Automation in Construction 131 (2021) 103917

Contents lists available at ScienceDirect

Automation in Construction
journal homepage: www.elsevier.com/locate/autcon

Review

Integrating disruptive technologies with facilities management: A literature

review and future research directions
Marco Marocco *, Ilaria Garofolo
Department of Civil Engineering and Architecture, University of Trieste, Trieste, Italy

A R T I C L E I N F O A B S T R A C T

Keywords: Facilities Management (FM) has received increasing attention from the Architecture, Engineering, Construction
Building information modelling (BIM) and Operation (AECO) industry over the last decade. Disruptive technologies have the potential to advance the
Digital twin (DT) Operation and Maintenance (O&M) phase in different application fields. With the increasing industry interest,
Facilities management (FM)
there is a need to review the current status of research developments in FM. In this review, 70 journal articles
Operation & maintenance (O&M)
published from 2011 to 2020 were reviewed. This paper aims to provide a comprehensive review of the ap­
Literature review
plications of disruptive technologies for FM, analyse research trends and identify research gaps and potential
future research directions. This article focuses on only academic articles including topics, such as operations and
maintenance, information management, emergency management and energy management. Eventually, it is
hoped that this review will provide researchers with clarity of where research endeavours are most needed and
helpful insights to address FM challenges.

1. Introduction Building information modelling (BIM) has had a sharp imple­

mentation in the Architectural, Engineering, Construction and Opera­
The International Facility Management Association defines Facilities tion (AECO) industry over the last twenty years, especially in the design
Management (FM) as “a profession that encompasses multiple disci­ and construction phases [7]. BIM can be regarded as a virtual process
plines to ensure functionality, comfort, safety and efficiency of the built that encompasses information in a single, virtual and shared model,
environment by integrating people, place, process and technology” [1]. allowing all building stakeholders to communicate and collaborate in an
The facility management phase is the longest phase in the building life effective way [8]. However, since buildings are dynamic environments
cycle and has a huge impact on the whole life of buildings by generating where changes occur every day, the added value of BIM for FM was
the overwhelming majority of expenses [2]. Indeed, design and con­ considered marginal [9]. There is a need to combine up-to-date and
struction phases usually account only for less than 15% of the total life- living data with static information to effectively support FM tasks. No
cycle cost of buildings [3], while the operational phase accounts for integration of information hampers decision-making processes, which
more than 60% of the building lifecycle for both cost and time [4]. become time-consuming and less intuitive [10].
To boost FM decision-making processes, a series of tasks, such as Digital Twin (DT) is a “digital model, which is a dynamic represen­
collection, storage, retrieval and analysis of data, is necessary. However, tation of an asset and mimics its real-world behaviour” [11]. DT includes
traditional FM software systems, such as Computer Maintenance and the model of the physical object, data from the object and a unique one-
Management System (CMMS) and Computer Aided Facility Manage­ to-one correspondence to the object [11]. BIM is regarded as the basis
ment (CAFM), cannot deal effectively with FM [5]. These technologies for the construction of digital twins [12], but needs data integration
usually exploit Computer Aided Design (CAD) systems, which cannot from other resources. Web-service platforms can exploit unlimited and
store and manage data as a centralised and unique repository and do not dynamic data storage space and achieve the objective of connecting
allow instant updates of all sources when some parts are modified. various kinds of data sources, avoiding separated and scattered data
Therefore, there is a fragmentation of data which is stored both in storage. Integrating cloud computing, BIM and IoT technologies pro­
different types of digital applications and in sheets of papers and/or file vides high fidelity operable dataset in real-time [13] and allows con­
notes [6]. ducting advanced analysis through Artificial Intelligence (AI) agents

* Corresponding author at: Piazzale Europa, 1, 34127 Trieste TS, Italy.

E-mail address: marco.marocco2@phd.units.it (M. Marocco).

https://doi.org/10.1016/j.autcon.2021.103917
Received 30 November 2020; Received in revised form 13 August 2021; Accepted 19 August 2021
Available online 27 August 2021
0926-5805/© 2021 Elsevier B.V. All rights reserved.
M. Marocco and I. Garofolo Automation in Construction 131 (2021) 103917

[14]. Other potential disruptive technologies, such as virtual reality therefore a threshold of both Citescore1 of the year 2019 and CiteScore
(VR), mixed reality (MR) and augmented reality (AR), can be used as Tracker of the year 2020 greater than or equals to 1.5 was set up. This
powerful tools for interacting with facilities [15]. choice allows analysing high-quality articles, which are peer-reviewed
With the increasing industry interest, there is a great need of ana­ carefully and present comprehensive results, and avoiding preliminary
lysing the potential applications of disruptive technologies for FM. Thus, results and hypothetical frameworks that may not have been validated
understanding the current situation and development in cutting-edge yet. As a result, 389 papers were discarded. Then, articles’ abstract were
technology and FM integration includes: analysed and if relevant, articles’ full text were skimmed. Articles
considered irrelevant were discarded. In addition, a few research articles
- What are the main application fields for FM? discovered by using the snowballing2 process were added. Eventually,
- What are the limitations that still remain in integrating disruptive the screening process turned out to 70 papers, as shown in Fig. 1.
technologies with FM activities?
- What are the research gaps and future research directions?
2.1. Bibliometric analysis

This paper aims to provide a comprehensive review of the current

The bibliometric analysis shows that most of the 70 reviewed journal
status and insights in supporting FM by analysing existing studies,
publications were produced in 15 countries, as shown in Fig. 2. Since
examining current limitations and identifying future research directions.
some of the analysed papers were written in collaboration among more
The scope of this article includes only academic articles, which deal with
researchers of different countries, the total amount of documents pro­
information management, energy management, emergency manage­
duced by countries is 91. Thus, these 15 countries account for 89% of
ment and maintenance management of buildings during the operational
this total amount. In particular, the most active countries are the UK
phase. Eventually, 70 research papers were reviewed over the last ten
(10), USA (9), China (9), Taiwan (9) and Canada (8), which account for
years. The rest of the paper is structured as follows: Section 2 focuses on
nearly 50% of the total amount of documents produced by countries.
the research methodology, Section 3 groups the reviewed articles into
These figures confirm the trend of interest of countries, which are more
four main application fields, Section 4 identifies research gaps and
advanced in terms of technology and processes, along with more expe­
future directions and Section 5 presents the conclusion.
rience in these fields. For example, the UK has proposed the Digital Built
Britain strategy for the development of the Level 3 BIM strategy, which
2. Methodology
will unlock significant savings and knowledge for assets and FM [19].
With reference to subject areas of papers, the bibliometric analysis re­
This study analyses the existing literature on the integration of
veals that Engineering was the most representative area, followed by
disruptive technologies with FM with a focus on recent applications in
Computer Science and Social Sciences, as shown in Fig. 3. The other
the operational phase. The bibliometric analysis conducted in this study
involved subject areas included Environmental Science, Physics and
aims to support the research by providing statistical data of the reviewed
Astronomy, Material Science, Decision Sciences, Chemistry and Energy.
research articles and potential insights about the application fields of
With the help of some scripts by using Pandas, which is an open
FM.
source data analysis and manipulation tool built, the bibliometric
The review process [16], consists of four main steps: (1) defining the
analysis also shows that the largest amount of reviewed papers were
research database and identifying keywords; (2) selecting, collecting
published in “Automation in Construction” with 26 out of the total
and discarding papers for content review; (3) in-depth analysis in Sec­
number of identified articles, as shown in Fig. 4 and Table 1. This was
tion 3 and (4) discussing the identified research gaps and future di­
followed by “Facilities” and “Advance Engineering Informatics”. In
rections in Section 4.
addition to this main group, which accounts for more than 50% of all
The method used to search articles is based on combining keywords,
reviewed papers, other articles were analysed from a group of five
such as “BIM” or “Building information model*”, “digital twin” and
journals including “Building and Environment”, “Applied Science
“smart buildings” with other keywords such as “FM”, “facilit* manag*”,
(Switzerland)”, “Visualisation in Engineering”, “Sensors (Switzerland)”
“operations and maintenance”, “asset management”, “emergency man­
and “Buildings”. The rest of the papers derived from journals where a
agement”, energy management”, “space management”. The initial
single publication was reviewed. Each reviewed research paper was
journal search identified 3677 articles by using the Scopus database,
published in sources with high citation impact, such as CiteScore, as
where duplicated papers were excluded.
shown in Fig. 5.
The data collection phase was critical to determine a set of articles
A keywords analysis was conducted by using CiteSpace, which is a
which allowed conducting a comprehensive review of the recent ap­
Java application for trends and pattern visualisation, and demonstrated
plications of the integration of disruptive technologies with FM, there­
that Building Information Modelling and FM were the most frequently
fore some features and limitation parameters of the research were
used keywords in the analysed papers, as shown in Fig. 6. In particular,
defined. First of all, Scopus was chosen as the main database where to
the figure shows that keywords can be divided into four main clusters,
search articles. Secondly, since the research scope focuses on the latest
which identify the four defined application fields, and three terms,
advancement, articles reviewed in this paper include only those pub­
namely building, building information modelling and facilities man­
lished in the last decade, namely from 2011 to 2020 (Table 1). Thirdly,
agement, that are connected and central for each cluster. Each word in
only papers written in English were reviewed. Lastly, only research ar­
the figure is linked with other related ones and its size and links to other
ticles were taken into consideration. After applying the above
words are displayed based on two features including occurrence fre­
mentioned filters, the journal search identified 1129 articles. Then, a
quency and centrality among words. Therefore, the more a keyword is
first check was conducted on articles’ title and keywords, discarding
frequent and central for a cluster, the more this keyword is big and have
articles which were considered irrelevant. As a result, 670 papers were
strong connections. Thanks to this keywords analysis, a preliminary idea
selected.
Only journals with high impact in the AECO industry were analysed,
1
CiteScore: “is the number of citations to documents (articles, reviews,
conference papers, book chapters, and data papers) by a journal over four years,
divided by the number of the same document types indexed in Scopus published
in those same four years” [17].
2
Snowballing: is a process based on retrieving papers by using citations of
relevant papers [18].

