. [136,189–194]
Effective construction management ESKIMO system Germany [195]
Construction of a two-storey building 3D printer India (Larsen & Toubro Company) [196]
Digital twins created from images of
Construction management robots with cameras and neural USA (Droxel company) [197]
networks
Design of buildings BIM technologies Great Britain, Russia, France, etc. [136,198–200]
Helsinki Airport and Tampere Deck
BIM technologies Finland [136]
and Arena
Reinforced concrete bridge BIM technologies Peru and France [136]
Pidekso Dam BIM technologies Indonesia [136]
Motorway with tunnels BIM technologies Norway [136]
Auckland International Airport BIM technologies New Zealand [199]
Shanghai Tower, Disneyland and
BIM technologies China [199,201]
Shanghai World Financial Center
City of New Alamein BIM technologies Egypt [201]
Autocad civil 3D and Autodesk Revit
Metro software, Synchro Pro and BIM Dubai (Acciona Company) [202]
technologies
Digital twins and shooting from a Germany, China, Czech Republic,
Road construction [203–206]
UAV Italy, USA, Malaysia, etc.

China and the USA are the undoubted leaders in the introduction of AI technologies
in different areas, including construction [207]. To form a new infrastructure, China has
already developed computing centers, deep artificial intelligence models and learning
platforms, thus, providing digital transformation in various areas of the real economy (vehi-
cles, manufacturing, building branch, etc.). Sufficient development of robotic technologies
with the support of artificial intelligence technologies expands the possibilities of their
application in construction, mining, disaster response and in types of work that require a
large amount of labor. The US surpasses China in terms of academic infrastructure, without
which the development of the necessary technologies slows down. The Beijing Academy of
Artificial Intelligence has already been established in China, and leading universities are
opening training courses in AI technologies.
By 2030, China will try to become a global innovation center in the area of AI. Accord-
ing to Roberts et al. [208] by this time, the growth of the main artificial intelligence industry
will amount to CNY 1 trillion again with the necessary development of an appropriate
legislative framework. According to the Chinese Academy of Sciences, artificial intelli-
gence is used to monitor the behavior of workers on building sites throughout China [209].
However, in 2021, the share of North American AI in construction accounted for the
largest market volume, which is provided by government initiatives in the field of AI in
the construction sector, a sufficient population with high purchasing power and constant
investments in automation [210].
The use of AI in logistics at a construction site can be useful for forecasting resource
needs, planning and optimizing delivery routes, automatic inventory management, im-
proving the efficiency of material and technical resource management processes, allocating
resources and manpower to increase productivity and reduce costs, detecting trends,
Sensors 2023, 23, 8740 22 of 35

anomalies and problems and making recommendations for improving processes and the
results of the work. But at the same time, it is necessary to integrate AI with various systems
and technologies, train personnel to work with it, ensure data security and protect personal
information, and all this requires significant organizational and financial costs.

11. Advantages and Disadvantages of AI in Construction

In the near future, AI and machine learning technologies will play a major role in im-
proving the quality of construction projects and the development of intelligent construction
in general [211]. However, these technologies have both advantages and limitations.
The use of innovative and breakthrough business models, in particular AI methods
and technologies adapted to collect data from construction sites and for existing construc-
tion applications, will allow for the support of existing construction companies and this
economic system in general. There is still no implementation of the platform approach in
the entire industry, but its partial implementation already shows great advantages. Artifi-
cial intelligence algorithms will allow the construction data to be structured and linked to
the data of other organizations, guaranteeing the technological and legal security of users.
The combined database will allow the use of various programs and applications in the
building branch [212]. However, when implementing AI methods in construction projects,
there are disadvantages, such as corruption, fragmentation of buildings, inappropriate
risk allocation, inefficient design, limited standardization, insufficient planning time and
delayed implementation of AI in the field of management and execution [213].
In the modern world, AI is rapidly changing the rules of the game in various industries,
including the construction segment. Combining AI technologies with the development
of computing resources will not only optimize processes and stimulate innovation but
also make them available everywhere, on many construction sites. For specialists in the
field of construction, it is important to understand all the advantages and disadvantages
of using AI in order to make informed decisions, overcome emerging difficulties and
ensure the economic efficiency of their company [214]. Companies can strategically im-
plement AI technologies, fully understanding the potential benefits (increased efficiency
and security, improved project planning, etc.) [70]. Khanzode and Sarode [215] and other
researchers [65,216–218] identified the main advantages and disadvantages of AI methods
(Table 5).

Table 5. Advantages and disadvantages of using AI in construction.

Advantages Disadvantages
Task completion speed High costs at the investment stage
Safe, efficient and timely execution of projects Technology costs
Ability to solve complex and stressful tasks Expensive and complex equipment repairs
Possibility of simultaneous execution of several processes Risks of unemployment of specialists
Reduction in administrative processes The need for training and retraining of the workforce
Reducing the cost of project implementation Efforts and funds for the maintenance of AI technologies
Forecasting of safety and quality risks at work sites Risks of hacker attacks and data leaks
Minimum number of errors Computational constraints
The possibility of long-term forecasting of events Presence of errors
The operation and efficiency of the system depends on an AI
Ability to process qualitative and quantitative factors
specialist
Reduction in operating costs Increasing technological dependence
Reduction in paperwork Dependence on electricity and internet connection
No need for control (operation without an operator) Limitations of the scope of application

Since AI-based programs and applications are susceptible to attacks by hackers and
cybercriminals, they lead to huge consequences for time, quality and cost. Most significantly,
the safety of builders can be jeopardized, which can lead to loss of life or life-threatening
accidents. For instance, a computer vision system that recognizes automated building
equipment can be hacked, which can lead to accidents. At this moment, there is a global
Sensors 2023, 23, 8740 23 of 35

shortage of personnel of the highest category, in particular, AI engineers who have the
necessary skills to lead serious developments in various industries. Also, investing in
AI-based equipment requires high costs. The cost of AI-based equipment maintenance
still remains high [4]. Forecasting of purchases and supplies of material and technical
resources for the construction of various kinds of facilities is not recommended without the
use of AI methods. The successful solution of a number of tasks of construction projects is
guaranteed by the application of AI methods, which iterates through numerous scenarios
in a limited time and is able to learn from calculated algorithms:
- The analysis and identification of the necessary stock of material values and the
turnover of inventory [214] determine the profitability of the construction business;
- The automation and aggregation of document flow, improvement of logistics manage-
ment by reducing the number of dealers and direct access to resource suppliers [219];
- Forecasting the demand for material resources and the formation of a supply sched-
ule [220].

12. Prospects and Limitations of AI Use in Construction

Despite the significant successes of AI in various fields, it still faces limitations in
understanding and performing complex tasks that require human judgment, creativity
and intuition. As a result, AI will not be able to completely replace the human factor in all
aspects of construction projects. There are several restrictions on the realization of AI in
construction [221]:
- Algorithms require sufficiently large amounts of data for accurate training (training
the system to detect a worker without a helmet required machine learning based on
millions of different images of a worker in a helmet), and insufficient data volume
excludes the possibility of conducting such training [221];
- The need to integrate all possible sources of information and provide conditions (a
single platform) for storing, accessing and processing the collected data [221];
- Most construction organizations do not have enough data to train their algorithms [222];
- The high level of requirements of data processing specialists can be met only by large
players in the construction business [223];
- For the proper functioning of AI technologies, human service is necessary (each event
or decision depends on several external factors leading to various consequences) [221].
In comparison with other industrial sectors, the digital transformation of processes in
the construction sector is significantly lagging behind [224]. The use of AI technologies and
the development of machine learning algorithms make it possible to optimize construction
processes at all stages of the construction of various objects [225]. Analysis of the condition
of the building branch in recent years shows that there is a growth in the costs of developing
and implementing AI algorithms in construction [226]. The accumulation of necessary data
occurs unevenly, at different stages of the construction project implementation [227]. The
BigData necessary for training AI neural networks are formed from video materials from
various construction sites, information and data obtained from security sensors, digital
models of buildings, etc. [228,229].
AI technologies and ML algorithms in the building branch have almost unlimited
application potential [24]. AI can maintain a high speed of complex decision making;
therefore, there will be a shift towards the algorithms of empowerment in the imple-
mentation of construction projects [230]. The introduction of design systems, the use of
autonomous construction equipment and robots in combination with the accumulation
of significant amounts of data for training neural networks will increase the efficiency of
the implementation of construction production facilities [231]. Artificial intelligence will
change the models of project implementation in the building branch and replace the human
workforce, reducing workplace injuries and increasing the efficiency of operations [232].
Artificial intelligence technologies, the Internet of Things and robotics can potentially re-
duce construction costs by up to 20.42%. Cameras attached to robots collect information; in
modern buildings, artificial intelligence plans plumbing and electrical systems. To prevent
Sensors 2023, 23, 8740 24 of 35

