Int BIMSIGSLR

Archives of Computational Methods in Engineering
https://doi.org/10.1007/s11831-021-09545-2
ORIGINAL PAPER
Integration of BIM and GIS for Construction Automation, a Systematic

Literature Review (SLR) Combining Bibliometric and Qualitative
Analysis
Sina Karimi1 · Ivanka Iordanova1
Received: 15 June 2020 / Accepted: 10 January 2021

© CIMNE, Barcelona, Spain 2021
Abstract
For several decades now, the construction industry is suffering from low productivity, especially in comparison to manu-
facturing industries which have succeeded to benefit from digitalization of their processes. Furthermore, scarceness of
qualified workforce is expected in the near future. Construction automation is introduced as a solution to these challenges.
The capabilities of construction robots are improving at an accelerated pace. They are starting to be used in non-laboratory
contexts for automating processes ranging from infrastructure inspection to digital fabrication. One fundamental requirement
of employing robots in construction is their autonomous positioning. Building information modelling (BIM) and geographic
information system (GIS) are now a necessity for the construction projects. Integration between BIM and GIS provides
holistic digital representation of the built environment that robots could potentially utilize for positioning purposes. Preced-
ing this research, a number of reviews have been conducted on BIM–GIS integration, but none studied it from automation
perspective. This research addresses this deficiency through a systematic literature review of the state-of-the-art on BIM–GIS
integration with the purpose of robot positioning and navigation on construction sites. Using software tools and “science-
mapping” methods, 236 papers were explored. Trends, challenges, potentials, and deficiencies identified and mapped. Cita-
tion patterns of journal articles along with the analysis of studies; visualized and analyzed. Bibliometric analysis is followed
by a thorough qualitative analysis of the articles identified by the systematic methodology indicating limitations of current
studies such as vertical navigation, inaccuracy, dynamics of construction sites, indoor-outdoor navigation. Requirements for
robot positioning using BIM–GIS integration are defined.
1 Introduction employing traditional methods, lack of implementing indus-