2
M. Marocco and I. Garofolo Automation in Construction 131 (2021) 103917

Table 1
Documents by year.
Source Title Year Total

2011 2013 2014 2015 2016 2017 2018 2019 2020

Advanced Engineering Informatics 1 1 0 1 1 0 1 0 0 5

Applied Energy 0 1 0 0 0 0 0 0 0 1
Applied Sciences (Switzerland) 0 0 0 0 0 0 1 1 0 2
Automation in Construction 1 1 1 1 0 4 5 7 6 26
Building and Environment 0 0 0 1 0 1 1 0 0 3
Buildings 0 0 0 1 0 0 0 1 0 2
Built Environment Project and Asset Management 0 0 0 1 0 0 0 0 0 1
Computers in Industry 0 0 0 0 0 0 0 1 0 1
Egyptian Informatics Journal 0 0 0 0 0 0 0 0 1 1
Energies 0 0 0 0 0 1 0 0 0 1
Facilities 0 0 1 0 0 0 0 5 0 6
Fire Safety Journal 0 0 0 0 0 1 0 0 0 1
IEEE Access 0 0 0 0 0 0 0 1 0 1
IEEE Sensors Journal 0 1 0 0 0 0 0 0 0 1
IEEE Transactions on Industrial Informatics 0 0 0 0 0 1 0 0 0 1
IT Professional 0 0 0 0 0 1 0 0 0 1
International Journal of Building Pathology and Adaptation 0 0 0 0 0 0 0 1 0 1
International Journal of Distributed Sensor Networks 0 0 0 0 0 0 0 0 1 1
International Journal of Precision Engineering and Manufacturing 0 0 0 0 1 0 0 0 0 1
Journal of Civil Engineering and Management 0 0 0 0 0 0 1 0 0 1
Journal of Computing in Civil Engineering 0 0 0 0 1 0 0 0 0 1
Journal of Management in Engineering 0 0 1 0 0 0 0 0 0 1
Journal of Structural Integrity and Maintenance 0 0 0 0 0 0 1 0 0 1
Proceedings of the IEEE 0 0 1 0 0 0 0 0 0 1
Scientific World Journal 0 0 1 0 0 0 0 0 0 1
Sensors (Switzerland) 0 0 0 0 0 0 0 1 1 2
Structure and Infrastructure Engineering 0 1 0 0 0 0 0 0 0 1
Sustainable Cities and Society 0 1 0 0 0 0 0 0 0 1
The Scientific World Journal 0 1 0 0 0 0 0 0 0 1
Visualisation in Engineering 0 0 0 1 1 0 0 0 0 2
Total 2 7 5 6 4 9 10 18 9 70

of the several application subfields which are connected to the four main information system which linked IFC objects, COBie data and historical
defined fields of application was generated and took into consideration FM work data together to enhance searching information efficiency. IFC
in the in-depth analysis. files were converted to the Web Ontology Language (OWL) by using the
resource description framework (RFD) and conserved BIM-based data
3. In-depth analysis of applications relationships. On the whole, what derives from this improvement is the
possibility to reduce costly and time-consuming inspections when these
To provide a comprehensive analysis of the reviewed papers, four are not extremely necessary.
main application fields were identified including information manage­ On the other hand, programmed and scheduled inspections are part
ment, maintenance management, energy management and emergency of maintenance contract obligations. To exploit this aspect efficiently,
management. The overall organisation of the reviewed papers is sum­ other studies focused on retrieving and tracking information on site by
marised in Table 2. For each of the fields, representative research papers exploiting different devices directly linked to BIM models. For example,
are analysed and discussed in this section. Lin et al. [6] proposed the mobile automated BIM-based facility man­
agement (BIMFM) system based on 2D barcode technology to acquire
3.1. Information management updated information to BIM models. One of the problems with this
system relates to the properties of 2D barcode labels, which are easily
Defects such as scattered information, limited communication damageable and can work only for short read distance. For these rea­
channels and lack of historical records, need solving by using a sons, Valero et al. [22] presented a system based on RFID tags attached
comprehensive platform able to integrate data [20]. Potential savings to furniture, where RFID readers identified elements thanks to their IDs,
for facility owners are generated whether collecting, storing and which were stored in FM and BIM databases. Likewise, Costin & Teizer
exploiting information is performed over facility life-cycle [4]. Current [23] developed an algorithm for real time resource location tracking in
research on disruptive technologies for information management fo­ indoor environments by using marketable passive RFID and linking each
cuses on localising and tracking building components, storing asset in­ tagged object in BIM models to a unique RFID ID tag. The algorithm
formation, and advanced visualisation and interaction with facilities differed from previous studies because it incorporated BIM model data
information. directly and used multilateration techniques that allowed accuracy
within 3 m of the actual location. However, the proposed system can
3.1.1. Localising & tracking building components work properly with fixed assets, but encountered problems dealing with
Localising building components effectively is a fundamental but also moveable ones. To solve this issue, Motamedi et al. [24] proposed the
challenging task when conducting FM activities. Traditionally, this task Cluster-based Movable Tag Localisation (CMTL) method based on RFID
is performed on-site by FM personnel, who rely on paper-based projects technology, which clustered reference tags that were spatially close.
and documents or on their personal knowledge and experience [2]. This method did not use wired Real-Time Location Systems (RTLS), but
However, since BIM elements are identified by a Globally Unique radio signals, which were propagated by fixed tags attached to fixed
Identifier (GUID) or Universal Unique Identifier (UUID), building ele­ assets. Omni-directional identical long-range tags were used to store
ments and their features can be filtered and searched remotely. Along data, such as position of assets, and stored data in XML format so as to
these lines, Kim et al. [21] proposed a semantic web-based FM simplify the mapping of BIM data. Another relevant solution was carried

3
M. Marocco and I. Garofolo Automation in Construction 131 (2021) 103917

Fig. 1. Flow diagram of articles’ collection.

4
M. Marocco and I. Garofolo Automation in Construction 131 (2021) 103917

Fig. 2. Documents by country or territory up to 15 countries.

Fig. 3. Documents by subject area.

out by Rashid et al. [25], who developed a point-and-click device based specific nomenclature in order to update, retrieve and collect them
on BIM and IoT integration to locate electric devices in buildings by easily [2]. Associating assets with their documents in databases is a
using an ultra-wideband (UWB)-based indoor positioning system and an critical task, but in most cases this process is conducted manually and
inertial measurement unit (IMU) for acquiring information of position therefore error-prone and time-consuming [2]. For this reason, Heaton
and orientation of users. To extend the application of localising building et al. [28] proposed a methodology to extract data from BIM models and
components, Baek et al. [26] presented an AR system which exploited transfer them directly into a relational database of asset management
BIM models to identify and visualise indoor position and orientation of systems by creating an asset classification system. However, FM data­
sanitary pipes for on-site maintenance. The system used an image-based bases are often fixed and do not allow improving data schemas ac­
indoor localisation which compares 2D BIM images with the user’s cording to user’ perspective without redeveloping the system. To this
perspective view by using the VGG pre-trained convolutional neural end, Kang & Choi [29] proposed an extendible database based on BIM
network (CNN). On balance, it can be noticed that combining BIM and metadata to integrate FM and BIM data of a research centre. The system
sensing technologies is the common denominator when dealing with exploited the extract, transform and load (ETL) process to generate a
localising and tracking building elements. Among these automatic data warehouse. Overall, it can be observed that an ongoing and auto­
identification and data capture technologies, it seems noticeable that matic process of information handover is an essential characteristic for
long-range wireless solutions are preferred above the others due to their collecting and storing building data effectively.
peculiarities and promising potentialities. Pervasive sensing technology can be also used to help FM personnel
to store maintenance records in databases [30]. Several studies have
3.1.2. Storing asset information been conducted on this topic over the last decade. For example, Zhan
Searching for specific documents and storing data are common daily et al. [31] proposed a system based on BIM and QR integration, which
operations which usually lead to a huge waste of time [27]. Assets data exploited deep-learning image classification algorithms to automatically
and information need to be stored in digital formats and catalogued with classify images according to their content. After conducting inspection

5
M. Marocco and I. Garofolo Automation in Construction 131 (2021) 103917

Fig. 4. Documents by journal.

Fig. 5. CiteScore publication by year.

tasks, the images of assets were collected by using smartphones and 3.1.3. Advanced visualisation and interaction with facilities information
classified and stored in an image database within a BIM knowledge re­ Advanced visualisation technologies have received a growing
pository. Motamedi et al. [5] proposed a prototype application which attention in the construction sector over the last decade due to its
mapped RFID components in BIM models by extending the IFC schema operational capability to support FM activities [36]. Exploiting graphic
and stored information in RFID tags associated with the BIM database. visualisation of buildings can provide clearer view and comprehension
Similarly, Motamedi et al. [32] proposed a framework based on RFID of facility information [27]. To this end, González et al. [37] presented
and BIM technology which used ifcXML format to store data and an indoor navigation support system based on ifcOWL ontology for
exploited a multi-level encryption based on an access control policy navigating BIM models by using Web-based technologies. The system
(ACP) along with role-based access control. Another relevant study was allowed BIM enquiry and navigation from any kind of device without
carried out by Pärn & Edwards [33], who developed the FinDD appli­ needing a specific software installed. Another relevant research was
cation programming interface plug-in for Revit which can store and conducted by Neuville et al. [38], who presented a RESTful web appli­
extract building data. A problem with this method concerns the usage of cation based on a 3D viewpoint management algorithm for improving
proprietary formats. To solve this issue, Lee et al. [34] developed the success rate and accuracy of visual counting of bad condition assets. The
V3DM+ web-based system, which exploited an API library for accessing application aimed to find out the most efficient camera position and
BIM models, to allow importing, processing and visualising data on the orientation and took into account four quality indicators including
internet. Along these lines and with reference to managing a great success rate in performing visual task, accuracy of visual task, certainty
amount of data exchange, Halmetoja et al. [35] presented the conditions degree of visual task and speed in carrying out visual task. On the whole,
data model (CDM) approach, which combined BIM with active data it is worth noticing that FM tools which do not rely on locally installed
generated by IoT technology to create the digital twin of the facility. On systems but can be used by using web applications are preferred to other
balance, what seems apparent is that using automatic identification and solutions due to their characteristics of availability and accessibility to
data capture technologies capable of associating assets with their doc­ building information.
uments in databases automatically and continuously is a common de­ Using BIM models as a basis for displaying and interacting with
nominator of development for storing asset information. available information is a valuable method, but other technologies can

6
M. Marocco and I. Garofolo Automation in Construction 131 (2021) 103917

Fig. 6. Keyword clustering.