errors during construction, performance and safety problems, the AI systems monitor
the interaction of workers, equipment and objects on the building site in real-time [233].
At all stages of strategic construction planning, it is possible to benefit from the use of
machine learning algorithms, for which there are practically no obstacles to application
in the building branch [234]. It is important to note that the issue of ethical and social
aspects of AI implementation for society and employees of the industry deserves special
attention [230].
The future of AI in the building sector, and the industry as a whole, is promising.
Technologies and computing power continue to develop, and it is obvious that AI is likely to
become an integral part of the construction process in the near future [235]. Companies that
are ready to adapt to these changes and invest in AI technologies in a timely manner will
have the opportunity to benefit and maintain a competitive advantage in the industry [236].
According to Allied Market Research [68], the cost of construction management soft-
ware will reach RUB 2.3 billion by 2028. In the long term, the introduction of AI technologies
in construction will depend on the industry’s ability to apply innovative approaches to
project management during their implementation, a conscious comparison of advantages
and disadvantages, and a willingness to invest in the development of the workforce. The
capabilities of artificial intelligence will transform the industry and make the emerging
artificial environment a common phenomenon [237].
AI methods and technologies can significantly improve work on construction sites,
providing new opportunities and increasing efficiency in various aspects of construction ac-
tivities. Currently, AI technologies support machines that have much higher computational
abilities than humans, for example, the ability to sift through huge amounts of data and use
this data to make better decisions. AI allows computers to adapt to new data—basically,
learning without explicit programming. AI can analyze large amounts of data and take
into account numerous factors (weather, time schedule, budget, etc.) to predict potential
problems and optimize the construction process. AI allows for the creation of automatic
control systems that can control and regulate various aspects of construction. Robotic
systems can perform repetitive tasks (laying bricks and installation of steel structures),
while maintaining the speed and quality of the work performed. Computer vision can
be used to automatically scan and analyze surfaces, identifying defects, cracks or other
problems. Smart building systems and AI can automatically adjust lighting, heating and
air conditioning to reduce energy costs and increase comfort and thereby optimize energy
consumption at the construction site. To ensure security management, video surveillance
systems based on AI technologies can automatically detect potentially dangerous situations
(workers falling, lack of protective equipment, etc.) and warn about them.
Additional bonuses appear when various artificial intelligence approaches interact
in one construction project, complementing each other. In construction, machine learning
can be used both for analyzing large amounts of data, pattern recognition, forecasting the
cost and duration of projects, and for automating decision-making processes. Robotics
at a construction site allow the use of autonomous robots that are programmed to per-
form operations such as laying bricks or plastering walls, dismantling, lifting and moving
heavy materials, etc. Computer vision can simultaneously be used to control the quality
of work, monitor safety on the construction site, detect defects in building materials, etc.
Knowledge-based design uses knowledge bases and expert systems, previously accumu-
lated knowledge and experience to create automated design processes and decision making
in construction, as well as to reduce time and improve the quality of the design.
In general, these different AI approaches in construction complement each other,
allowing automation and optimization of various aspects of construction activities to
improve efficiency, safety and quality of work. Obviously, regardless of the industry, the
potential of AI is huge; with a competent combination of data analysis, machine learning
and other AI technologies it can increase efficiency, reduce costs and optimize any working
environment. Thanks to the ability of AI to analyze data, learn and make predictions, the
future possibilities of its application at construction sites will only increase.
Sensors 2023, 23, 8740 25 of 35

13. Consequences of the Use of Artificial Intelligence in Politics

Over the past few years, there has been significant progress in the field of AI in areas
such as smart vehicles and smart buildings, medical robots, communications and intelligent
education systems. It is expected that these achievements will have consequences in several
areas of politics and government (economic, social, educational, etc.), and increasing
attention is being paid to it around the world [88].
In the documents, both the US and the UK, it is noted that when considering ap-
proaches to AI regulation, attention should be paid to ensuring that such approaches do
not hinder innovation and progress. Moreover, the UK report calls for the creation of an
AI Commission, which will be tasked with determining the principles of AI development
and application, providing recommendations to the government and promoting public
dialogue [238]. The draft report of the European Parliament also recognizes that existing
legal regimes and doctrines can be applied to robotics but notes that “the current legal
framework will not be enough to cover the damage caused by new generation robots.”
Therefore, it requires the adoption of the EU Directive on the rules of civil law in relation
to robotics as well as the guiding ethical framework for the design, production and use of
robots [88]. It is also proposed to create a European Agency for Robotics and AI to provide
technical, ethical and regulatory knowledge to support the EU and its member states [93].
The political implications of AI also attract the attention of intergovernmental organi-
zations such as the United Nations, the G7 and the Organization for Economic Cooperation
and Development. Governments increasingly believe that AI will benefit from international
cooperation in promoting research and development as well as in identifying appropriate
responses to relevant problems [239]. In the United States, the government is encouraged to
develop an AI-related international engagement strategy and collaborate with other stake-
holders in developing AI standards. The draft report of the European Parliament calls for
international harmonization of technical standards, mainly in order to avoid risks of market
fragmentation and to uniformly solve consumer problems. It also encourages international
cooperation in setting regulatory standards under the auspices of the UN [240].
However, markers of the relative danger of uncontrolled AI technologies require
protective legal and policy measures. In September 2017, Oren Etzioni, CEO of the Allen
Institute for Artificial Intelligence, recommended that regulators focus on the “tangible
impact” of AI, rather than “defining” or “harnessing” the “amorphous” field. At the same
time, as the field and risks of AI materialize, legislators and politicians are shifting towards
how to balance the risks of national security with the promotion of technological advances
and access to information [241]. The EU Law on Artificial Intelligence regulates specific
applications such as law enforcement, voting and social networks. China, Brazil and other
countries apply their own approaches to regulation [241].
Meanwhile, US Congressional leaders dealing with AI technologies suggest it could be
months before a comprehensive federal law is passed to supplement their meager existing
regulations. The John S. McCain National Defense Authorization Act of 2019 struggles to
define AI. The 2020 National AI Initiative Act largely instructs the President and federal
agencies to assess the potential and risks of artificial intelligence [242].
Taken together, legal maneuvers to regulate AI entail many constitutional and other
legal problems concerning jurisdiction, powers, business interests, civil liberties and indi-
vidual rights. The wave of upcoming lawsuits has already begun with numerous class-
action lawsuits against industry leaders on the grounds that their chatbots improperly
acquired copyrighted data on the Internet. In June 2023, a public figure in Georgia stated
that ChatGPT had provided defamatory information suggesting that he had embezzled
money [243].
New legal efforts to respond positively to AI-related public health and safety risks
are centered around international agreements based on existing governance systems that
comply with humanitarian law. Stakeholders are calling for the creation of broad coalitions
from several countries that manage “democracy-affirming” technologies that promote
human rights and fight authoritarianism. US diplomatic agreements with the EU, Israel,
Sensors 2023, 23, 8740 26 of 35

India and other jurisdictions are under negotiation. It is hoped that a global consensus on AI
control can be achieved through these and other alliances [244]. All political consequences
should be taken into account and weighed when developing and implementing the use of
AI in order to mitigate its negative consequences and transform it into a positive tool for
society.

14. Conclusions
AI methods are of great importance for the development of the construction industry.
Their application has the potential to organize a more reliable, automated, intensive and
cost-effective construction process. The construction industry is facing a productivity
challenge and countless other challenges that could potentially be solved with AI. With
the increase in the amount of data generated throughout the life cycle of a building and
the advent of other digital technologies, AI has the opportunity to use this data and the
capabilities of other technologies to improve construction processes. To answer the research
questions in this work, the extent to which artificial intelligence technologies are used in
construction was studied. Concepts, types, components and subfields of AI are briefly
explained, as well as works using these subfields. The basic information on the areas
of application, advantages, limitations and benefits of each subdomain of AI used in
construction was presented.
This study shows that although some artificial intelligence technologies have been used
in construction research, recent advances in these technologies have improved significantly,
and the introduction of new, more powerful AI technologies is relatively slow. Moreover,
this study identified and discussed some additional opportunities and open research
challenges for AI research in construction. Although the application of AI is gradually
increasing, its relevance is further enhanced by other new trends, such as BIM, IoT, quantum
computing, augmented reality, cybersecurity and blockchain. We investigated some of
the problems affecting the implementation of AI in the industry and solved them with the
help of recommendations. This study is a useful source of information for researchers and
practitioners about relevant AI applications and research in the construction industry. This
study presented a clear understanding of internal capabilities as well as potential obstacles
in application areas. Construction stakeholders such as regulators, decision makers, highly
skilled workers and digital enthusiasts can use the information provided in this study to
clearly identify the path to AI adoption and reduce the potential risks associated with it.
The development of the construction industry is primarily associated with solutions
based on AI technologies. At the same time, a small amount of machine-readable docu-
mentation and a low percentage of the use of BIM modeling technologies remain a limiting
factor. There are opinions that there are two areas where AI can be used: integrated devel-
opment of territories (with the help of AI solutions, it is much faster and cheaper to search,
select and analyze territories to involve them in the process of integrated development) and
the use of AI in expert work (AI algorithms are gradually replacing experts when analyzing
the results of engineering surveys, project documentation, dataset, machine processing and
data indexing). However, if efficiency increases by reducing labor costs, then it is necessary
to solve the ethical issues that arise as well as the issue of retraining people. Another
solution implemented in construction using AI is to monitor project financing and inform
banks about facilities for which construction progress requires attention or verification.
In general, the use of AI technologies in construction can improve processes, increase
productivity, reduce costs and improve safety. This can be especially useful in modern
conditions when construction is becoming more complex and requires more attention to
detail.