trial approaches of construction processes, taking little bene-
Productivity has always been an issue in construction indus- fit from the use of digital tools and communication technolo-
try [1]. According to Scape Group, 58% of construction sup- gies [4]. Numerous studies, consequently, are carried out to
pliers and contractors identify scarceness of qualified work- tackle this issue. Barbosa et al. [5] propose adaptation of
force as the major challenge of improving the productivity technology, through leveraging cross-functional teams, and
of the construction industry in the near future [2]. Studies implementing brand new technology simultaneously with
indicate that the construction industry is falling behind the the training for it. Another study identifies the privileges
overall global improvement in productivity [3]. A great num- of applying Scrum strategy from design to the construction
ber of reasons have been identified, such as persistence of phase [6]. Agarwal et al. [7] have developed a framework
to better exploit and leverage current technologies namely,
‘rapid digital mapping’, ‘Building Information Modelling’
* Sina Karimi (BIM), ‘collaboration within a digital workplace’, ‘Internet
[email protected] of Things’ (IoT), and ‘future-proof design and construction’.
Ivanka Iordanova ‘Future-proof design’ is mainly referred to as future antici-
[email protected] pation design and development methods not to detriment
1 the future of the existing buildings [8]. Some researchers
Department of Construction Engineering, École de
Technologie Supérieure (ÉTS), 1100 Notre‑Dame St W, propose that the construction industry should undergo a
Montreal, QC H3C 1K3, Canada deep transformation in order to be able to adopt advanced
13
Vol.:(0123456789)
S. Karimi, I. Iordanova
technology. According to Bock [3], the required change in challenging. Wang et al. [12] argue that the different focuses
construction industry comes from the current emerging tech- of BIM and GIS causes the integration challenge. The for-
nology referred to as “Industry 4.0”. mer focuses on building components while the latter—on
In that direction, Bowmaster and Rankin [9] have recently geospatial information and environment around the building.
modified the multidimensional framework proposed by Fro- BIM is more concerned with the internal details of building
ese and Rankin [10], with the purpose to examine the level projects forming micro-level data, whereas GIS is special-
of maturity of Canadian construction industry in respect to ized in geospatial analysis. Nevertheless, BIM and GIS have
industry 4.0 technologies. The authors conclude that very great potential to be used together for the robot navigation
little research has been carried out with regard to ‘construc- and positioning. GIS would provide geo-referenced locations
tion-based’ automation and robotics in Canada causing a gap enabling robots to generate a navigation path, and BIM—
of ‘prototype development’ in ‘cyber-physical systems’ of semantic and geometrical information of the building or
navigating and positioning [9]. infrastructure, thus helping robots to detect obstacles and
Performing research on one of the pillars of Industry ultimately generating a navigable path.
4.0, García de Soto et al. [4] investigate the productivity This study is the first stage of a larger research project
of digital fabrication in construction industry with a robot aiming at using BIM and GIS for robots’ positioning on
fabricating a complex concrete wall. The results show higher construction sites in order to reduce the complexity of the
productivity when robotically fabricating a wall in compari- current navigation generating methods by using the common
son to a conventional method, and provide evidence that digital environment of the construction project. To be able
employing robots would enhance construction productivity. to define the research focus, initially, current studies in BIM
Actually, a robot can be associated with every on-site and construction robotics and their characteristics are inves-
inspection or digital fabrication practice, and a key part of tigated. Then, GIS and construction robotics are explored
the process is determining the robot’s position. Therefore, to examine how GIS can contribute to construction robots’
positioning of robots becomes a fundamental step in con- navigation. In addition, the related works with regard to
struction inspection or digital fabrication. BIM and GIS integration are studied to examine the current
A virtual representation of the project and its environ- solutions and to identify existing limitations. The research
ment can provide a holistic overview of the construction contributes to the scientific knowledge on robots’ navigation
in relation to the existing infrastructures and the surround- on a construction site by systemizing the-state-of-the-art in
ing environment. Today, two well developed technologies the domain, and by identifying the requirements of robot’s
namely Building Information Modelling (BIM) and Geo- navigation on construction sites, integrated with BIM and
graphic Information System (GIS) provide the digital envi- GIS. The practical application of the projected results will
ronment for facilitating the analysis and management of make possible the automatic construction of more complex
spatial and non-spatial data [11]. BIM, basically, represents and ’real-life’ building elements, integrating heterogeneous
geometric and semantic functions of construction projects building systems. Ultimately, this has the potential to affect
and provides a shared database enabling construction prac- positively the productivity, health and safety on construction
titioners to collaborate effectively [12]. BIM facilitates data sites, as well as the quality and sustainability of a project.
management of buildings’ lifecycle including design, con-
struction, operation, and maintenance of built assets [13]. On
the other hand, GIS provides location-related analysis along 2 Methodology
with spatial representation of built environment in various
fields of science [14]. Provided the capability of spatial data The methodology used in this study is Systematic Litera-
analysis by GIS, it is applicable to a broad range of practices ture Review (SLR), which is defined as the identification,
including in construction industry [12]. In addition, several evaluation, and interpretation of a field of research that can
GIS-based simulation studies have been conducted for vari- be reproduced with the same protocol by other researchers
ous purposes, which makes this field of knowledge more [23]. The utilized SLR employs a combination of qualitative
practical [15]. analysis [24] and bibliometric network visualization referred
Built environment stakeholders and geospatial special- to as “science mapping” [25]. The former focuses on quali-
ists have investigated the integration of BIM and GIS in tatively examining the papers collected through science
various research topics and practical applications, such as mapping co-occurrence method, and the latter provides a
Smart City [16], urbanization [17], internet of things (IoT) comprehensive overview of the status in the field. Common
[18], noise assessment [19], energy consumption [20], flood methods studied in science mapping are “keyword co-occur-
influence evaluation [21], and environmental data analysis rence,” “citation relations,” and “co-authorship relations.”
[22]. Despite the great benefits of BIM and GIS integra- Bibliometric network visualization facilitates the analysis of
tion, the process and methodology of such integration are a vast number of scientific networks by visualizing patterns
13
Integration of BIM and GIS for Construction Automation, a Systematic Literature Review (SLR)…
systematically in bibliographical databases [26]. Science As presented on the figure, the initial keywords co-occur-
mapping is capable of denoting the potentials of a specific rence analysis is performed on the Web of Science database.
field. In the context of the current study it is about the poten- It allows for new keywords to be added to the paper identi-
tials of BIM and GIS integration in relation to construc- fication process. This yielded 236 papers. At the next step,
tion robot navigation and positioning. Figure 1 illustrates after having identified the journals whose focus is the closest
the overall relationships between bibliometric analysis and to our target domain, the search is extended to include the
qualitative analysis. Keywords co-occurrence is the mutual following databases: Scopus, Engineering Village, IEEE and
step in both analyses. Its results in bibliometric analysis is the ISARC proceedings, thus resulting 1730 papers. After
used in qualitative analysis in order to identify the most rel- eliminating the duplicated and the inclusions, we obtain
evant articles. 1021 papers. Their titles, keywords and abstracts are then
Figure 2 illustrates the overall methodology used to iden- carefully read to determine their relevance to the targeted
tify the most relevant papers to be investigated in the context domain. Thus, finally, 64 scholarly papers are qualitatively
of the qualitative analysis of this study. In the first phase, this analyzed. The following points present in detail the meth-
methodology uses the keywords co-occurrence conducted in odology and the results of the science mapping part of this
the bibliometric analysis. research.
Fig. 1 Overall Relationship