Synchronisation (BVRS) system based on the metadata interpretation

Table 2
system of a BIM database, and a cloud-based infrastructure, which
Summary of the reviewed papers by fields of application.
aimed to automatically update BIM models and visualise them in VR in
Fields of Subfields Related research real time. Other research integrated more than one type of visual­
application
isations. For example, Patti et al. [43] presented an Android-based
Information Localising and tracking [5,6,15,21–26,28–35,37,38,40–45] mobile application to visualise 3D BIM models in AR and VR, which
Management building components
allowed storing and querying physical environmental parameters
Storing asset information
Advanced visualisation correlated to models. Hou et al. [44] proposed the Augmented Visual­
and interaction with ised Plant Management (AVPM) system prototype based on AR, AV and
facilities information information modelling (IM) integration to facilitate locating faulty
Maintenance Work order management [10,46–53,55–61,63–65] components and accessing data for a petroleum refinery. To boost real-
Management Decision-making
processes for
time collaboration between field worker and facility manager, El
maintenance Ammari & Hammad [45] proposed the Collaborative Markless Mixed
Detecting asset faults and Reality Facility Management System (CBIM3R-FMS) prototype based on
inspecting building BIM and Mixed Reality (MR) integration where facility managers see
assets
what field workers see both in a synchronised and non-synchronised
Predictive maintenance
Energy Real-time energy [69–76,78–87] ways. Overall, it can be seen that although there is no common prac­
Management monitoring tice regarding when using one kind of technology rather than the others,
Assessing and optimising using BIM in combination with diverse advanced visualisation tech­
energy building nologies can be considered the main denominator for improving the
performance
interaction and usability of building information.
Emergency Emergency response and [89,90,92,94–98]
Management path optimisation
Hazards monitoring
3.2. Maintenance management

improve this task. To enhance the visualisation of BIM models several

Building maintenance encompasses all the tasks which are critical to
studies have been focused on visual information provided by Augmented allow an efficient operation of facilities. Four critical topics in this field
Reality (AR), Virtual Reality (VR), Augmented Virtuality (AV) and
are discussed, including work order management, decision-making
Mixed Reality (MR). These systems allow FM personnel to perform processes for maintenance, detecting assets faults and inspecting
operation and maintenance tasks more efficiently than traditional
building assets, and predictive maintenance.
methods based on manual and phone assistance [39] [15]. For instance,
Irizarry et al. [40] proposed the conceptual ambient intelligent (AmI)
3.2.1. Work order management
environment based on BIM mobile AR (MAR) technology and aerial
drones. Williams et al. [41] presented the BIM2MAR application based
3.2.1.1. Tracking work order information. Work order (WO) stores in­
on BIM and AR to visualise data on mobile technologies for FM activities
formation about maintenance and repair works of facilities. FM
in a healthcare environment by using the Web Graphics Library through
personnel can exploit WO to generate feedback about problems and
JavaScript. Similarly, Du et al. [42] proposed the BIM-VR Real time
solutions [27]. However, tracking work orders needs appropriate

7
M. Marocco and I. Garofolo Automation in Construction 131 (2021) 103917

systems and methods to be carried out without losing valuable infor­ Almarshad [51] proposed an integrated system based on BIM and the
mation. Traditionally, WO data is collected by using paper or digital Case-Based Reasoning (CBR) knowledge-based technique to support
documents which are not linked with other facility information, e.g. preventive and corrective maintenance decisions for building mainte­
space, component and system, making complicated any kind of perfor­ nance in public organisations. Integrating BIM and CBR library by using
mance comparison. To this end, Lin & Su [46] proposed the BIM-based an IFC protocol, the system stored and managed building maintenance
facility maintenance management (BIMFMM) system which allowed information. Information and knowledge of building maintenance ac­
accessing and modifying 3D BIM models to acquire, track and update tivities were collected both during and after tasks by using a specific
maintenance records by using web technologies. A mobile device sub­ developed taxonomy for building maintenance. The collected informa­
system supported access to information on site rather than using PC tion helped FM staff to learn from previous experiences and understand
desktops. Overall, what seems essential for handling facility information how buildings deteriorate. To enhance the previous decision-making
is an ongoing connection between tools that can manage building in­ system, Motawa & Almarshad [52] implemented the Analytical Hier­
formation, such as geometric features, attributes and details, and those archy Process (AHP) in the CBR module for retrieving the most similar
which can deal with maintenance data. knowledge cases. After assigning attributes to knowledge cases to
identify similar situations, a weighting score based on the pairwise
3.2.1.2. Optimal path for sequence of work orders. When FM personnel comparisons decision criteria was assigned to each attribute for ranking
conduct maintenance works, a path that allows them to efficiently cases. Another study was carried out by Ma et al. [53], who presented a
operate in sequence should be identified. However, hardly are WOs data-driven approach based on the Reliability Centred Maintenance
performed following a spatio-temporal logic, therefore leading to time- (RCM) process to support decision-making processes of equipment
consuming and inefficient tasks. To solve this issue, Chen et al. [47] maintenance. The process took into account the need of three features
proposed a facility maintenance management framework based on BIM including large diversity of equipment systems, data repetition of
technology which aimed to perform automatic scheduling of mainte­ equipment replacements and data extracted from BIM and GIS models.
nance work orders. After individuating failure elements, the system To comply with these requirements, the RCM process was modified by
calculated optimised maintenance path based on the BIM model by using quantitative decision-making models and the Monte-Carlo Simu­
using the A* algorithm and automatically scheduled work orders by lation.Although nearly 70–80% of operational accidents, such as falls,
using a modified Dijkstra algorithm based on distance, problem type, electrical shocks and bruises, happen due to human errors [54], decision
emergency level and location. However, this system did not consider support systems for safe maintenance are relatively scarce. To this end,
multiple maintenance teams at the same time, leading to potential work Wetzel & Thabet [55] proposed a framework based on BIM and data
interference. On balance, what seems to be apparent is that systems processing and rule based system (DPRBS) which classified and organ­
which can take into consideration the optimisation of maintenance ised safe maintenance and repair practice for supporting decision-
paths in terms of both time and space are regarded as worthwhile for making processes of maintenance. Once safety information was stored
enhancing maintenance operations. into a singular repository, the DPRBS generated a safety protocol for
maintenance by using a question-based system. Overall, it is worth
3.2.2. Decision-making processes for maintenance noticing that the common practice for boosting decision-making pro­
Records of maintenance interventions and up-to-date data are crit­ cesses stems from the collaborative working between BIM tools and
ical information to enable effective decision-making processes for diverse techniques and methods of processing maintenance data.
maintenance management. To this end, several studies have been con­
ducted on this topic over the last ten years. For instance, Nascimento 3.2.3. Detecting asset faults and inspecting building assets
et al. [48] investigated the integration of the applicability level of visual Maintenance and repair activities begin when assets which do not
management in PDCA (Plan, Do, Check and Act) cycle with BIM and work properly or have failings are detected. However, checking every
Lean methodology to enhance preventive maintenance planning and single asset at a specific level is time-consuming and expensive, there­
control. The framework exploited workers’ opinions written on A3 fore assets need to be distinguished between critical and non-critical
sheets, meetings with professionals with different expertise and BIM ones. Predicting processes can detect possible defects in advance and
models of facilities to support awareness of problems and collaboration consequently enable cost savings. To this end, GhaffarianHoseini et al.
among workers, and reduce maintenance risks and time. Along the lines [56] presented the Integrated Knowledge-based Building Management
of visual comprehension for enhancing decision-making processes, System (nD BIM-IKBMS) framework based on nD BIM applications to
Motamedi et al. [10] proposed the FM Visual Analysis System (FMVAS) provide automatic detection and identification of operational faults. The
framework which exploited BIM-based colour visual analysis to better system exploited a cloud database to store condition monitoring data
identify failure root-cause and conduct maintenance. BIM models were and BIM models to manage and maintain asset information. Data and
linked with data stored in FM software applications through the Revit information were used by the Case-Based Reasoning system and the
Application Programming Interface (API), while the system engine Rule-Based Reasoning system to retrieve similarities when fault
queried and extracted information by using asset/location IDs. Simi­ occurred. The fault detection and isolation (FDI) module used AI algo­
larly, to reduce the cognitive load of FM data spreadsheets, Yalcinkaya & rithms, such as digraphs, fault trees and qualitative physics to recognise,
Singh [49] proposed the VisualCOBie platform based on BIM and Ge­ classify and locate faults from the data stored in the database and alert
stalt’s principles of visual perception. The platform exploited graph- FM personnel. Another relevant research was conducted by Lu et al.
based visual representation of COBie spreadsheet data to enhance data [57], who proposed a system based on Digital Twin to detect anomalies
query. On balance, it can be noticed that technologies that allow visual of centrifugal pump vibration in the HVAC system during O&M phase.
comprehension of maintenance issues have been gaining momentum for The Bayesian on-line change point detection method (BOCPD) was used
enhancing decision-making processes. to detect change points, i.e. where generative parameters of vibration
Other studies have relied on collected historical data about building drift and deviate from normal values. An object-oriented BIM visual­
elements to provide useful insights into abnormalities and causes of is­ isation displayed and pinpointed elements by extending the IFC schema
sues [27]. For instance, Vitiello et al. [50] presented a method based on and connecting each element to data collected by sensor systems
semi-probabilistic approach and BIM technology in order to assess through a primary key in a database. On the whole, what can be
economic performance and losses of buildings caused by seismic events. considered as the main denominator of development is the automatic
BIM models acted as databases where information, such as fragility detection and identification of potential operational faults by exploiting
parameters, was stored and retrieved when necessary. Motawa & up-to-date and dynamic data.
Once faults are detected, there is a need of inspecting building assets

8
M. Marocco and I. Garofolo Automation in Construction 131 (2021) 103917

to perform building components maintenance. However, typically, Eventually, the approach provided topic-specific questions for
tradespeople waste time to pinpoint objects which need maintaining enhancing information collection by creating structured forms based on
during on-site maintenance [58]. To this end, Diao & Shih [59] proposed previous predictions. Another relevant study was carried out by Cheng
a BIM-based Augmented Reality Maintenance System (BARMS) to help et al. [65], who proposed a framework based on BIM, IoT and FMM
FM personnel conduct inspection and work by using a smartphone-based systems integration for data-driven predictive maintenance of mechan­
platform. The system provided safety maintenance routes, MEP infor­ ical, electrical and plumbing (MEP) components. The framework con­
mation and 3D animated operating instructions by integrating BIM and sisted of two layers including an information layer and an application
AR visualisation. Thanks to embedded flashlights of portable devices, layer. The former collected and integrated data from BIM models, sen­
such as tablets or smartphones, maintenance inspection can be con­ sors and FMM systems in real-time by using the Building Automation
ducted both during high and low light conditions. Similarly, Chen et al. and Control Networks (BACnet). The latter conducted predictive main­
[60] proposed a BIM AR FSE system to get easy access to information for tenance by using artificial neural networks (ANN) and support vector
fire safety equipment (FSE) inspections. Information requirements for machine (SVM) machine learning algorithms. This layer allowed FM
FSE inspections were stored into a cloud database, along with COBie staff to notice fault alarming early, predict future failing timing and
spreadsheets and OBJ files, to manage equipment inspection and reduce overtime costs for preparing maintenance tools. On balance, it is
maintenance. The system integrated BIM and AR technology by creating worth noticing that enabling an efficient retrieval and extraction of
a set of AR inspection points in BIM models and applied the iBeacon valuable data from information management systems is the common
technology for data transmission and visualisation of AR virtual ele­ denominator to simulate long-term degradation and track trends of
ments. On balance, it can be noticed that advanced visualisation tech­ building system deterioration.
nology is a frequent governing theme used for facilitating on-site
inspections. 3.3. Energy management