Author Contributions: Conceptualization, S.I.; methodology, A.K. and R.Z.; analyzed and interpreted
the data, A.K. and R.Z.; formal analysis, S.I.; writing—review and editing, S.I. and A.R.; project
administration, A.R. All authors have read and agreed to the published version of the manuscript.
Sensors 2023, 23, 8740 27 of 35

Funding: This research was funded by the Ministry of Science and Higher Education of the Russian
Federation, agreement 075-15-2022-1195.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Data are contained within the article.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Choi, S.D.; Guo, L.; Kim, J.; Xiong, S. Comparison of fatal occupational injuries in construction industry in the United States,
South Korea, and China. Int. J. Ind. Ergon. 2019, 71, 64–74. [CrossRef]
2. Pan, Y.; Zhang, L. Roles of artificial intelligence in construction engineering and management: A critical review and future trends.
Autom. Constr. 2021, 122, 103517. [CrossRef]
3. Zeng, L.; Li, R.Y.M. Construction safety and health hazard awareness in Web of Science and Weibo between 1991 and 2021. Saf.
Sci. 2022, 152, 105790. [CrossRef]
4. Abioye, S.; Oyedele, L.O.; Akanbi, L.; Ajayi, A.; Delgado, J.M.D.; Bilal, M.; Olugbenga, O.A.; Ahmed, A. Artificial intelligence
in the construction industry: A review of present status, opportunities and future challenges. J. Build. Eng. 2021, 44, 103299.
[CrossRef]
5. Chien, C.F.; Dauzère-Pérès, S.; Huh, W.T.; Jang, Y.J.; Morrison, J.R. Artificial intelligence in manufacturing and logistics systems:
Algorithms, applications, and case studies. Int. J. Prod. Res. 2020, 58, 2730–2731. [CrossRef]
6. Chui, M.; McKinsey, S.F. Global Institute Study, Artificial Intelligence, The Next Digital Frontier; McKinsey and Company: New York,
NY, USA, 2017; p. 80.
7. Young, D.; Panthi, K.; Noor, O. Challenges involved in adopting BIM on the construction jobsite. EPiC Ser. Built Environ. 2021, 2,
302–310.
8. Delgado, J.M.D.; Oyedele, L. Digital Twins for the built environment: Learning from conceptual and process models in manufac-
turing. Adv. Eng. Inf. 2021, 49, 101332. [CrossRef]
9. Bello, S.A.; Oyedele, L.O.; Akinade, O.O.; Bilal, M.; Delgado, J.M.D.; Akanbi, L.A.; Ajayi, A.O.; Owolabi, H.A. Cloud computing
in construction industry: Use cases, benefits and challenges. Autom. Construct. 2021, 122, 103441. [CrossRef]
10. Delgado, J.M.D.; Oyedele, L. Deep learning with small datasets: Using autoencoders to address limited datasets in construction
management. Appl. Soft Comput. 2021, 112, 107836. [CrossRef]
11. Gbadamosi, A.; Mahamadu, A.; Manu, P.; Akinade, O.; Sierra, F.; Lam, T.T.; Alzaatreh, A. A BIM based approach for optimization
of construction and assembly through material selection. In Proceedings of the Creative Construction, Ljubljana, Slovenia, 30
June–3 July 2018; pp. 955–962. [CrossRef]
12. Gbadamosi, A.-Q.; Mahamadu, A.-M.; Oyedele, L.O.; Akinade, O.O.; Manu, P.; Mahdjoubi, L.; Aigbavboa, C. Offsite construction:
Developing a BIM-Based optimizer for assembly. J. Clean. Prod. 2019, 215, 1180–1190. [CrossRef]
13. Zaher, M.; Greenwood, D.; Marzouk, M. Mobile augmented reality applications for construction projects. Construct. Innov. 2018,
18, 152–166. [CrossRef]
14. Mrosla, L.; Koch, V.; Both, P.V. Quo vadis AI in Architecture? Survey of the current possibilities of AI in the architectural practice.
Des. Artif. Intell. 2019, 2, 45–54. [CrossRef]
15. Steenson, M.W. Architectural Intelligence: How Designers, and Architects Created the Digital Landscape; The MIT Press: Cambridge,
MA, USA, 2017.
16. Poh, C.; Ubeynarayana, C.U.; Goh, Y.M. Safety leading indicators for construction sites: A machine learning approach. Autom.
Constr. 2018, 93, 375–386. [CrossRef]
17. Palacios, A.T.; Cordero, J.M.E.B.; Bello, M.R.; Palacios, E.T.; Martínez González, J.L. New applications of 3D SLAM on risk
management using unmanned aerial vehicles in the construction industry. In Drones-Applications; Dekoulis, G., Ed.; IntechOpen:
London, UK, 2018; Available online: https://www.intechopen.com/chapters/58901 (accessed on 25 July 2023).
18. Wei, C.; Li, Y. Design of energy consumption monitoring and energy-saving management system of intelligent building based on
the Internet of things. In Proceedings of the International Conference on Electronics, Communications and Control (ICECC),
Ningbo, China, 9–11 September 2011; pp. 3650–3652. [CrossRef]
19. Lanko, A.; Vatin, N.; Kaklauskas, A. Application of RFID combined with blockchain technology in logistics of construction
materials. In Proceedings of the Business Technologies for Sustainable Urban Development (MATEC), Saint Petersburg, Russia,
20–22 December 2017. [CrossRef]
20. Chen, H.-P.; Ying, K.-C. Artificial Intelligence in the Construction Industry: Main Development Trajectories and Future Outlook.
Appl. Sci. 2022, 12, 5832. [CrossRef]
21. Regona, M.; Yigitcanlar, T.; Xia, B.; Li, R.Y.M. Opportunities and Adoption Challenges of AI in the Construction Industry: A
PRISMA Review. J. Open Innov. Technol. Mark. Complex. 2022, 8, 45. [CrossRef]
22. Shukla, A.K.; Janmaijaya, M.; Abraham, A.; Muhuri, P.K. Engineering applications of artificial intelligence: A bibliometric analysis
of 30 years (1988–2018). Eng. Appl. Artif. Intell. 2019, 85, 517–532. [CrossRef]
Sensors 2023, 23, 8740 28 of 35

23. Gorodnova, N.V. Application of Artificial Intelligence and Nanotechnology in the Investment and Construction Sector in Russia.
Vestnik NSUEM 2021, 3, 81–95. [CrossRef]
24. Xin, X.; Tu, Y.; Stojanovic, V.; Wang, H.; Shi, K.; He, S.; Pan, T. Online reinforcement learning multiplayer non-zero sum games of
continuous-time Markov jump linear systems. Appl. Math. Comput. 2022, 412, 126537. [CrossRef]
25. The Future of AI: Construction-Grade Artificial Intelligence. Available online: https://onekeyresources.milwaukeetool.com/en/
ai-in-construction (accessed on 27 June 2023).
26. Horst, D.J.; de Andrade Júnior, P.P. Sustainability of the Steel Industry: A Systematic Review. Biointerface Res. Appl. Chem. 2023,
13, 525. [CrossRef]
27. Erdélyi, D.; Kern, Z.; Nyitrai, T.; Hatvani, I.G. Predicting the spatial distribution of stable isotopes in precipitation using a machine
learning approach: A comparative assessment of random forest variants. GEM 2023, 14, 14. [CrossRef]
28. Frehywot, S.; Vovides, Y. An equitable and sustainable community of practice framework to address the use of artificial intelligence
for global health workforce training. Hum. Resour. Health 2023, 21, 45. [CrossRef]
29. Baghdadi, A.; Lama, S.; Singh, R.; Sutherland, G.R. Tool-tissue force segmentation and pattern recognition for evaluating
neurosurgical performance. Sci. Rep. 2023, 13, 9591. [CrossRef] [PubMed]
30. Wei, Y.; Han, C.; Yu, Z. An environment safety monitoring system for agricultural production based on artificial intelligence,
cloud computing and big data networks. J. Cloud Comput. 2023, 12, 83. [CrossRef]
31. Cheng, Q.; Chunhong, Z.; Qianglin, L. Development and application of random forest regression soft sensor model for treating
domestic wastewater in a sequencing batch reacto. Sci. Rep. 2023, 13, 9149. [CrossRef]
32. Marwaha, J.S.; Raza, M.M.; Kvedar, J.C. The digital transformation of surgery. Digit. Med. 2023, 6, 103. [CrossRef]
33. Burgueño, A.M.; Aldana-Martín, J.F.; Vázquez-Pendón, M.; Barba-González, C.; Jiménez Gómez, Y.; García Millán, V.; Navas-
Delgado, I. Scalable approach for high-resolution land cover: A case study in the Mediterranean Basin. J. Big Data 2023, 10, 91.
[CrossRef]
34. Babović, Z.; Bajat, B.; Barac, D.; Bengin, V.; Ðokić, V.; Ðord̄ević, F.; Drašković, D.; Filipović, N.; French, S.; Furht, B.; et al. Teaching
computing for complex problems in civil engineering and geosciences using big data and machine learning: Synergizing four
different computing paradigms and four different management domains. J. Big Data 2023, 10, 89. [CrossRef]
35. Guinot, L.; Ando, K.; Takahashi, S.; Iwata, H. Analysis of implicit robot control methods for joint task execution. ROBOMECH J.
2023, 10, 12. [CrossRef]
36. Lew, A.J.; Jin, K.; Buehler, M.J. Designing architected materials for mechanical compression via simulation, deep learning, and
experimentation. Comput. Mater. 2023, 9, 80. [CrossRef]
37. Sidlauskiene, J.; Joye, Y.; Auruskeviciene, V. AI-based chatbots in conversational commerce and their effects on product and price
perceptions. Electron. Mark. 2023, 33, 24. [CrossRef] [PubMed]
38. Prosekov, A.; Rada, A.; Kuznetsov, A.; Osintseva, M.; Petrik, N. Features of using unmanned aircraft as an element of digitalization
of industry in the Siberian federal district of Russia. AIP Conf. Proc. 2022, 2632, 020037. [CrossRef]
39. Hsu, H.-C. A study on computer vision for facial emotion recognition. Sci. Rep. 2023, 13, 8425. [CrossRef]
40. Zayed, S.M.; Attiya, G.; El-Sayed, A.; Sayed, A.; Hemdan, E.E.-D. An Efficient Fault Diagnosis Framework for Digital Twins Using
Optimized Machine Learning Models in Smart Industrial Control Systems. Int. J. Comput. Intell. Syst. 2023, 16, 69. [CrossRef]
41. Rada, A.O.; Kuznetsov, A.D. Digital inventory of agricultural land plots in the Kemerovo Region. Foods Raw Mater. 2022, 10,
206–215. [CrossRef]
42. Obianyo, J.I.; Udeala, R.C.; Alaneme, G.U. Application of neural networks and neuro-fuzzy models in construction scheduling.
Sci. Rep. 2023, 13, 8199. [CrossRef] [PubMed]
43. Yan, X.; Liu, X.; Zhao, C.; Chen, G.-Q. Applications of synthetic biology in medical and pharmaceutical fields. Signal Transduct.
Target. Ther. 2023, 8, 199. [CrossRef] [PubMed]
44. Bačiulienė, V.; Bilan, Y.; Navickas, V.; Lubomír, C. The Aspects of Artificial Intelligence in Different Phases of the Food Value and
Supply Chain. Foods 2023, 12, 1654. [CrossRef]
45. Sicard, J.; Barbe, S.; Boutrou, R.; Bouvier, L.; Delaplace, G.; Lashermes, G.; Théron, L.; Vitrac, O.; Tonda, A. A primer on predictive
techniques for food and bioresources transformation processes. J. Food Process Eng. 2023, 46, e14325. [CrossRef]
46. Kryukov, K.M.; Metlev, A.M. Possibilities of using artificial intelligence in construction Engineering. Bull. Don. 2022, 10, 1–10. (In
Russian)
47. Warszawski, A.; Sangrey, D.A. Robotics in building construction. J. Constr. Eng. Manag. 1985, 111, 260–280. [CrossRef]
48. Warszawski, A. Robots in the construction industry. Robotica 1986, 4, 181–188. [CrossRef]
49. Adachi, R.; Mogi, S.; Tsuji, H.; Nakano, T.; Shiozaki, H.; Ohtani, H.; Ohba, K.; Shinohara, H.; Matsui, H.; Jitsukawa, T.; et al.
Application of intelligent robot arc-welding system to large-sized steel construction. Hitachi Zosen Tech. Rev. 1985, 46, 32–39.
50. Hirokazu, K.S.; Nakada, T.; Akagi, F. Method of optimal work assignment for robot’s hand used in construction. Nippon. Kikai
Gakkai Ronbunshu C Hen/Trans. Jpn. Soc. Mech. Eng. Part. C 1983, 49, 243–249. [CrossRef]
51. Paulson, B.C. Automation and robotics for construction. J. Constr. Eng. Manag. 1985, 111, 190–207. [CrossRef]
52. Levitt, R.E.; Kartam, N.A.; Kunz, J.C. Artificial intelligence techniques for generating construction project plans. J. Constr. Eng.
Manag. 1988, 114, 329–343. [CrossRef]
53. McGartland, M.R.; Hendrickson, C.T. Expert systems for construction project monitoring. J. Constr. Eng. Manag. 1985, 111,
293–307. [CrossRef]
Sensors 2023, 23, 8740 29 of 35