3.2.4. Predictive maintenance Buildings account for more than 40% of the final energy consump­
Maintenance can be reactive/corrective, programmed/preventive tion in the European Union [66], therefore reducing building energy
and predictive [51]. While the first one responds to a cause of failure or consumption is one of the most important environmental topics [67].
breakdown [51], programmed and predictive maintenance aim to act in Dividing energy consumption based on end user, space conditioning
advance in order to prevent possible deteriorations and failings. accounts for almost 50% of the total energy consumption, lighting
Several studies have focused on predictive maintenance over the last nearly 20% and office equipment 20% [68]. Energy management sys­
decade due to its huge potentiality. For example, Wu & Lepech [61] tems are used to monitor and analyse energy utilisation of buildings both
presented a BIM-based framework to generate deterioration and dura­ over a selected period of time and in real-time [2]. Current research on
bility assessment models for predicting performance behaviour and disruptive technologies for energy management focuses on real-time
visualise results in BIM models by exploiting the Revit API. The frame­ energy monitoring and assessing and optimising energy building
work aimed to integrate multi-physics models of chloride-induced performance.
corrosion of reinforced concrete structures to simulate long-term dete­
rioration. Information stored in BIM models were extracted and pro­ 3.3.1. Real-time energy monitoring
cessed by using three models in Matlab including the Finite Element Energy management systems are fundamental tools to check and
Model for modelling transport of heat and matter, the FEM for modelling control costs and help FM personnel in avoiding energy wastage. Along
electrochemical processes at the reinforcement surface and the FEM- these lines, sensors can capture environmental conditions, such as
lattice model for modelling crack formation in concrete domain.Other concentration of CO2, temperature and humidity, which are funda­
advanced analyses can be performed by exploiting information man­ mental for internal comfort [69]. Exploiting real-time monitoring and
agement systems to track trends of building system deterioration [62]. automated control of these parameters, savings in terms of energy can be
For example, Peng et al. [63] proposed an approach based on BIM and achieved [2]. For instance, unoccupied spaces can be pinpointed by
Data Mining (DM) to analyse complex data records and provide insights sensors, which at that point turn off light remotely [2]. Several studies
and patterns for the O&M phase of an airport. After transforming the have focused on BIM and IoT integration for energy management over
BIM relational database into a data warehouse, the approach exploited the last decade. Kang et al. [70] proposed a prototype system based on
three DM techniques to analyse the data warehouse including clustering BIM and IoT technology to improve the management and the monitoring
algorithms, such as the k-means clustering algorithm, outlier detection, of an office. Environmental data were collected through different sen­
such as the local density-based algorithm and the cluster-based frequent sors including DHT11 for temperature and humidity monitoring and
pattern mining algorithm based on the Apriori algorithm. Overall, what PT550 CdS for ambient light. Data was saved in two different databases
seems apparent is that exploiting static data stored in BIM models in in accordance with its properties, namely BIM data was collected in the
conjunction with a significant set of accurate and ongoing maintenance MongoDB database through BSON format, whereas environmental data
data is the common practice for predictive building maintenance was stored in the InfluxDB database. One of the main problems of this
analysis. system related to the issue of separated data silos. To solve this problem,
Challenges arise when dealing with work orders since this data is Zhang et al. [71] presented a cloud-based platform based on BIM tech­
usually free-form and collected in an unstructured way. To this end, nology and an Arduino sensing network that integrated static and real-
McArthur et al. [64] developed a FM-BIM integration approach based on time heterogeneous energy consumption and environmental data to
two sets of supervised machine learning models, which classified and monitor the performance of a commercial and an office building. The
analysed the unstructured written complaints description of work orders system stored static data, such as building basic information, building
to identify, classify and predict WO categories and subcategories. Term geometry and topology, and stream data, such as energy consumption
Frequency (TF), Term Frequency-Inverse Document Frequency (TF- and environmental information, into the HBase cloud-based NoSQL
IDF), Random Forests and Frequent Itemset Analysis (FIA) algorithms database. Other studies focused on wider scenarios by combining
were exploited for information retrieval, identifying the most frequent diverse static information sources to dynamic ones. For instance, Brundu
and the prevalence of representative words and for textual classification, et al. [72] proposed an IoT software infrastructure which integrated BIM
respectively. BIM models allowed FM personnel to visualise work or­ and GIS technology to monitor and manage energy consumption in city
ders, which were inputted as shared parameters, in terms of space and districts and simulate energy analyses from the district level to the
time by displaying colour-coded floor plans, which showed the fre­ building room level. To avoid dependence on closed proprietary sys­
quency and the spatial clustering of WOs for each time period. tems, Dave et al. [73] developed Otaniemi 3D, a web platform based on

9
M. Marocco and I. Garofolo Automation in Construction 131 (2021) 103917

BIM and IoT integration, to monitor energy usage, occupancy and user Data Warehouse Core and the Information Representation layer. The
comfort for a campus by using open messaging standards, such as O-MI3 first one collected data from wired and wireless sensing devices and
and O-DF.4 Other contributions emphasised the visual part of moni­ extracted data in the ifcXML and CSV data types from BIM tools and
toring energy consumption. For example, Chang et al. [74] proposed a sensors, respectively. Moreover, the ETL tool transformed the data types
platform able to transfer sensor data to BIM models to support energy- according to the data warehouse requirements and loaded the data into
savings decision-making processes by using coloured visualisation of dimensional and fact data tables of the Data Warehouse Core. The sec­
the indoor comfort. The system collected energy data from sensors ond one stored data, identified the dynamic data and aggregated multi-
deployed in indoor spaces through the Arduino protocol and exported dimensional data. The last one displayed the building performance to
the data in BIM models by using the Firefly suite plug-in, which linked building owners, facility managers, occupants and building technicians.
Arduino with Dynamo. The indoor comfort displayed in BIM models was To boost the availability and accessibility to building energy informa­
based on the Predictive Mean Vote (PMV) index value, which took into tion, other studies allow FM personnel to get access to energy infor­
consideration a grid visualisation of the planar space of the target room. mation with any device. For example, McGlinn et al. [82] developed the
Along the same lines of giving prominence to energy data visualisation, web-interface BuildVis, which aimed to provide facility managers with
Wu & Liu [75] developed the COZyBIM system based on BIM and IoT suggestions for setting energy-saving strategies. Historical data collected
technology to monitor and manage indoor comfort levels of space usage, at five minutes time intervals and stored into a SQL-database were
thermal comfort and air quality of a conference room. Data was captured examined to define correlations, rules and energy consumption predic­
by using Newtonsoft.json programme and the processor of Arduino Yun tion by using data mining techniques, such as a linear regression and a
and then stored into the Seeed Studio Wio Node’s user database by using decision tree classification. Likewise, Lee et al. [83] presented an energy
Wi-Fi connection. Data visualisation was performed by exploiting the management and operating system based on BIM models to enhance
Autodesk Revit API. Focussing on more specific building parts for energy Building Energy Management Systems (BEMS) efficiency by using web
monitoring, Khajavi et al. [76] proposed a method based on WSN and browser models. The BACnet connected energy consumption with the
BIM technology to create the digital twin of an office building facade. Building Automation System (BAS) by using XML files, whereas Revit.
Transferring data was performed by connecting sensors to an IoT NET API connected BIM models with the system. Three energy-saving
gateway through Bluetooth Low Energy (BLE) and the gateway to a methods were applied including geothermal heat, lighting and air-
cloud platform through Wi-Fi. Only three environmental parameters conditioning, and photovoltaics systems. According to [84], energy ef­
were analysed including light, temperature and relative humidity. On ficiency can be achieved by performing the BIM optimisation based on
balance, it can be noticed that an ongoing and comprehensive integra­ BIM techniques and automatic adjustments, rather than conducting the
tion between static and real-time information which derives from traditional optimisation technique based on manually operations. To
different and diverse sources is the common practice for building energy this end, Petri et al. [84] presented a methodology based on BIM for
monitoring. supporting optimisation of electrical and thermal energy consumption.
Taking into account indoor environmental conditions and acceptable
3.3.2. Assessing and optimising energy building performance indoor comfort level for occupants, the method defined the best setting
As well as monitoring energy consumption, assessing and optimising of inlet air temperature by using the non-domination asynchronous
energy building performance are the other two main tasks to be sequel genetic algorithm NSGA-II. Along these lines, another relevant
considered for reducing energy waste, but also for improving building work was carried out by Bonci et al. [85], who proposed a BIM-based
occupant well-being. On average, people spend 80% of their life in cyber physical system which aimed to automatically conduct multi-
buildings, therefore it is critical to provide occupants with a healthy and objective real-time monitoring of facilities and assess their perfor­
comfortable location to live and work [77]. However, facilities are mance. The system used a virtual simulation laboratory to emulate the
subjected to deterioration, failures and damaging events during the building behaviour and show changes within environments by using the
operational phase. Thus, assessing and optimising building performance Revit DB-link. Moreover, a decision support tool managed the efficiency
are fundamental tasks to guarantee a satisfactory level of service [78]. of the entire system and advised potential corrective actions based on
Several studies have been conducted on this topic over the last ten years. overall throughput effectiveness (OTE) parameters. Similarly, Rogage
Costa et al. [79] proposed the multi-criteria tool MultiOpt based on BIM et al. [86] developed a sensor visualisation platform to link data
and building energy simulation models to conduct energy performance captured from real-time and static sources to spatial data of BIM models
management and building renovation operations. The tool exploited the for visualising and optimising building performance of residential
Key Factors methodology, which took into consideration energy con­ buildings. Fixing a set of thresholds, the system alerted FM personnel
sumption and occupants comfort. Along the lines of using energy whether a performance issue occurred. For instance, it compared hu­
simulation models, Abdelalim et al. [80] provided a framework based on midity, overheating, under-heating and light levels with the designed
BIM and building performance simulation (BPS), which aimed to pro­ ones and generated an alert whether the level exceeded the threshold.
duce insights for energy consumption enhancements by evaluating and Moreover, integrating data retrieved by sensors and qualitative ques­
visualising energy flows and related costs. After adjusting the BIM model tionnaires, a matrix of building performance as to energy and comfort
for energy simulations, the model was converted to a BPS model and use was generated to identify data requirements for supporting energy
calibrated based on sub-hourly measured data during winter and sum­ savings. As well as performance issues, thresholds can take into
mer seasons. Displaying data of energy flows and associated costs was consideration regulatory information to check energy performance
performed by using the Sankey diagram, which helped to discover sys­ regulations. To this end, Zhong et al. [87] proposed an ontology-based
tem inefficiencies and opportunities for energy savings. Other research framework based on BIM, environmental information and regulatory
focused on integrating more energy data sources to enhance building information to conduct environmental compliance checking of an en­
performance. For example, Gökçe & Gökçe [81] proposed the Holistic ergy station. The framework consisted of four layers including infor­
Multi-Dimensional Information Management System based on open in­ mation acquisition, ontology development, semantic processing and
formation exchange and BIM tools to optimise energy consumption by application system. Data obtained from different sources, such as BIM
integrating and analysing energy data sets. The system consisted of three models and sensors, was transformed and integrated into the resource
parts including Extraction Transformation and Loading (ETL) layer, description framework (RDF) data model. The SPARQL rules checked
the indoor environmental indicators, such as indoor air quality and
environmental noise, according to the requirements of constraints. On
3
O-MI = Open Messaging Interface balance, what seems apparent is that using BIM models as a starting
4
O-DF = Open Data Format point is the common practice to create building energy models without

10
M. Marocco and I. Garofolo Automation in Construction 131 (2021) 103917

starting from scratch, but also emulate and simulate the building Dijkstra’s algorithm to calculate the shortest path, Cheng et al. [95]
behaviour and check the compliance with environmental regulations. added a penalty factor equation for calculating safe paths. On balance, it
can be noticed that visual route systems which can exploit 3D visual­
3.4. Emergency management isations along with filtered information which involves building indoor
and outdoor features are the common denominators to detect the posi­
Emergency is mainly caused by human-caused emergencies such as tion of hazards and allow escaping and rescuing occupants in emergency
asset failures, chemical spills and fire, but also derives from natural di­ situations.
sasters, e.g. tornadoes and earthquakes [2]. FM personnel need to
localise hazards as soon as possible, get access easily to building infor­ 3.4.2. Hazards monitoring
mation and provide emergency response. Current research on disruptive Hazards need detecting in advance in order to prevent possible
technologies for emergency management focuses on emergency damages and enable efficient troubleshootings. Several studies have
response and path optimisation, and hazard monitoring. been conducted on this topic over the last decade. For example, Van­
decasteele et al. [96] proposed a methodology based on linking BIM and
3.4.1. Emergency response and path optimisation photographs taken from visual and thermal cameras to improve recog­
Emergency situations differ one from another in terms of gravity, nition and localisation of emergency situations. After classifying scenes
time and context, but also depend on three main interactions including with labels extracted from images and assigning the spatio-temporal
human-human, human-building and human-emergency interactions history of the localisation, objects were detected in images and
[88]. When an emergency occurs, such as a blaze, occupants are not compared with historical images to detect emergencies. Other recent
aware of the position where trouble has just happened and can behave in research focused on the application of smart sensors to anticipate and
three different ways including escaping immediately, ignoring the detect hazards. For instance, Parn et al. [97] proposed the Confined
alarms and helping to solve the problem [89]. To improve the decision- Spaces Safety Monitoring System (COSMoS) to monitor environmental
making processes of responding to emergency situations, Ma & Wu [90] conditions and human behaviours along with collecting archival records
developed a fire emergency management system which combined of data from confined spaces to generate insights for asset management
multiple data sensing networks with video surveillance to check and safety performance. The system exploited a wireless sensor network
whether the emergency was controllable or not and provided the (WSN), audible warning alarms and BIM models to collect real-time
optimal path according to occupants’ behaviour. The system consisted data, alert FM personnel to issues and visualise potential hazards,
of four modules including fire intelligence monitoring, fire-warning, respectively. Moreover, the whole system allowed using machine algo­
fire-response and fire-treatment modules to detect fire, alert occupants rithms to improve emergency management within facilities and learning
and provide a countermeasure to emergency. The system exploited BIM from experience to design future practice. As well as emergencies,
models for summarising the functional requirements of occupants ac­ hazards can happen during FM activities, such as maintenance tasks. To
cording to the different behaviour choices and visualising the provided this end, Arslan et al. [98] proposed the Worker Trajectory Analysis
path. Limitations of this system relate to the active guidance, which System (WoTAS) which identified the most likely high-risk working
relied only on the visual route of 3D BIM models, and the incapacity of areas by detecting places of unsafe workers movements and rotation in
conducting simulation, namely the system worked only after the fire real-time during FM activities. To define the risk areas, Regions of In­
occurred. Overall, it is worth noticing that relying on and using accurate terest (RoIs) and Point of Interest (PoI), where users stay for longer
and up-to-date information captured from sensing devices is a common duration, were pinpointed in BIM models by collecting data through
practice for making relevant decisions in hazardous scenarios. Bluetooth Low Energy (BLE). The system exploited the Hidden Markov
In the case of escaping, occupants need to have access to the fastest Model (HMM) and the Viterbi algorithm for extracting multifaceted
and safest escape path in the short-term period [91]. To this end, trajectory characteristics, which describe object behaviours. On balance,
Tashakkori et al. [92] proposed the Indoor Emergency Spatial Model it is worth noticing that combining BIM models with remote, ongoing
(IESM) to visualise 3D BIM indoor environments, pinpoint detailed se­ and real-time monitoring conducted by wireless sensing devices is the
mantic and geographic information and optimise route finding by using common practice for recognising and localising emergency situations.
the Esri GIS platform. The system exploited the IFC schema by filtering
only necessary data including indoor building information, dynamic and 4. Research gaps and future directions
semantic building information and outdoor emergency information.
Space containing hazards were highlighted and the position of windows The research of the integration of disruptive technologies with FM
for escaping in emergency situations were identified in the BIM models. has shown increasing interest and development over the last decade.
Although the shortest path might be the solution, it may trap occupants This section focuses on discussing the research gaps identified during the
into the hazardous location [93]. For this reason, Zhang et al. [94] content analysis and proposing future research directions. Furthermore,
proposed a fire evacuation management framework based on the total this section presents two recommendations for future research, which
risk level for generating customised evacuation routes by using BIM are real-time building security and gamification approach for O&M
models. The framework used indoor real-time location systems (RTLS) management.
such as Bluetooth low energy (BLE) to retrieve information about
emergencies and exploit them to calculate the weighted summation of 4.1. Digital technology implementation in facilities management processes
the risk level index value of all risk parameters for all segments along the
route. The process of adopting digital technology for FM purposes is strictly
On the other hand, first responders and rescuers do not have connected to elements including people, processes and technology.
comprehensive information of facilities and 3D data visualisation to get Although FM has been integrated with disruptive technologies for more
awareness easily when navigating buildings. To solve these issues, than a decade, there still remains a change-resistant attitude to new
Cheng et al. [95] developed the BIM-based Intelligent Fire Prevention ways of planning and conducting FM tasks. The identified gaps in the
and Disaster Relief System, which aimed to provide fire-fighters and implementation of disruptive technologies in FM activities stem from
occupants with optimised rescue and evacuation routes, respectively. three main issues. Firstly, a lack of coherent information management
The web system collected information by using Bluetooth technology hampers effective application. Indeed, there is often a lack of BIM
sensors and displayed 3D real-time and dynamic route-information of models as a starting point to manage building information or informa­
facilities through the Revit Application Programming Interface by tion contained into BIM models is scattered and not accurate due to
providing voice and arrow localisation functions. As well as using the reluctance of project stakeholders to share information, along with

11
M. Marocco and I. Garofolo Automation in Construction 131 (2021) 103917

different efficiency in managing building data by professionals during to a continuous creation of new relevant FM information and conse­
the design and construction phase. Although owners know data required quently potential insights to act. Advanced techniques of data analysis,
to successfully perform the operation phase, whole information needed such as machine learning, have been developed to analyse data and
for FM is still not contracted [3]. To ensure the accuracy and integrity of provide useful insights in an effective manner. However, data is sub­
the data source and the usability of this data in the operational phase, jected to errors based on its quality, therefore developing efficient
major efforts should be taken to ensure that classification systems, which decision-making processes continues to be a challenge worthy of
are data infrastructures applied to building information in order to investigation. Thus, more studies should focus on machine learning al­
effectively catalogue and sort the information generated during the gorithms integration to create accurate prediction models for perform­
design and construction phases, are developed and documented. Thus, ing more flexible analysis without losing data and capable of generating
there is still a great need to involve FM staff during the design and data for lost signals. On the other hand, visual outcomes in the form of
construction phases to give advice to set up an environment that would tables and graphs can facilitate the understanding and interpretation of
be easier to manage. Future research agenda should involve more data, especially for non-experts of data analysis. Thus, future agenda
research into creating mechanisms and procedures for quality manage­ should also focus on developing decision-making systems based on
ment of BIM models in every phase according to FM requirements. graphical representation of data. Furthermore, considering experience
Secondly, there is still a lack of the technical skills to manage cutting- and knowledge of workers is critical to guide decisions, but often
edge technology in the operational phase, especially for maintenance neglected, as usually a limited number of focus groups are involved
management. Indeed, modelling and maintaining the models of facil­ during research. Defining performance indicators for strategic decision
ities, along with collecting and analysing accurate maintenance data methods based on both data and worker expertise could be a topic of
need knowledge, competencies and processes which are not standard in interest.
the FM context. For this reason, many organisations prefer to outsource
FM activities due to its low strategic consideration, savings on associated 4.3. User-centred facilities management
costs and reduction of personnel head-count [3]. Another barrier iden­
tified during the content analysis which concerns all the identified fields The main goal of FM is to provide building owners and occupants
of application is the lack of real cases for validating approaches and with facilities, which can support their business processes and a satis­
systems. This problem has as effect a slow improvement of innovation factory level of healthy and comfortable location to work and live,
and technology and the absence of benefits awareness. Thus, more respectively. Considering only hypothetical situations and data
studies based on defining upsides, benefits and return of investment of retrieved by sensors and devices when guiding decision-making pro­
using disruptive technologies for FM are needed. Future research should cesses simplifies the boundary conditions rather than representing real
focus on extending fields of application by considering different settings situations. For example, assumptions for evacuation route in the emer­
and use scenarios and giving emphasis and a clear understanding of cost gency management usually take into consideration only good physical
savings. characteristics of occupants. Neglected features, such as gender, age and
mobility-impairment considerations, are paramount to be considered
4.2. Digital twin platforms [94]. With reference to the energy management, an example could be
the air temperature and humidity that users want when they work or live
The importance of graphical data falls down significantly at the end in a building. Thus, not taking into account real users’ needs and feed­
of the design phase, while information regarding attributes and details back can lead to misleading perception of effective facilities manage­
increases sharply in the construction and operational phase [99]. ment. Future agenda should focus more on conducting user-centred
Technology deployed in buildings provides FM personnel with huge post-occupancy evaluations by creating performance parameters and
amounts of data to manage maintenance, energy and emergency issues, indicators according to human requirements, behaviours and influence
therefore developing platforms to perform data classification, integra­ over buildings.
tion and analysis for facilities management are needed. Potential
research directions for all identified fields of application can focus on 4.4. Integrating facilities management with interactive visualisation
current gaps and technical limitations including i) lack of data integra­ technologies
tion systems, ii) cybersecurity and iii) complexity of decision-making
processes. Traditionally, on-site tasks rely on paper-based documents, phone
Deploying sets of sensors and devices to capture up-to-date data from assistance and personal knowledge and experience of facilities managers
buildings and better understanding what kind of sensor best fits the [39] [15] [2]. Nevertheless, FM tasks can be leveraged by using cutting-
specific data collection are critical steps. Nevertheless, different data edge technology, which allow FM personnel to perform their work more
collection devices store information into different formats and data­ efficiently than traditional methods, especially in the maintenance and
bases, leading to the issue of lack of interoperability and separated data emergency management scenarios. Virtual Reality, Mixed Reality and
silos. Thus, semantics-based web relational databases, which are capable Augmented Reality have gained attention over the last decade, but
of performing automatic transfer and processing of data, are needed. several issues are not still addressed [102]. Firstly, handheld devices are
Future research should focus more on developing Digital Twin platforms not comfortable to support dynamic tasks as they need at least one hand
by integrating cloud computing, BIM and IoT technologies to manage to be held, therefore hampering FM tasks. On the other hand, although
and have access to real-time facilities data. head-mounted display is regarded as a possible solution, FM personnel
Nevertheless, exploiting web technologies exposes systems to cyber- are still sceptical of its usage in real cases [26]. Future research should
attacks, which can cause several issues, such as unintended spillover of focus on implementing and improving vocal commands on both hand­
data and disruption of network services [100]. Dealing with cyberse­ held devices and not for boosting the process of querying and obtaining
curity issues is a critical task, therefore a secure and private environment information in real case studies.
is needed to gather reliable and accurate information [101]. Future A technical challenge occurs when working outside under strong
research studies should focus more on blockchain, IoT and BIM tech­ sunlight. This condition hinders the correct operation of the system,
nology integration, which can provide authorised stakeholders with leading to instability in visualisation [44]. Another technical issue refers
data they need and avoid inappropriate access to information to those to the mismatch level of detail between BIM and AR technologies. Thus,
who are not authorised. further studies in these directions are required. Other potential research
Another challenge refers to decision-making processes. This issue is can focus on current limitations including i) accurate matching between
due to the fact that FM data is not only static but also dynamic, leading virtual camera and physical camera in mobile devices and ii) focal