54. Furuta, H.; Tu, K.-S.; Yao, J.T.P. Structural engineering applications of expert systems. Comput. Aided Des. 1985, 17, 410–419.
[CrossRef]
55. Ashley, D.B.; Levitt, R.E. Expert systems in construction: Work in progress. J. Comput. Civ. Eng. 1987, 1, 303–311. [CrossRef]
56. Williams, T.P.; Kangari, R. Knowledge-based system for planning a construction operation. Microcomput. Civ. Eng. 1988, 3,
345–353. [CrossRef]
57. Saleh, M.A.E. Automation of quantity surveying in construction projects. J. Archit. Eng. 1999, 5, 141–148. [CrossRef]
58. Scdano, B.G.; Cerrolaza, M.; Alarcon, E. QL-CONST1: An expert system for quality level prediction in concrete structures. Civ.
Eng. Syst. 1988, 5, 206–212. [CrossRef]
59. Hirayama, Y. Automatic design system for custom-made houses. Fujitsu Sci. Tech. J. 1981, 17, 27–45.
60. Althoff, K.-D.; Kratz, N.; Richter, M.M.; Spieker, P. Planning systems and artificial intelligence. Ann. Oper. Res. 1988, 16, 1–12.
[CrossRef]
61. Bowen, P.A.; Edwards, P.J. Uncertainty and incomplete information in the evaluation of building projects. Invest. Anal. J. 1985, 14,
11–21. [CrossRef]
62. Hendrickson, C.; Zozaya-Gorostiza, C.; Rehak, D.; Baracco-Miller, E.; Lim, P. Expert system for construction planning. J. Comput.
Civ. Eng. 1987, 1, 253–269. [CrossRef]
63. Vladimirova, I.L.; Bareshenkova, K.A. Digital engineering in the field procurement in the implementation of investment and
construction projects. Econ. Entrep. Law 2020, 10, 377–394. (In Russian) [CrossRef]
64. Mamadjanov, A.M.; Yusupov, S.M.; Sadirov, S. Advantages and the Future of CNC Machines. Sci. Prog. 2021, 2, 1638–1647.
65. How Does Artificial Intelligence Impact the Construction Industry? Available online: https://requestum.com/blog/artificial-
intelligence-in-construction-industry (accessed on 2 July 2023).
66. Artificial Intelligence in the Construction Industry. Available online: https://thinkml.ai/artificial-intelligence-in-the-construction-
industry (accessed on 10 July 2023).
67. Ghosh, T.; Martinsen, K. Generalized approach for multi-response machining process optimization using machine learning and
evolutionary algorithms. Eng. Sci. Technol. 2020, 23, 650–663. [CrossRef]
68. Reinventing Construction through a Productivity Revolution. Available online: https://www.mckinsey.com/capabilities/
operations/our-insights/reinventing-construction-through-a-productivity-revolution (accessed on 10 July 2023).
69. Wan, S.; Li, D.; Gao, J.; Roy, R.; He, F. A collaborative machine tool maintenance planning system based on content management
technologies. Int. J. Adv. Manuf. Technol. 2018, 94, 1639–1653. [CrossRef]
70. Jamwal, A.; Agrawal, R.; Sharma, M.; Kumar, A.; Kumar, V.; Garza-Reyes, J.A.A. Machine learning applications for sustainable
manufacturing: A bibliometric-based review for future research. J. Enterp. Inf. Manag. 2022, 35, 566–596. [CrossRef]
71. Chen, C.; Liu, Y.; Kumar, M.; Qin, J.; Ren, Y. Energy consumption modelling using deep learning embedded semi-supervised
learning. Comput. Ind. Eng. 2019, 135, 757–765. [CrossRef]
72. Nasir, V.; Sassani, F. A review on deep learning in machining and tool monitoring: Methods, opportunities, and challenges. Int. J.
Adv. Manuf. Technol. 2021, 115, 2683–2709. [CrossRef]
73. Moore, J.; Stammers, J.; Dominguez-Caballero, J. The application of machine learning to sensor signals for machine tool and
process health assessment. Proc. Inst. Mech. Eng. B J. Eng. Manuf. 2021, 235, 1543–1557. [CrossRef]
74. Kaewunruen, S.; Xu, N. Digital Twin for Sustainability Evaluation of Railway Station Buildings. Front. Built Environ. 2018, 4, 77.
[CrossRef]
75. Kaewunruen, S.; Sresakoolchai, J.; Huang, J.; Zhu, Y.; Ngamkhanong, C.; Remennikov, A.M. Machine Learning Based Design of
Railway Prestressed Concrete Sleepers. Appl. Sci. 2022, 12, 10311. [CrossRef]
76. Sresakoolchai, J.; Kaewunruen, S. Integration of building information modeling (BIM) and artificial intelligence (AI) to detect
combined defects of infrastructure in the railway system. In Resilient Infrastructure. Lecture Notes in Civil Engineering; Kolathayar,
S., Ghosh, C., Adhikari, B.R., Pal, I., Mondal, A., Eds.; Springer Nature: Singapore, 2021; Volume 202, pp. 377–386. [CrossRef]
77. Alkaissy, M.; Arashpour, M.; Golafshani, E.M.; Hosseini, M.R.; Khanmohammadi, S.; Bai, Y.; Feng, H. Enhancing construction
safety: Machine learning-based classification of injury types. Saf. Sci. 2023, 162, 106102. [CrossRef]
78. Anwer, S.; Li, H.; Umer, W.; Antwi-Afari, M.F.; Mehmood, I.; Yu, Y.; Haas, C.; Wong, A. Yu Lok Identification and Classification
of Physical Fatigue in Construction Workers Using Linear and Nonlinear Heart Rate Variability Measurements. J. Constr. Eng.
Manag. 2023, 149, 04023057. [CrossRef]
79. Luo, X.; Li, X.; Goh, Y.M.; Song, X.; Liu, Q. Application of machine learning technology for occupational accident severity
prediction in the case of construction collapse accidents. Saf. Sci. 2023, 163, 106138. [CrossRef]
80. Chern, W.-C.; Hyeon, J.; Nguyen, T.V.; Asari, V.K.; Kim, H. Context-aware safety assessment system for far-field monitoring.
Autom. Constr. 2023, 149, 104779. [CrossRef]
81. Chen, W.; Li, C.; Guo, H. A lightweight face-assisted object detection model for welding helmet use. Expert. Syst. Appl. 2023, 2211,
119764. [CrossRef]
82. Dadashi Haji, M.; Behnam, B.; Sebt, M.H.; Ardeshir, A.; Katooziani, A. BIM-Based Safety Leading Indicators Measurement Tool
for Construction Sites. Int. J. Civ. Eng. 2023, 21, 265–282. [CrossRef]
83. Ramadan, A.S.; Elgendi, E.O. A review of optimization techniques and algorithms used for FRP applications in civil engineering.
J. Eng. Appl. Sci. 2023, 70, 61. [CrossRef]
Sensors 2023, 23, 8740 30 of 35

84. Wang, Y.; Dai, F.; Jia, R.; Wang, R.; Sharifi, H.; Wang, Z. A novel combined intelligent algorithm prediction model for the tunnel
surface settlement. Sci. Rep. 2023, 13, 9845. [CrossRef]
85. Sánchez-Garrido, A.J.; Navarro, I.J.; García, J.; Yepes, V. A systematic literature review on modern methods of construction in
building: An integrated approach using machine learning. J. Build. Eng. 2023, 7315, 106725. [CrossRef]
86. Wu, X.; Lu, G.; Wu, Z. Remote sensing technology in the construction of digital twin basins: Applications and prospects. Water
2023, 15, 2040. [CrossRef]
87. Wang, S.; Ma, Q.; Liang, H. System simulation of land use spatial planning method and environment management strategy
analysis by using machine vision. Soft Comput. 2023, 27, 5985–5994. [CrossRef]
88. Baduge, S.K.; Thilakarathna, S.; Perera, J.S.; Arashpour, M.; Sharafi, P.; Teodosio, B.; Shringi, A.; Mendis, P. Artificial intelligence
and smart vision for building and construction 4.0: Machine and deep learning methods and applications. Autom. Constr. 2022,
2022, 104440. [CrossRef]
89. Sayed, Y.A.K.; Ibrahim, A.A.; Tamrazyan, A.G.; Fahmy, M.F.M. Machine-learning-based models versus design-oriented models
for predicting the axial compressive load of FRP-confined rectangular RC columns. Eng. Struct. 2023, 28515, 116030. [CrossRef]
90. Jia, Y.; Wang, J.; Shou, W.; Hosseini, M.R.; Bai, Y. Graph neural networks for construction. Autom. Constr. 2023, 154, 104984.
[CrossRef]
91. Wang, D.; Chang, F. Application of machine learning-based BIM in green public building design. Soft Comput. 2023, 13, 9031–9040.
[CrossRef]
92. Mowafy, N.; El Zayat, M.; Marzouk, M. Parametric BIM-based life cycle assessment framework for optimal sustainable design. J.
Build. Eng. 2023, 7515, 106898. [CrossRef]
93. Al-Musawi, R.; Naimi, S. Evaluation of Construction Project’s Cost Using BIM Technology. Math. Model. Eng. Probl. 2023, 10,
469–476. [CrossRef]
94. Vakaj, E.; Cheung, F.; Cao, J.; Tawil, A.-R.H.; Patlakas, P. An ontology-based cost estimation for offsite construction. J. Inf. Technol.
Constr. 2023, 28, 220–245. [CrossRef]
95. Simić, N.; Ivanišević, N.; Nedeljković, Ð.; Senić, A.; Stojadinović, Z.; Ivanović, M. Early Highway Construction Cost Estimation:
Selection of Key Cost Drivers. Sustainability 2023, 15, 5584. [CrossRef]
96. Maaouane, M.; Chennaif, M.; Zouggar, S.; Krajačić, G.; Amrani, S.; Zahboune, H. Cost-effective design of energy efficiency
measures in the building sector in North Africa using Building Information Modeling. EnergyBuild 2023, 2941, 113283. [CrossRef]
97. Ranjbar, N.; Balali, A.; Valipour, A.; Pignatta, G.; Wei, S. Identification and prioritization of energy consumption optimization
strategies in the building industry using the hybrid swara-bim model. J. Green Build. 2023, 18, 37–69. [CrossRef]
98. Zhang, L.; Yao, H.; Fu, Y.; Chen, Y. Comparing subjective and objective measurements of contract complexity in influencing
construction project performance: Survey versus machine learning. J. Manag. Eng. 2023, 39, 04023017. [CrossRef]
99. Ottoni, A.L.; Novo, M.S.; Costa, D.B. Deep Learning for vision systems in Construction 4.0: A systematic review. Signal Image
Video Process 2023, 17, 1821–1829. [CrossRef]
100. Alsugair, A.M.; Alsanabani, N.M.; Al-Gahtani, K.S. Forecasting the Final Contract Cost on the Basis of the Owner’s Cost
Estimation Using an Artificial Neural Network. Buildings 2023, 13, 786. [CrossRef]
101. Qiu, Y.; Chen, H. A systematic review of the knowledge domain of institutional theory in construction project management. Eng.
Constr. Archit. Manag. 2023, 30, 2523–2544. [CrossRef]
102. Salahaldain, Z.; Naimi, S.; Alsultani, R. Estimation and Analysis of Building Costs Using Artificial Intelligence Support Vector
Machine. Math. Model. Eng. Probl. 2023, 10, 405–411. [CrossRef]
103. Xu, Y.; Zhou, Y.; Sekula, P.; Ding, L. Machine learning in construction: From shallow to deep learning. Dev. Built Environ. 2021, 6,
100045. [CrossRef]
104. Wang, L.; Cheng, Y.; Zhang, Y. Exploring the risk propagation mechanisms of supply chain for prefabricated building projects. J.
Build. Eng. 2023, 741, 106771. [CrossRef]
105. Koc, K.; Ekmekcioğlu, Ö.; Işık, Z. Developing a probabilistic decision-making model for reinforced sustainable supplier selection.
Int. J. Prod. Econ. 2023, 259, 108820. [CrossRef]
106. Sadatnya, A.; Sadeghi, N.; Sabzekar, S.; Khanjani, M.; Tak, A.N.; Taghaddos, H. Machine learning for construction crew
productivity prediction using daily work reports. Autom. Constr. 2023, 152, 104891. [CrossRef]
107. Geng, S.; Luo, Q.; Liu, K.; Li, Y.; Hou, Y.; Long, W. Research status and prospect of machine learning in construction 3D printing.
Case Stud. Constr. Mater. 2023, 18, e01952. [CrossRef]
108. Ahila, P.R.; Arivazhagan, S.; Arun, M. Crack recognition on concrete structures based on machine crafted and hand crafted
features. Expert. Syst. Appl. 2023, 228, 120447. [CrossRef]
109. Alsakka, F.; El-Chami, I.; Yu, H.; Al-Hussein, M. Computer vision-based process time data acquisition for offsite construction.
Autom. Constr. 2023, 149, 104803. [CrossRef]
110. Zou, M.; Yu, J.; Lv, Y.; Lu, B.; Chi, W.; Sun, L. A Novel Day-to-Night Obstacle Detection Method for Excavators Based on Image
Enhancement and Multisensor Fusion. IEEE Sens. J. 2023, 10, 10825–10835. [CrossRef]
111. Abu Kharmeh, S.; Natsheh, E.; Sulaiman, B.; Abuabiah, M.; Tarapiah, S. Indoor WiFi-Beacon Dataset Construction Using
Autonomous Low-Cost Robot for 3D Location Estimation. Appl. Sci. 2023, 13, 6768. [CrossRef]
112. Oke, A.E.; Aliu, J.; Oluwasefunmi Fadamiro, P.; Akanni, P.O.; Stephen, S.S. Attaining digital transformation in construction: An
appraisal of the awareness and usage of automation techniques. J. Build. Eng. 2023, 6715, 105968. [CrossRef]
Sensors 2023, 23, 8740 31 of 35