12
M. Marocco and I. Garofolo Automation in Construction 131 (2021) 103917

length and depth of field between virtual camera and physical camera return of investment of using innovative technology for FM, ii) involving
[41]. FM professionals during the design and construction phases for creating
A collaborative system, which allows multiple maintenance teams to mechanisms and procedures for quality management of BIM models, iii)
perform their tasks simultaneously, remains a challenge worthy of integrating blockchain, IoT and BIM technologies, iii) defining perfor­
investigation. Thus, a recommendation for future research concerns mance indicators for strategic decision methods based on both data and
developing a platform which allows multiple FM teams to work at the worker expertise and iv) user-centred post-occupancy evaluations. Po­
same time in the same building zone by sharing information, such as tential FM applications, such as real-time building security and gamifi­
position and time span of tasks and whether including potential hazards. cation approach for O&M management, are suggested in this paper.
Finally, it is hoped that this paper will contribute to improving and
4.5. Real-time building security providing helpful insights for FM. Since FM aims to enhance the quality
of life of occupants and the productivity of business, advancements will
There has been limited research on some of O&M topics, such as real- necessitate the joint efforts of professionals and academics to overcome
time building security, over the last decade. Monitoring movements of the current challenges.
people has always caused conflicting issues regarding privacy [69].
However, keeping the record of occupant movements can provide Declaration of Competing Interest
relevant information for future checks and analyses. For example, theft
monitoring systems, emergency escape route planning, hotdesking and The authors declare that they have no known competing financial
space management can be potential enhanced applications. In partic­ interests or personal relationships that could have appeared to influence
ular, during emergency situations, such as COVID-19 pandemic, multi­ the work reported in this paper.
ple data sensing networks can be integrated with video surveillance to
store relevant information about occupants. Combining cutting-edge Acknowledgements
monitoring technology with a smart working approach for providing
safer and more comfortable working spaces according to users’ behav­ This research did not receive any specific grant from funding
iours could be a topic of interest. Future agenda should focus more on agencies in the public, commercial, or not-for-profit sectors.
developing policies and systems that can track user movement prefer­
ences without having trouble with privacy issues. References

4.6. Gamification approach for O&M management [1] What is FM - Definition of Facility Management. https://www.ifma.org/abou
t/what-is-facility-management. (Accessed 25 February 2020).
[2] Burcin Becerik-Gerber, Farrokh Jazizadeh, Nan Li, Gulben Calis, Application
Alhough FM is not among the core competences of organisations, its areas and data requirements for BIM-enabled facilities management, J. Constr.
relevance for an optimal building operation is widely recognised and Eng. Manag. 138 (3) (Mar. 2012) 431–442, https://doi.org/10.1061/(ASCE)
unquestionable. However, occupants often underrate or even neglect the CO.1943-7862.0000433.
[3] E. Teicholz, Bridging the AEC/FM technology gap, J. Facil. Manag. (2004) 1–8
importance of buildings management and energy savings. Due to these [Online]. Available: https://docplayer.net/64590079-Bridging-the-aec-fm-tech
reasons, issues in terms of wastage and failures can occur. A solution to nology-gap-eric-teicholz-ifma-fellow.html.
increase engagement for buildings management refers to applying real [4] P. Scarponcini, Editorial: time for an integrated approach to facility management,
J. Comput. Civ. Eng. 10 (1) (Jan. 1996), https://doi.org/10.1061/(ASCE)0887-
situations into games [103]. Games mainly fix objectives, give in­ 3801(1996)10:1(3), pp. 3–3.
structions and encourage the creation of teams, which are three critical [5] A. Motamedi, M.M. Soltani, S. Setayeshgar, A. Hammad, Extending IFC to
topics for organisations. Gamification can become a critical ally of FM by incorporate information of RFID tags attached to building elements, Adv. Eng.
Inform. 30 (1) (Jan. 2016) 39–53, https://doi.org/10.1016/j.aei.2015.11.004.
assigning points to occupants based on their reports on warnings of [6] Y.-C. Lin, Y.-C. Su, Y.-P. Chen, Developing mobile BIM/2D barcode-based
wasteful building components and rewarding them according to the automated facility management system, Sci. World J. 2014 (2014) 1–16, 374735,
obtained score. With the help of gamification, proactive behaviour and https://doi.org/10.1155/2014/374735.
[7] R. Volk, J. Stengel, F. Schultmann, Building information modeling (BIM) for
more cognitive interest to reduce waste are incentivised. Future studies
existing buildings — literature review and future needs, Autom. Constr. 38 (Mar.
related to FM should include gamification-based systems to increase 2014) 109–127, https://doi.org/10.1016/j.autcon.2013.10.023.
business savings. [8] S. Azhar, Building information modeling (BIM): trends, benefits, risks, and
challenges for the AEC industry, Leadersh. Manag. Eng. 11 (3) (Jul. 2011)
241–252, https://doi.org/10.1061/(ASCE)LM.1943-5630.0000127.
5. Conclusion [9] A. Bosch, L. Volker, A. Koutamanis, BIM in the operations stage: bottlenecks and
implications for owners, Built Environ. Project Asset Manag. 5 (3) (Jul. 2015)
This paper contributes to the body of knowledge by presenting an in- 331–343, https://doi.org/10.1108/BEPAM-03-2014-0017.
[10] A. Motamedi, A. Hammad, Y. Asen, Knowledge-assisted BIM-based visual
depth review of the current applications of the integration of disruptive analytics for failure root cause detection in facilities management, Autom. Constr.
technologies with FM, examining current limitations and identifying 43 (Jul. 2014) 73–83, https://doi.org/10.1016/j.autcon.2014.03.012.
potential future research directions. A total of 70 articles, published [11] Q. Lu, X. Xie, J. Heaton, A.K. Parlikad, J. Schooling, From BIM towards digital
twin: Strategy and future development for smart asset management, in: Service
between 2011 and 2020, were identified and reviewed. Four topics Oriented, Holonic and Multi-Agent Manufacturing Systems for Industry of the
including Information Management, Maintenance Management, Energy Future vol. 853, Springer International Publishing, Cham, 2020, pp. 392–404,
Management and Emergency Management were analysed. The content https://doi.org/10.1007/978-3-030-27477-1_30.
[12] S. Howell, Y. Rezgui, Beyond BIM: Knowledge Management for a Smarter Future. St
analysis of the existing literature shows that BIM was the main tech­ Albans: IHS Markit, ISBN: 978–1–84806-476-8, 2018.
nology covered in the reviewed articles. A promising starting point for [13] S. Tang, D.R. Shelden, C.M. Eastman, P. Pishdad-Bozorgi, X. Gao, A review of
boosting FM includes developing Digital Twin platforms by integrating building information modeling (BIM) and the internet of things (IoT) devices
integration: present status and future trends, Autom. Constr. 101 (May 2019)
cloud computing, BIM and IoT technologies. Although the research 127–139, https://doi.org/10.1016/j.autcon.2019.01.020.
trends demonstrate that there is a seamless increasing interest in inte­ [14] C. Boje, A. Guerriero, S. Kubicki, Y. Rezgui, Towards a semantic construction
grating disruptive technologies with FM, several challenges still hinder digital twin: directions for future research, Autom. Constr. 114 (Jun. 2020)
103179, https://doi.org/10.1016/j.autcon.2020.103179.
this process – change-resistant attitude to new ways of planning and
[15] M. Gheisari, J. Irizarry, Investigating human and technological requirements for
conducting FM tasks, technical skills to manage cutting-edge technol­ successful implementation of a BIM-based mobile augmented reality environment
ogy, a lack of real cases for validating approaches and systems, data in facility management practices, Facilities 34 (1/2) (Feb. 2016) 69–84, https://
integration systems, cybersecurity and complexity of decision-making doi.org/10.1108/F-04-2014-0040.
[16] R. Santos, A.A. Costa, A. Grilo, Bibliometric analysis and review of building
processes. Based on the observed limitations, future research di­ information Modelling literature published between 2005 and 2015, Autom.
rections include i) more studies based on defining upsides, benefits and Constr. 80 (Aug. 2017) 118–136, https://doi.org/10.1016/j.autcon.2017.03.005.