113. Wu, C.; Li, X.; Jiang, R.; Guo, Y.; Wang, J.; Yang, Z. Graph-based deep learning model for knowledge base completion in constraint
management of construction projects. Comput. Aided Civ. Infrastruct. Eng. 2023, 38, 702–719. [CrossRef]
114. Wen, H.; Liu, L.; Zhang, J.; Hu, J.; Huang, X. A hybrid machine learning model for landslide-oriented risk assessment of
long-distance pipelines. J. Environ. Manag. 2023, 342, 118177. [CrossRef]
115. Lin, P.; Wu, M.; Zhang, L. Probabilistic safety risk assessment in large-diameter tunnel construction using an interactive and
explainable tree-based pipeline optimization method. Appl. Soft Comput. 2023, 143, 110376. [CrossRef]
116. Wuni, I.Y.; Shen, G.Q.; Antwi-Afari, M.F. Exploring the design risk factors for modular integrated construction projects. Constr.
Innov. 2023, 23, 213–228. [CrossRef]
117. Osorio-Sandoval, C.A.; Crick, G.; Collinge, W.H.; Farghaly, K.; Mosleh, M.H.; Manu, P.; Cheung, C.M. Dataset of characterized
construction safety risks and related treatments. Data Brief 2023, 48, 109293. [CrossRef]
118. Zheng, Y.; Zhang, Y.; Lin, J. BIM–based time-varying system reliability analysis for buildings and infrastructures. J. Build. Eng.
2023, 761, 106958. [CrossRef]
119. Eastman, C. The use of computers instead of drawings in building design. Art 1975, 63, 46–50.
120. Sacks, R.; Girolami, M.; Brilakis, I. Building information modelling, artificial intelligence and construction tech. Dev. Built Environ.
2020, 4, 100011. [CrossRef]
121. Zhang, Y. Safety management of civil engineering construction based on artificial intelligence and machine vision technology.
Adv. Civ. Eng. 2021, 2021, 3769634. [CrossRef]
122. Construction Safety Statistics for 2022. Available online: https://www.bigrentz.com/blog/construction-safety-statistics (accessed
on 30 May 2023).
123. Safety+Health. Available online: www.safetyandhealthmagazine.com (accessed on 30 May 2023).
124. Health and Safety Statistics. Available online: www.hse.gov.uk (accessed on 30 May 2023).
125. Bigham, G.F.; Adamtey, S.; Onsarigo, L.; Jha, N. Artificial intelligence for construction safety: Mitigation of the risk of fall. In
Intelligent Systems and Applications (IntelliSys 2018); Arai, K., Kapoor, S., Bhatia, R., Eds.; Springer International Publishing: Cham,
Switzerland, 2019; Volume 2, pp. 1024–1037.
126. Razi, P.Z.; Sulaiman, S.K.; Ali, M.I.; Ramli, N.I.; Saad, M.S.H.; Jamaludin, O.; Doh, S.I. How Artificial Intelligence Changed the
Construction Industry in Safety Issues. IOP Conf. Ser. Earth Environ. Sci. 2023, 1140, 012004. [CrossRef]
127. Ayhan, B.U.; Tokdemir, O.B. Safety assessment in megaprojects using artificial intelligence. Saf. Sci. 2019, 118, 273–287. [CrossRef]
128. Gheraibia, Y.; Kabir, S.; Aslansefat, K.; Sorokos, I.; Papadopoulos, Y. Safety+ AI: A novel approach to update safety models using
artificial intelligence. IEEE Access 2019, 7, 135855–135869. [CrossRef]
129. Yeon-Chul, K.; Wi Seong, Y.; Yoonseok, S. Application of artificial neural networks to prediction of construction safety accidents.
J. Korean Soc. Hazard. Mitig. 2017, 17, 7–14.
130. Li, R.Y.M.; Leung, T.H. Computer vision and hybrid reality for construction safety risks: A pilot study. In Fourth International
Congress on Information and Communication Technology (ICICT 2019); Springer Nature: Singapore, 2020; Volume 2, pp. 17–22.
131. Choi, J.; Gu, B.; Chin, S.; Lee, J.S. Machine learning predictive model based on national data for fatal accidents of construction
workers. Autom. Constr. 2020, 110, 102974. [CrossRef]
132. Wu, J.; Cai, N.; Chen, W.; Wang, H.; Wang, G. Automatic detection of hardhats worn by construction personnel: A deep learning
approach and benchmark dataset. Autom. Constr. 2019, 106, 102894. [CrossRef]
133. Fang, W.; Zhong, B.; Zhao, N.; Love, P.E.; Luo, H.; Xue, J.; Xu, S. A deep learning-based approach for mitigating falls from height
with computer vision: Convolutional neural network. Adv. Eng. Inform. 2019, 39, 170–177.
134. Fang, W.; Ding, L.; Luo, H.; Love, P.E. Falls from heights: A computer vision-based approach for safety harness detection. Autom.
Constr. 2018, 91, 53–61.
135. Huang, X.; Wu, X.; Usmani, A. Perspectives of using artificial intelligence in building fire safety. In Handbook of Cognitive and
Autonomous Systems for Fire Resilient Infrastructures; Springer International Publishing: Cham, Switzerland, 2022; pp. 139–159.
136. Zhang, J.; Zi, L.; Hou, Y.; Wang, M.; Jiang, W.; Deng, D. A deep learning-based approach to enable action recognition for
construction equipment. Adv. Civ. Eng. 2020, 2020, 8812928. [CrossRef]
137. Zhu, C.; Zhu, J.; Bu, T.; Gao, X. Monitoring and Identification of Road Construction Safety Factors via UAV. Sensors 2022, 22, 8797.
[CrossRef]
138. 140Mohan, M.; Varghese, S. Artificial intelligence enabled safety for construction sites. Int. Res. J. Eng. Technol. 2019, 6, 1–6.
139. Fang, W.; Love, P.E.; Luo, H.; Ding, L. Computer vision for behaviour-based safety in construction: A review and future directions.
Adv. Eng. Inform. 2020, 43, 100980. [CrossRef]
140. Guo, B.H.; Zou, Y.; Fang, Y.; Goh, Y.M.; Zou, P.X. Computer vision technologies for safety science and management in construction:
A critical review and future research directions. Saf. Sci. 2021, 135, 105130. [CrossRef]
141. Tay, J.; Shi, P.; He, Y.; Nath, T. Application of computer vision in the construction industry. In Proceedings of the 10th World
Construction Symposium, Colombo, Sri Lanka, 24–26 June 2022; Sandanayake, Y.G., Gunatilake, S., Waidyasekara, K.G.A.S., Eds.;
pp. 125–137. Available online: http://dl.lib.uom.lk/handle/123/19960 (accessed on 30 May 2023). [CrossRef]
142. Anysz, H.; Foremny, A.; Kulejewski, J. Comparison of ANN classifier to the neuro-fuzzy system for collusion detection in the
tender procedures of road construction sector. IOP Conf. Ser. Mater. Sci. Eng. 2020, 471, 112064. [CrossRef]
143. Sarkar, D.; Dhaneshwar, D.; Raval, P. Automation in Monitoring of Construction Projects through BIM-IoT-Blockchain Model. J.
Inst. Eng. (India) Ser. A 2023, 104, 317–333. [CrossRef]
Sensors 2023, 23, 8740 32 of 35

144. Pan, Y.; Zhang, L. Integrating BIM and AI for Smart Construction Management: Current Status and Future Directions. Arc.
Comput. Methods Eng. 2023, 30, 1081–1110. [CrossRef]
145. Cocco, L.; Tonelli, R.; Marchesi, M. A System Proposal for Information Management in Building Sector Based on BIM, SSI, IoT
and Blockchain. Future Internet 2022, 14, 140. [CrossRef]
146. Ke, W. The construction of enterprise’s financial supply chain management under blockchain technology. Expert Syst. 2023; in
press. [CrossRef]
147. Liu, H.; Han, S.H.; Zhu, Z. Blockchain Technology toward Smart Construction: Review and Future Directions. J. Constr. Eng.
Manag. 2023, 149, 03123002. [CrossRef]
148. Zhang, X.; Liu, T.; Rahman, A.; Zhou, L. Blockchain Applications for Construction Contract Management: A Systematic Literature
Review. J. Constr. Eng. Manag. 2023, 149, 03122011. [CrossRef]
149. Fu, Y.; Xu, C.; Zhang, L.; Chen, Y. Control, coordination, and adaptation functions in construction contracts: A machine-coding
model. Autom. Constr. 2023, 152, 104890. [CrossRef]
150. Shuai, B. A rationale-augmented NLP framework to identify unilateral contractual change risk for construction projects. Comput.
Ind. 2023, 149, 103940. [CrossRef]
151. Cheng, J.C.P.; Liu, H.; Gan, V.J.L.; Das, M.; Tao, X.; Zhou, S. Construction cost management using blockchain and encryption.
Autom. Constr. 2023, 152, 104841. [CrossRef]
152. Zimon, G.; Nakonieczny, J.; Chudy-Laskowska, K.; Wójcik-Jurkiewicz, M.; Kochański, K. An Analysis of the Financial Liquidity
Management Strategy in Construction Companies Operating in the Podkarpackie Province. Risks 2022, 10, 5. [CrossRef]
153. Dimitriou, L.; Marinelli, M.; Fragkakis, N. Early bill-of-quantities estimation of concrete road bridges: An artificial intelligence-
based application. Public Work. Manag. Policy 2018, 23, 127–149. [CrossRef]
154. Juszczyk, M. The challenges of nonparametric cost estimation of construction works with the use of artificial intelligence tools.
Procedia Eng. 2017, 196, 415–422. [CrossRef]
155. Peng, H.; Wu, H.; Wang, J.; Dede, T. Research on the prediction of the water demand of construction engineering based on the BP
neural network. Adv. Civ. Eng. 2020, 2020, 8868817. [CrossRef]
156. Lin, X.; Lu, Y. Research on mathematical model of cost budget in the early stage of assembly construction project based on
improved neural network algorithm. Math. Probl. Eng. 2020, 2020, 3674929. [CrossRef]
157. Beljkaš, Ž.; Knežević, M.; Rutešić, S.; Ivanišević, N. Application of artificial intelligence for the estimation of concrete and
reinforcement consumption in the construction of integral bridges. Adv. Civ. Eng. 2020, 2020, 8645031. [CrossRef]
158. Gondia, A.; Siam, A.; El-Dakhakhni, W.; Nassar, A.H. Machine learning algorithms for construction projects delay risk prediction.
J. Constr. Eng. Manag. 2020, 146, 04019085. [CrossRef]
159. Minli, Z.; Shanshan, Q. Research on the application of artificial neural networks in tender offer for construction projects. Phys.
Procedia 2012, 24, 1781–1788. [CrossRef]
160. Kim, J.M.; Bae, J.; Son, S.; Son, K.; Yum, S.G. Development of model to predict natural disaster-induced financial losses for
construction projects using deep learning techniques. Sustainability 2021, 13, 5304. [CrossRef]
161. Sharma, S.; Ahmed, S.; Naseem, M.; Alnumay, W.S.; Singh, S.; Cho, G.H. A survey on applications of artificial intelligence for
pre-parametric project cost and soil shear-strength estimation in construction and geotechnical engineering. Sensors 2021, 21, 463.
[CrossRef]
162. Mojtahedi, S.F.; Ahmadihosseini, A.; Sadeghi, H. An Artificial Intelligence Based Data-Driven Method for Forecasting Unconfined
Compressive Strength of Cement Stabilized Soil by Deep Mixing Technique. Geotech. Geol. Eng. 2023, 41, 491–514. [CrossRef]
163. Jiang, Y.; Bai, Y.; Han, S. Determining ground elevations covered by vegetation on construction sites using drone-based orthoimage
and convolutional neural network. J. Comput. Civ. Eng. 2020, 34, 04020049. [CrossRef]
164. Wang, X.Q.; Chen, P.; Chow, C.L.; Lau, D. Artificial-intelligence-led revolution of construction materials: From molecules to
Industry 4.0. Matter 2023, 6, 1831–1859. [CrossRef]
165. Zhong, B.; Wu, H.; Ding, L.; Love, P.E.; Li, H.; Luo, H.; Jiao, L. Mapping computer vision research in construction: Developments,
knowledge gaps and implications for research. Autom. Constr. 2019, 107, 102919. [CrossRef]
166. Practical Application of BIM Technologies on the Example of a Moscow Developer. Available online: https://vc.ru/life/321724
-prakticheskoe-primenenie-bim-tehnologiy-na-primere-moskovskogo-developera (accessed on 10 July 2023).
167. Cao, Y.; Zandi, Y.; Agdas, A.S.; Wang, Q.; Qian, X.; Fu, L.; Wakil, K.; Selmi, A.; Issakhov, A.; Roco Videla, Á. A review study of
application of artificial intelligence in construction management and composite beams. Steel Compos. Struct. 2021, 39, 685–700.
168. Elhegazy, H.; Chakraborty, D.; Elzarka, H.; Ebid, A.M.; Mahdi, I.M.; Aboul Haggag, S.Y.; Abdel Rashid, I. Artificial intelligence
for developing accurate preliminary cost estimates for composite flooring systems of multi-storey buildings. J. Asian Archit. Build.
Eng. 2022, 21, 120–132. [CrossRef]
169. Rahimian, F.P.; Seyedzadeh, S.; Oliver, S.; Rodriguez, S.; Dawood, N. On-demand monitoring of construction projects through a
game-like hybrid application of BIM and machine learning. Autom. Constr. 2020, 110, 103012. [CrossRef]
170. Braun, A.; Borrmann, A. Combining inverse photogrammetry and BIM for automated labeling of construction site images for
machine learning. Autom. Constr. 2019, 106, 102879. [CrossRef]
171. Jiang, Y.; Bai, Y. Estimation of construction site elevations using drone-based orthoimagery and deep learning. J. Constr. Eng.
Manag. 2020, 146, 04020086. [CrossRef]
Sensors 2023, 23, 8740 33 of 35

172. Li, D.; Cong, A.; Guo, S. Sewer damage detection from imbalanced CCTV inspection data using deep convolutional neural
networks with hierarchical classification. Autom. Constr. 2019, 101, 199–208. [CrossRef]
173. Pandian, A.P. Artificial intelligence application in smart warehousing environment for automated logistics. J. Artif. Intell. 2019, 1,
63–72.
174. Klumpp, M. Automation and artificial intelligence in business logistics systems: Human reactions and collaboration requirements.
Int. J. Logist. Res. Appl. 2018, 21, 224–242. [CrossRef]
175. Oleśków-Szłapka, J.; Wojciechowski, H.; Domański, R.; Pawłowski, G. Logistics 4.0 maturity levels assessed based on GDM
(grey decision model) and artificial intelligence in logistics 4.0-trends and future perspective. Procedia Manuf. 2019, 39, 1734–1742.
[CrossRef]
176. Strandhagen, J.O.; Vallandingham, L.R.; Fragapane, G.; Strandhagen, J.W.; Stangeland, A.B.H.; Sharma, N. Logistics 4.0 and
emerging sustainable business models. Adv. Manuf. 2017, 5, 359–369. [CrossRef]
177. Zhang, Y. The application of artificial intelligence in logistics and express delivery. J. Phys. Conf. Ser. 2019, 1325, 012085. [CrossRef]
178. Woschank, M.; Rauch, E.; Zsifkovits, H. A review of further directions for artificial intelligence, machine learning, and deep
learning in smart logistics. Sustainability 2020, 12, 3760. [CrossRef]
179. Arabi, S.; Haghighat, A.; Sharma, A. A deep-learning-based computer vision solution for construction vehicle detection. Comput.-
Aided Civ. Infrastruct. Eng. 2020, 35, 753–767. [CrossRef]
180. Rashid, K.M.; Louis, J. Times-series data augmentation and deep learning for construction equipment activity recognition. Adv.
Eng. Inform. 2019, 42, 100944. [CrossRef]
181. Qian, X.; Papadonikolaki, E. Shifting trust in construction supply chains through blockchain technology. Eng. Constr. Archit.
Manag. 2021, 28, 584–602. [CrossRef]
182. Xu, S.; Wang, J.; Shou, W.; Ngo, T.; Sadick, A.M.; Wang, X. Computer vision techniques in construction: A critical review. Arch.
Comput. Methods Eng. 2021, 28, 3383–3397. [CrossRef]
183. Ercan, M.F.; Wang, R.B. Computer Vision-Based Inspection System for Worker Training in Build and Construction Industry.
Computers 2022, 11, 100. [CrossRef]
184. Kwiatek, C.; Sharif, M.; Li, S.; Haas, C.; Walbridge, S. Impact of augmented reality and spatial cognition on assembly in
construction. Autom. Constr. 2019, 108, 102935. [CrossRef]
185. Rada, A.O.; Kuznetsov, A.D.; Zverev, R.E.; Timofeev, A.E. Automation of monitoring construction works based on laser scanning
from unmanned aerial vehicles. Nanotechnol. Constr. 2023, 15, 373–382. [CrossRef]
186. Digital Transformation in the Construction and Engineering Industry—Part Two. Available online: https://pwc.blogs.com/
industry_perspectives/2017/05/digital-transformation-in-the-construction-and-engineering-industry-part-two.html (accessed
on 30 May 2023).
187. Digital Twins and Digital Doubles: Populating our Virtual Worlds. Available online: https://www.futuregrasp.com/digital-
twins-and-digital-doubles (accessed on 30 May 2023).
188. China Is Using AI and 3D Printing to Build a 590-Foot-Tall Dam without the Need for Human Workers, Scientists Say. Available
online: https://www.businessinsider.com/china-robots-ai-3d-print-dam-no-human-labor-2022-5 (accessed on 10 July 2023).
189. China: Artificial Intelligence Used to Monitor Workers’ Safety on Construction Sites. Available online: https://www.business-
humanrights.org/en/latest-news/china-artificial-intelligence-used-to-monitor-workers-safety-on-construction-sites (accessed
on 10 July 2023).
190. How Construction Control Is Carried Out with the Help of IAS “Monitoring”: The Experience of RosKapStroy. Available online:
https://digital-build.ru/kak-proishodit-czifrovoj-kontrol-stroitelstva-pri-pomoshhi-ias-monitoring-opyt-roskapstroj (accessed
on 10 July 2023).
191. SITE SAFETY AT SIEMENS →→→SIEMENS Contracted DEKRA to Develop and Implement a Comprehensive Safety Concept
in Accordance with International Standards for the Construction of a New Digital Factory in Nanjing, China. Available online:
https://gb2021.dekra-online.de/en/kickstart/service-divisions/consulting (accessed on 9 July 2023).
192. Construction Site Security. Video Monitoring and Security Solutions Built for the Construction Industry. Available online:
https://calibercommunications.ca/construction-security (accessed on 10 July 2023).
193. Newly Developed Safety Monitoring System for Construction Sites. Available online: https://indiaeducationdiary.in/newly-
developed-safety-monitoring-system-for-construction-sites (accessed on 10 July 2023).
194. Enhancing Workers’ Safety-Towards Automated Industrial Safety Monitoring Systems. Available online: https://social-
innovation.hitachi/en-in/knowledge-hub/viewpoint/industrial-safety-monitoring-systems (accessed on 10 July 2023).
195. Artificial Intelligence the Way to The fully Automated Construction Site. Available online: https://openexperience.de/en/
themen/Kuenstliche-Intelligenz-im-Bau (accessed on 10 July 2023).
196. The Rise of Artificial Intelligence in Construction Industry. Available online: https://gharpedia.com/blog/artificial-intelligence-
in-construction (accessed on 10 July 2023).
197. Artificial Intelligence in Construction. Examples of AI for the Construction Industry. Available online: https://elport.ru/articles/
iskusstvennyiy_intellekt_v_stroitelstve_primeryi_ii_dlya_stroitelnoy_otrasli (accessed on 10 July 2023).
198. Digital Technologies in Construction. Available online: https://tadviser.com/index.php/Article:Digital_Technologies_in_
Construction (accessed on 10 July 2023).
Sensors 2023, 23, 8740 34 of 35