13
M. Marocco and I. Garofolo Automation in Construction 131 (2021) 103917

[17] How are CiteScore metrics used in Scopus?, in: Scopus: Access and use Support [44] L. Hou, et al., Combining photogrammetry and augmented reality towards an
Center, 2020. https://service.elsevier.com (accessed Oct. 23, 2020). integrated facility management system for the oil industry, Proc. IEEE 102 (2)
[18] A. Sayers, Tips and tricks in performing a systematic review, Br. J. Gen. Pract. 58 (Feb. 2014) 204–220, https://doi.org/10.1109/JPROC.2013.2295327.
(547) (Feb. 2008) 136, https://doi.org/10.3399/bjgp08X277168. [45] K. El Ammari, A. Hammad, Remote interactive collaboration in facilities
[19] HM Government, Digital Built Britain, Accessed: Mar. 31, 2020. [Online]. management using BIM-based mixed reality, Autom. Constr. 107 (Nov. 2019)
Available: https://assets.publishing.service.gov.uk/government/uploads/syst 102940, https://doi.org/10.1016/j.autcon.2019.102940.
em/uploads/attachment_data/file/410096/bis-15-155-digital-built-britain-level- [46] Y.-C. Lin, Y.-C. Su, Developing mobile- and BIM-based integrated visual facility
3-strategy.pdf, 2015. maintenance management system, Sci. World J. 2013 (2013) 1–10, https://doi.
[20] Q. Lu, L. Chen, S. Lee, X. Zhao, Activity theory-based analysis of BIM org/10.1155/2013/124249.
implementation in building O&M and first response, Autom. Constr. 85 (Jan. [47] W. Chen, K. Chen, J.C.P. Cheng, Q. Wang, V.J.L. Gan, BIM-based framework for
2018) 317–332, https://doi.org/10.1016/j.autcon.2017.10.017. automatic scheduling of facility maintenance work orders, Autom. Constr. 91
[21] K. Kim, H. Kim, W. Kim, C. Kim, J. Kim, J. Yu, Integration of ifc objects and (Jul. 2018) 15–30, https://doi.org/10.1016/j.autcon.2018.03.007.
facility management work information using semantic web, Autom. Constr. 87 [48] D.L.M. de Nascimento, O.L.G. Quelhas, M.J. Meiriño, R.G.G. Caiado, S.D.
(Mar. 2018) 173–187, https://doi.org/10.1016/j.autcon.2017.12.019. J. Barbosa, P. Ivson, Facility management using digital obeya room by integrating
[22] E. Valero, A. Adán, F. Bosché, Semantic 3D reconstruction of furnished interiors bim-lean approaches – an empirical study, J. Civ. Eng. Manag. 24 (8) (Dec. 2018)
using laser scanning and RFID technology, J. Comput. Civ. Eng. 30 (4) (Jul. 581–591, https://doi.org/10.3846/jcem.2018.5609.
2016), https://doi.org/10.1061/(ASCE)CP.1943-5487.0000525, 04015053. [49] M. Yalcinkaya, V. Singh, VisualCOBie for facilities management: a BIM
[23] A.M. Costin, J. Teizer, Fusing passive RFID and BIM for increased accuracy in integrated, visual search and information management platform for COBie
indoor localization, Vis. Eng. 3 (1) (Dec. 2015), https://doi.org/10.1186/s40327- extension, Facilities 37 (7/8) (May 2019) 502–524, https://doi.org/10.1108/F-
015-0030-6. 01-2018-0011.
[24] A. Motamedi, M.M. Soltani, A. Hammad, Localization of RFID-equipped assets [50] U. Vitiello, V. Ciotta, A. Salzano, D. Asprone, G. Manfredi, E. Cosenza, BIM-based
during the operation phase of facilities, Adv. Eng. Inform. 27 (4) (Oct. 2013) approach for the cost-optimization of seismic retrofit strategies on existing
566–579, https://doi.org/10.1016/j.aei.2013.07.001. buildings, Autom. Constr. 98 (Feb. 2019) 90–101, https://doi.org/10.1016/j.
[25] K.M. Rashid, J. Louis, K.K. Fiawoyife, Wireless electric appliance control for autcon.2018.10.023.
smart buildings using indoor location tracking and BIM-based virtual [51] I. Motawa, A. Almarshad, A knowledge-based BIM system for building
environments, Autom. Constr. 101 (May 2019) 48–58, https://doi.org/10.1016/j. maintenance, Autom. Constr. 29 (Jan. 2013) 173–182, https://doi.org/10.1016/
autcon.2019.01.005. j.autcon.2012.09.008.
[26] F. Baek, I. Ha, H. Kim, Augmented reality system for facility management using [52] I. Motawa, A. Almarshad, Case-based reasoning and BIM systems for asset
image-based indoor localization, Autom. Constr. 99 (Mar. 2019) 18–26, https:// management, Built Environ. Project Asset Manag. 5 (3) (Jul. 2015) 233–247,
doi.org/10.1016/j.autcon.2018.11.034. https://doi.org/10.1108/BEPAM-02-2014-0006.
[27] A. Akcamete, X. Liu, B. Akinci, J.H. Garrett, Integrating and visualizing [53] Z. Ma, Y. Ren, X. Xiang, Z. Turk, Data-driven decision-making for equipment
maintenance and repair work orders in BIM: lessons learned from a prototype, in: maintenance, Autom. Constr. 112 (Apr. 2020) 103103, https://doi.org/10.1016/
Presented at the 11th International Conference on Construction Applications of j.autcon.2020.103103.
Virtual Reality, 2011, https://doi.org/10.25643/bauhaus-universitaet.1468. [54] N. Leveson, A new accident model for engineering safer systems, Saf. Sci. 42 (4)
[28] J. Heaton, A.K. Parlikad, J. Schooling, Design and development of BIM models to (Apr. 2004) 237–270, https://doi.org/10.1016/S0925-7535(03)00047-X.
support operations and maintenance, Comput. Ind. 111 (Oct. 2019) 172–186, [55] E.M. Wetzel, W.Y. Thabet, The use of a BIM-based framework to support safe
https://doi.org/10.1016/j.compind.2019.08.001. facility management processes, Autom. Constr. 60 (Dec. 2015) 12–24, https://
[29] T.-W. Kang, H.-S. Choi, BIM perspective definition metadata for interworking doi.org/10.1016/j.autcon.2015.09.004.
facility management data, Adv. Eng. Inform. 29 (4) (Oct. 2015) 958–970, https:// [56] A. GhaffarianHoseini, et al., ND BIM-integrated knowledge-based building
doi.org/10.1016/j.aei.2015.09.004. management: inspecting post-construction energy efficiency, Autom. Constr. 97
[30] C.-H. Ko, N.-F. Pan, C.-C. Chiou, Web-based radio frequency identification facility (Jan. 2019) 13–28, https://doi.org/10.1016/j.autcon.2018.10.003.
management systems, Struct. Infrastruct. Eng. 9 (5) (May 2013) 465–480, [57] Q. Lu, X. Xie, A.K. Parlikad, J.M. Schooling, Digital twin-enabled anomaly
https://doi.org/10.1080/15732479.2010.546804. detection for built asset monitoring in operation and maintenance, Autom.
[31] J. Zhan, X.J. Ge, S. Huang, L. Zhao, J.K.W. Wong, S.X. He, Improvement of the Constr. 118 (Oct. 2020) 103277, https://doi.org/10.1016/j.autcon.2020.103277.
inspection-repair process with building information modelling and image [58] S. Lee, Ö. Akin, Augmented reality-based computational fieldwork support for
classification, Facilities 37 (7/8) (May 2019) 395–414, https://doi.org/10.1108/ equipment operations and maintenance, Autom. Constr. 20 (4) (Jul. 2011)
F-01-2018-0005. 338–352, https://doi.org/10.1016/j.autcon.2010.11.004.
[32] A. Motamedi, R. Saini, A. Hammad, B. Zhu, Role-based access to facilities [59] P.-H. Diao, N.-J. Shih, BIM-based AR maintenance system (BARMS) as an
lifecycle information on RFID tags, Adv. Eng. Inform. 25 (3) (Aug. 2011) intelligent instruction platform for complex plumbing facilities, Appl. Sci. 9 (8)
559–568, https://doi.org/10.1016/j.aei.2011.03.004. (Apr. 2019) 1592, https://doi.org/10.3390/app9081592.
[33] E.A. Pärn, D.J. Edwards, Conceptualising the FinDD API plug-in: a study of BIM- [60] Y.-J. Chen, Y.-S. Lai, Y.-H. Lin, BIM-based augmented reality inspection and
FM integration, Autom. Constr. 80 (Aug. 2017) 11–21, https://doi.org/10.1016/ maintenance of fire safety equipment, Autom. Constr. 110 (Feb. 2020) 103041,
j.autcon.2017.03.015. https://doi.org/10.1016/j.autcon.2019.103041.
[34] W.-L. Lee, M.-H. Tsai, C.-H. Yang, J.-R. Juang, J.-Y. Su, V3DM+: BIM interactive [61] J. Wu, M.D. Lepech, Incorporating multi-physics deterioration analysis in
collaboration system for facility management, Vis. Eng. 4 (1, 5) (Dec. 2016), building information modeling for life-cycle management of durability
https://doi.org/10.1186/s40327-016-0035-9. performance, Autom. Constr. 110 (Feb. 2020) 103004, https://doi.org/10.1016/
[35] E. Halmetoja, The conditions data model supporting building information models j.autcon.2019.103004.
in facility management, Facilities 37 (7/8) (May 2019) 484–501, https://doi.org/ [62] P. Parsanezhad, An overview of information logistics for FM&O business
10.1108/F-11-2017-0112. processes, in: eWork and eBusiness in Architecture, Engineering and Construction,
[36] V. Balali, M. Noghabaei, A. Heydarian, K. Han, Improved stakeholder CRC Press, 2014, pp. 719–725, https://doi.org/10.1201/b17396-117.
communication and visualizations: real-time interaction and cost estimation [63] Y. Peng, J.-R. Lin, J.-P. Zhang, Z.-Z. Hu, A hybrid data mining approach on BIM-
within immersive virtual environments, in: Construction Research Congress 2018, based building operation and maintenance, Build. Environ. 126 (Dec. 2017)
New Orleans, Louisiana, Mar. 2018, pp. 522–530, https://doi.org/10.1061/ 483–495, https://doi.org/10.1016/j.buildenv.2017.09.030.
9780784481264.051. [64] J.J. McArthur, N. Shahbazi, R. Fok, C. Raghubar, B. Bortoluzzi, A. An, Machine
[37] E. González, J.D. Piñeiro, J. Toledo, R. Arnay, L. Acosta, An approach based on learning and BIM visualization for maintenance issue classification and enhanced
the ifcOWL ontology to support indoor navigation, Egypt. Inform. J. 22 (1) (Mar. data collection, Adv. Eng. Inform. 38 (Oct. 2018) 101–112, https://doi.org/
2020) 1–13, https://doi.org/10.1016/j.eij.2020.02.008. 10.1016/j.aei.2018.06.007.
[38] R. Neuville, J. Pouliot, R. Billen, Identification of the best 3D viewpoint within [65] J.C.P. Cheng, W. Chen, K. Chen, Q. Wang, Data-driven predictive maintenance
the BIM model: application to visual tasks related to facility management, planning framework for MEP components based on BIM and IoT using machine
Buildings 9 (7) (Jul. 2019) 1–18, https://doi.org/10.3390/buildings9070167. learning algorithms, Autom. Constr. 112 (Apr. 2020) 103087, https://doi.org/
[39] J. Wang, Y. Feng, C. Zeng, S. Li, An augmented reality based system for remote 10.1016/j.autcon.2020.103087.
collaborative maintenance instruction of complex products, in: IEEE International [66] P. Cox, M. Fisher Boel, Directive 2002/91/ec of the European Parliament and of
Conference on Automation Science and Engineering (CASE), 2014, pp. 309–314. the council of 16 December 2002 on the energy performance of buildings, Off. J.
Aug. ISSN: 2161-8089, https://doi.org/10.1109/CoASE.2014.6899343. Eur. Communities (2003) 65–71. Accessed: Mar. 03, 2020. [Online]. Available:
[40] J. Irizarry, M. Gheisari, G. Williams, K. Roper, Ambient intelligence environments https://eur-lex.europa.eu/legal-content/en/ALL/?uri=CELEX:32002L0091.
for accessing building information: a healthcare facility management scenario, [67] Dodge data and analytics, in: World Green Building Trends 2018, 2018. Accessed:
Facilities 32 (3/4) (Feb. 2014) 120–138, https://doi.org/10.1108/F-05-2012- Mar. 03, 2020. [Online]. Available, https://www.worldgbc.org/sites/default/fi
0034. les/World%20Green%20Building%20Trends%202018%20SMR%20FINAL%20
[41] G. Williams, M. Gheisari, P.-J. Chen, J. Irizarry, BIM2MAR: An efficient BIM 10-11.pdf.
translation to mobile augmented reality applications, J. Manag. Eng. 31 (1) (Jan. [68] A. Stumpf, H. Kim, E. Jenicek, Early design energy analysis using BIMs (Building
2015), https://doi.org/10.1061/(ASCE)ME.1943-5479.0000315. A4014009. Information Models), in: Construction Research Congress 2009, Seattle,
[42] J. Du, Z. Zou, Y. Shi, D. Zhao, Zero latency: real-time synchronization of BIM data Washington, United States, 2009, pp. 426–436, https://doi.org/10.1061/41020
in virtual reality for collaborative decision-making, Autom. Constr. 85 (Jan. (339)44. Apr.
2018) 51–64, https://doi.org/10.1016/j.autcon.2017.10.009. [69] J. Xu, K. Chen, A.E. Zetkulic, F. Xue, W. Lu, Y. Niu, Pervasive sensing technologies
[43] E. Patti, et al., Information modeling for virtual and augmented reality, IT Prof. for facility management: a critical review, Facilities 38 (1/2) (Aug. 2019)
19 (3) (2017) 52–60, https://doi.org/10.1109/MITP.2017.43. 161–180, https://doi.org/10.1108/F-02-2019-0024.