199. Inspiring BIM Projects You Need to See. Available online: https://constructionblog.autodesk.com/successful-bim-projects/
(accessed on 10 July 2023).
200. Building Routes to Economic Growth and Prosperity. Available online: https://www.bechtel.com/services/infrastructure/civil-
infrastructure (accessed on 10 July 2023).
201. Application of BIM in Steel Structures Design. Available online: https://steelexplained.com/application-of-bim-in-steel-
structures-design/ (accessed on 10 July 2023).
202. DUBAI METRO 2020. Available online: https://www.acciona.com/projects/dubai-metro-2020/?_adin=0924833216 (accessed on
10 July 2023).
203. Siebert, S.; Teizer, J. Mobile 3D mapping for surveying earthwork projects using an Unmanned Aerial Vehicle (UAV) system.
Autom. Constr. 2014, 41, 1–14. [CrossRef]
204. Digital Tools in Road Construction. Available online: https://innovation.strabag.com/en/project/digital-tools-in-road-
construction (accessed on 10 July 2023).
205. Driving Digitalisation in Road Construction. Available online: https://www.dwtc.com/en/industry-insights/driving-
digitalisation-in-road-construction (accessed on 10 July 2023).
206. Road and Bridge Digital Twins in Action: Four Case Studies. Available online: https://www.cadalyst.com/collaboration/digital-
twin/road-and-bridge-digital-twins-action-four-case-studies-75827 (accessed on 10 July 2023).
207. China’s AI Implementation is Edging Ahead of the US. Available online: https://www.forbes.com/sites/craigsmith/2023/01/14
/chinas-ai-implementation-is-edging-ahead-of-the-us/?sh=7dc2e92c2dfb (accessed on 10 July 2023).
208. Roberts, H.; Cowls, J.; Morley, J.; Taddeo, M.; Wang, V.; Floridi, L. The Chinese approach to artificial intelligence: An analysis of
policy, ethics, and regulation. AI Soc. 2021, 36, 59–77. [CrossRef]
209. Artificial Intelligence (AI) in China’s Construction Industry (2016–2025)-Analysis on AI Spending with 15+ KPIs, Market Size
& Forecast Across 6+ Application Segments, AI Domains, and Technology. Available online: https://www.businesswire.
com/news/home/20190329005250/en/Artificial-Intelligence-AI-in-Chinas-Construction-Industry-2016-2025{-}{-}-Analysis-
on-AI-Spending-with-15-KPIs-Market-Size-Forecast-Across-6-Application-Segments-AI-Domains-and-Technology{-}{-}-
ResearchAndMarkets.com (accessed on 10 July 2023).
210. Yan, J.-K.; Zheng, Z.; Zhou, Y.-C.; Lin, J.-R.; Deng, Y.-C.; Lu, X.-Z. Recent Research Progress in Intelligent Construction: A
Comparison between China and Developed Countries. Buildings 2023, 13, 1329. [CrossRef]
211. Martinez, P.; Al-Hussein, M.; Ahmad, R. A scientometric analysis and critical review of computer vision applications for
construction. Autom. Constr. 2019, 107, 102947. [CrossRef]
212. Oprach, S.; Bolduan, T.; Steuer, D.; Vössing, M.; Haghsheno, S. Building the future of the construction industry through artificial
intelligence and platform thinking. Digit. Welt 2018, 3, 40–44. [CrossRef]
213. McAleenan, P. Moral responsibility and action in the use of artificial intelligence in construction. Proc. Inst. Civ. Eng. Manag.
Procure. Law 2020, 173, 166–174. [CrossRef]
214. Alekseeva, T.R. BIM technologies and artificial intelligence in infrastructural construction. BST Bull. Constr. Equip. 2022, 1, 52–54.
(In Russian)
215. Khanzode, K.C.A.; Sarode, R.D. Advantages and disadvantages of artificial intelligence and machine learning: A literature review.
Int. J. Libr. Inform. Sci. 2020, 9, 30–36.
216. Soori, M.; Arezoo, B.; Habibi, M. Dimensional and geometrical errors of three-axis CNC milling machines in a virtual machining
system. Comput. Aided Des. 2013, 45, 1306–1313. [CrossRef]
217. Soori, M.; Arezoo, B.; Habibi, M. Tool deflection error of three-axis computer numerical control milling machines, monitoring and
minimizing by a virtual machining system. J. Manuf. Sci. Eng. 2016, 138, 081005. [CrossRef]
218. Yang, Y.; Hu, T.; Ye, Y.; Gao, W.; Zhang, C. A knowledge generation mechanism of machining process planning using cloud
technology. J. Ambient. Intell. Hum. Comput. 2019, 10, 1081–1092. [CrossRef]
219. Shishkunova, D.V.; Ishhenko, A.V. Construction Production Logistics: Problems and Solutions. Inzhenernyj Vestn. Dona 2020, 1.
Available online: http://Ivdon.ru/ru/magazine/archive/N1y2020/6292 (accessed on 2 September 2023).
220. Chui, M.; Manyika, J.; Miremadi, M.; Henke, N.; Chung, R.; Nel, P.; Malhotra, S. Notes from the AI Frontier: Applications and
Value of Deep Learning. Available online: https://www.mckinsey.com/featured-insights/artificial-intelligence/notes-from-the-
ai-frontier-applications-and-value-of-deep-learning (accessed on 10 July 2023).
221. Pandey, K.; Datta, S. Hot machining of difficult-tocut materials: A review. Mater. Today Proc. 2021, 44, 2710–2715. [CrossRef]
222. Shokrani, A.; Betts, J. A new hybrid minimum quantity lubrication system for machining difficult-tocut materials. CIRP Ann.
2020, 69, 73–76. [CrossRef]
223. Zhao, J.; Liu, Z.; Wang, B.; Hu, J. PVD AlTiN coating effects on tool-chip heat partition coefficient and cutting temperature rise in
orthogonal cutting Inconel 718. Int. J. Heat. Mass. Transf. 2020, 163, 120449. [CrossRef]
224. Pradhananga, P.; Elzomor, M.; Santi Kasabdji, G. Identifying the challenges to adopting robotics in the us construction industry. J.
Constr. Eng. Manag. 2021, 147, 05021003. [CrossRef]
225. Marzouk, M.; Azab, S.; Metawie, M. BIM-based approach for optimizing life cycle costs of sustainable building. J. Clean. Prod.
2018, 188, 217–226. [CrossRef]
226. Bagherzadeh, A.; Budak, E. Investigation of machinability in turning of difficult-to-cut materials using a new cryogenic cooling
approach. Tribol. Int. 2018, 119, 510–520. [CrossRef]
Sensors 2023, 23, 8740 35 of 35

227. De Galarreta, S.R.; Jeffers, J.R.; Ghouse, S. A validated finite element analysis procedure for porous structures. Mater. Des. 2020,
189, 108546. [CrossRef]
228. Bykov, I.A. Artificial intelligence as a source of political judgments. J. Political Stud. 2020, 4, 23–33.
229. Zhilin, V.V.; Safaryan, O.A. Artificial intelligence in data storage systems. Bull. Don. State Tech. Univ. 2020, 20, 196–200. [CrossRef]
230. Arroyo, P.; Schöttle, A.; Christensen, R. The Ethical and Social. Dilemma of AI Uses in The Construction Industry. In Proceedings
of the 29th Annual Conference of the International Group for Lean Construction (IGLC), Lima, Peru, 12–18 July 2021. [CrossRef]
231. Ilvitskaya, S.V.; Lobkova, T.V. Green housing architecture and Green BIM technologies. Archit. Constr. Russ. 2018, 1, 108–113.
232. Karaguzel, U.; Olgun, U.; Uysal, E.; Budak, E.; Bakkal, M. Increasing tool life in machining of difficultto-cut materials using
nonconventional turning processes. Int. J. Adv. Manuf. Technol. 2015, 77, 1993–2004. [CrossRef]
233. Parida, A.K.; Maity, K. Study of machinability in heat-assisted machining of nickel-base alloy. Measurement 2021, 170, 108682.
[CrossRef]
234. Hegab, H.; Umer, U.; Soliman, M.; Kishawy, H.A. Effects of nano-cutting fluids on tool performance and chip morphology during
machining Inconel 718. Int. J. Adv. Manuf. Technol. 2018, 96, 3449–3458. [CrossRef]
235. Soori, M.; Arezoo, B. A Review in Machining-Induced Residual Stress. J. New Technol. Mater. 2022, 12, 64–83.
236. Marinelli, M. From Industry 4.0 to Construction 5.0: Exploring the Path towards Human–Robot Collaboration in Construction.
Systems 2023, 11, 152. [CrossRef]
237. Soori, M.; Asmael, M.; Solyalı, D. Recent Development in Friction Stir Welding Process: A Review. SAE Int. J. Mater. Manuf. 2021,
14, 63–80. [CrossRef]
238. Păvăloaia, V.-D.; Necula, S.-C. Artificial Intelligence as a Disruptive Technology—A Systematic Literature Review. Electronics
2023, 12, 1102. [CrossRef]
239. Syed, A.S.; Sierra-Sosa, D.; Kumar, A.; Elmaghraby, A. IoT in smart cities: A survey of technologies, practices and challenges.
Smart Cities 2021, 4, 429–475. [CrossRef]
240. Alahi, M.E.E.; Sukkuea, A.; Tina, F.W.; Nag, A.; Kurdthongmee, W.; Suwannarat, K.; Mukhopadhyay, S.C. Integration of IoT-
Enabled Technologies and Artificial Intelligence (AI) for Smart City Scenario: Recent Advancements and Future Trends. Sensors
2023, 23, 5206. [CrossRef]
241. Nikitas, A.; Michalakopoulou, K.; Njoya, E.T.; Karampatzakis, D. Artificial Intelligence, Transport and the Smart City: Definitions
and Dimensions of a New Mobility Era. Sustainability 2020, 12, 2789. [CrossRef]
242. Summary of AI Provisions from the National Defense Authorization Act 2023. Available online: https://hai.stanford.edu/
summary-ai-provisions-national-defense-authorization-act-2023 (accessed on 25 September 2023).
243. European Parliamentary Research Service. Available online: https://www.europarl.europa.eu/RegData/etudes/STUD/2020/6
34452/EPRS_STU(2020)634452_EN.pdf (accessed on 25 September 2023).
244. Gaumond, E.; Régis, C. Assessing Impacts of AI on Human Rights: It’s Not Solely About Privacy and Nondiscrimination. Avail-
able online: https://www.lawfaremedia.org/article/assessing-impacts-of-ai-on-human-rights-it-s-not-solely-about-privacy-
and-nondiscrimination (accessed on 25 September 2023).

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.