14
M. Marocco and I. Garofolo Automation in Construction 131 (2021) 103917

[70] K. Kang, J. Lin, J. Zhang, BIM- and IoT-based monitoring framework for building Build. Environ. 141 (Aug. 2018) 127–142, https://doi.org/10.1016/j.
performance management, J. Struct. Integr. Maint. 3 (4) (Oct. 2018) 254–261, buildenv.2018.05.046.
https://doi.org/10.1080/24705314.2018.1536318. [88] R. Zhu, J. Lin, B. Becerik-Gerber, N. Li, Human-building-emergency interactions
[71] Y.-Y. Zhang, K. Kang, J.-R. Lin, J.-P. Zhang, Y. Zhang, Building information and their impact on emergency response performance: a review of the state of the
modeling–based cyber-physical platform for building performance monitoring, art, Saf. Sci. 127 (Jul. 2020) 104691, https://doi.org/10.1016/j.
Int. J. Distrib. Sensor Netw. 16 (2) (Feb. 2020), https://doi.org/10.1177/ ssci.2020.104691.
1550147720908170. [89] C.M. Zhao, S.M. Lo, S.P. Zhang, M. Liu, A post-fire survey on the pre-evacuation
[72] F.G. Brundu, et al., IoT software infrastructure for energy management and human behavior, Fire. Technol 45 (1) (Mar. 2009) 71–95, https://doi.org/
simulation in smart cities, IEEE Trans. Indus. Inform. 13 (2) (Apr. 2017) 832–840, 10.1007/s10694-007-0040-6.
https://doi.org/10.1109/TII.2016.2627479. [90] G. Ma, Z. Wu, BIM-based building fire emergency management: combining
[73] B. Dave, A. Buda, A. Nurminen, K. Främling, A framework for integrating BIM and building users’ behavior decisions, Autom. Constr. 109 (Jan. 2020) 102975,
IoT through open standards, Autom. Constr. 95 (Nov. 2018) 35–45, https://doi. https://doi.org/10.1016/j.autcon.2019.102975.
org/10.1016/j.autcon.2018.07.022. [91] M.-P. Kwan, J. Lee, Emergency response after 9/11: the potential of real-time 3D
[74] K.-M. Chang, R.-J. Dzeng, Y.-J. Wu, An automated IoT visualization BIM platform GIS for quick emergency response in micro-spatial environments, Comput.
for decision support in facilities management, Appl. Sci. 8 (7) (Jul. 2018) 1086, Environ. Urban. Syst. 29 (2) (Mar. 2005) 93–113, https://doi.org/10.1016/j.
https://doi.org/10.3390/app8071086. compenvurbsys.2003.08.002.
[75] I.-C. Wu, C.-C. Liu, A visual and persuasive energy conservation system based on [92] H. Tashakkori, A. Rajabifard, M. Kalantari, A new 3D indoor/outdoor spatial
BIM and IoT technology, Sensors 20 (1) (Dec. 2019) 139, https://doi.org/ model for indoor emergency response facilitation, Build. Environ. 89 (Jul. 2015)
10.3390/s20010139. 170–182, https://doi.org/10.1016/j.buildenv.2015.02.036.
[76] S.H. Khajavi, N.H. Motlagh, A. Jaribion, L.C. Werner, J. Holmstrom, Digital twin: [93] K.A.F.A. Samah, B. Hussin, A.S.H. Basari, Modification of Dijkstra’s algorithm for
vision, benefits, boundaries, and creation for buildings, IEEE Access 7 (2019) safest and shortest path during emergency evacuation, Appl. Math. Sci. 9 (29–32)
147406–147419, https://doi.org/10.1109/ACCESS.2019.2946515. (2015) 1531–1541, https://doi.org/10.12988/ams.2015.49681.
[77] R. Yang, L. Wang, Multi-agent based energy and comfort management in a [94] J. Zhang, J. Guo, X. Liu, H. Xiong, D. Zhang, A framework for an intelligent and
building environment considering behaviors of occupants, in: 2012 IEEE Power personalized fire evacuation management system, Sensors 19 (14) (Jul. 2019)
and Energy Society General Meeting, 2012, pp. 1–7, https://doi.org/10.1109/ 3128, https://doi.org/10.3390/s19143128.
PESGM.2012.6345671. Jul. [95] M.-Y. Cheng, K.-C. Chiu, Y.-M. Hsieh, I.-T. Yang, J.-S. Chou, Y.-W. Wu, BIM
[78] T. Torfs, et al., Low power wireless sensor network for building monitoring, IEEE integrated smart monitoring technique for building fire prevention and disaster
Sensors J. 13 (3) (Mar. 2013) 909–915, https://doi.org/10.1109/ relief, Autom. Constr. 84 (Dec. 2017) 14–30, https://doi.org/10.1016/j.
JSEN.2012.2218680. autcon.2017.08.027.
[79] A. Costa, M.M. Keane, J.I. Torrens, E. Corry, Building operation and energy [96] F. Vandecasteele, B. Merci, S. Verstockt, Fireground location understanding by
performance: monitoring, analysis and optimisation toolkit, Appl. Energy 101 semantic linking of visual objects and building information models, Fire Saf. J. 91
(Jan. 2013) 310–316, https://doi.org/10.1016/j.apenergy.2011.10.037. (Jul. 2017) 1026–1034, https://doi.org/10.1016/j.firesaf.2017.03.083.
[80] A. Abdelalim, W. O’Brien, Z. Shi, Data visualization and analysis of energy flow [97] E.A. Parn, D. Edwards, Z. Riaz, F. Mehmood, J. Lai, Engineering-out hazards:
on a multi-zone building scale, Autom. Constr. 84 (Dec. 2017) 258–273, https:// digitising the management working safety in confined spaces, Facilities 37 (3/4)
doi.org/10.1016/j.autcon.2017.09.012. (Feb. 2019) 196–215, https://doi.org/10.1108/F-03-2018-0039.
[81] H.U. Gökçe, K.U. Gökçe, Holistic system architecture for energy efficient building [98] M. Arslan, C. Cruz, D. Ginhac, Semantic trajectory insights for worker safety in
operation, Sustain. Cities Soc. 6 (1) (Feb. 2013) 77–84, https://doi.org/10.1016/ dynamic environments, Autom. Constr. 106 (Oct. 2019) 102854, https://doi.org/
j.scs.2012.07.003. 10.1016/j.autcon.2019.102854.
[82] K. McGlinn, B. Yuce, H. Wicaksono, S. Howell, Y. Rezgui, Usability evaluation of a [99] D. Ghodasara, A. Patel, N. Bhatt, T. Thaker, Application of building information
web-based tool for supporting holistic building energy management, Autom. modeling in facility management: a case study of a commercial project’, in:
Constr. 84 (Dec. 2017) 154–165, https://doi.org/10.1016/j.autcon.2017.08.033. Proceedings of the Creative Construction Conference 2019, 2019, pp. 792–799,
[83] D. Lee, G. Cha, S. Park, A study on data visualization of embedded sensors for https://doi.org/10.3311/CCC2019-108.
building energy monitoring using BIM, Int. J. Precis. Eng. Manuf. 17 (6) (Jun. [100] D. Bhamare, M. Zolanvari, A. Erbad, R. Jain, K. Khan, N. Meskin, Cybersecurity
2016) 807–814, https://doi.org/10.1007/s12541-016-0099-4. for industrial control systems: a survey, Comput. Secur. 89 (Feb. 2020) 101677,
[84] I. Petri, S. Kubicki, Y. Rezgui, A. Guerriero, H. Li, Optimizing energy efficiency in https://doi.org/10.1016/j.cose.2019.101677.
operating built environment assets through building information modeling: a case [101] I.V. Lokshina, M. Greguš, W.L. Thomas, Application of integrated building
study, Energies 10 (8) (Aug. 2017) 1167, https://doi.org/10.3390/en10081167. information modeling, IoT and blockchain technologies in system design of a
[85] A. Bonci, A. Carbonari, A. Cucchiarelli, L. Messi, M. Pirani, M. Vaccarini, A cyber- smart building, Procedia Comput. Sci. 160 (2019) 497–502, https://doi.org/
physical system approach for building efficiency monitoring, Autom. Constr. 102 10.1016/j.procs.2019.11.058.
(Jun. 2019) 68–85, https://doi.org/10.1016/j.autcon.2019.02.010. [102] U. Rehman, S. Cao, Augmented reality-based indoor navigation: a comparative
[86] K. Rogage, A. Clear, Z. Alwan, T. Lawrence, G. Kelly, Assessing building analysis of handheld devices vs. Google glass, IEEE Trans. Human-Mach. Syst. 47
performance in residential buildings using BIM and sensor data, Int. J. Build. (1) (2017) 140–151, https://doi.org/10.1109/THMS.2016.2620106.
Pathol. Adapt. 38 (1) (Sep. 2019) 176–191, https://doi.org/10.1108/IJBPA-01- [103] J.W. Rice, The gamification of learning and instruction: game-based methods and
2019-0012. strategies for training and education, Int. J. Gam. Comp. Mediated Simul. 4 (4)
[87] B. Zhong, C. Gan, H. Luo, X. Xing, Ontology-based framework for building (Oct. 2012) 81–83, https://doi.org/10.4018/jgcms.2012100106.
environmental monitoring and compliance checking under BIM environment,

15