of the SLR methodology with
bibliometric and qualitative
analyses
Fig. 2 Context determination

framework to search relevant
scholarly journals
13
3 Keywords Co‑occurrence Analysis identified and subsequently a qualitative analysis is con-

ducted to investigate features and functions in this regard.
The purpose of this analysis is to provide a holistic over- • GIS and Construction Robotics: (gis OR “geographic
view of construction robots’ navigation and positioning information system*”) AND (automat* OR “construct*
employing BIM and GIS integration. The search strategy robot*” OR “digital fabrication” OR dfab) AND (navi-
is initially to investigate robots’ navigation with BIM–GIS gat* OR traject* OR path*)
integration together. It gave an empty result (0 papers). Figure 4 demonstrates the hidden concepts of GIS
Hence, to attain the aforementioned goal, the bibliomet- and construction robotics and how GIS helps navigating
ric study is divided into 3 keyword clusters to investigate and positioning robots. Similar to BIM and construction
their relation with each other. In this step, each category is robotics keywords analysis, one important application of
explored to include the most relevant papers for the quali- GIS in robotics is “navigation”. The other concepts such
tative analysis presented in Sect. 4.2. as “path planning,” “digital elevation models,” “algo-
rithm,” “tracking,” “gps,” “remote sensing” play differ-
• BIM and Construction Robotics: (bim OR “building ent roles in GIS and construction robotics domain. To
information model*”) AND (automat* OR robot* OR understand the functions of each concept in navigating
“digital fabrication” OR dfab) AND (navigat* OR tra- construction robots with GIS, detailed qualitative analy-
ject* OR path*) sis is carried out (in Sect. 4.2) to identify features and
As illustrated on Fig. 3, the combination of BIM and methods presented in the papers.
construction robotics comprises various subdomains, • BIM and GIS: (bim OR “building information model*”)
which indicates the applications and the potentials of AND (gis OR “geographic information system*”)
BIM in construction automation especially in construc- Many subdomains of the BIM and GIS interaction are
tion robotics. Figure 3, also reveals various technologies revealed on Fig. 5. The purpose of conducting this analy-
employed for robots. It illustrates that “navigation” is sis is to collect data needed to integrate building informa-
a field of study that researchers work on, and suggests tion models and geographic information system in the
that the application of BIM in the construction industry desired research direction. Figure 5 indicates the current
can be related to robot navigation. Additionally, Fig. 3 applications of such an integration. “facility manage-
denotes adjacent subdomains around it, namely: “point ment,” “layout,” “smart city,” “optimization,” “integra-
cloud”, “path planning”, “indoor navigation”, “indoor tion,” and “indoor” are some of those. It is also important
modeling”. To explore the application(s) of BIM in CR, to mention that IFC (Industry Foundation Classes) and
scholarly papers categorized under each subdomain are CityGML (City Geography Markup Language) are open
standard data model and exchange format for BIM and
Fig. 3 Keywords network of

BIM and construction robotics
13

GIS and construction robotics

BIM and GIS
GIS respectively. Both appear in the keywords network ied qualitatively to provide a comprehensive understand-
as subdomains. As the two previous categories presented, ing of current contributions and requirements of integrat-
journal papers categorized under each subdomain is stud- ing them.
13
All the sets of data derived from Web of Science Core Col- of research for qualitative analysis. The fundamental step
lection, initially, are submitted to VOSviewer to construct of bibliometric analysis is to determine the contexts of
a co-occurrence network of keywords, subsequently, are research, which are relevant to the objective of the survey.
imported to Gephi to create customizable and detailed visu- In other words, the choice of keywords for the search in the
alization and more importantly to run further analysis. With databases, models the entire bibliometric analysis [26].
Gephi computing tool, similar concepts such as “building
information modelling”, “building information modelling” 3.1 Identification
and BIM are merged.
Cobo et al. [26] provide a review of a number of avail- Once the keywords network is formed, the subdomains and
able tools for the purpose of bibliometric analysis, namely: their keywords identified, the search proceeds a step for-
Bibexcel, CiteSpace, CoPalRed, IN-SPIRE, Leydesdorff’s ward and refines the context determination to the desired
Software, Network Workbench Tool, Sci2 Tool, Vantage- ones. Based on the procedure described in Sect. 3.1, Web
Point, VOSViewer. The authors conduct a survey to find the of Science Core Collection refinement provides the articles,
advantages and drawbacks of each of the above-mentioned which are the relevant ones to the scope of the study. It is
tools. In addition, Van Eck and Waltman [25] compare two noteworthy to mention that the identified papers are pub-
“general network analysis tools”—Pajek [27] and Gephi [28] lished by 2020. The papers collected by now are the only
in order to analyze the visualize networks. The authors con- ones available in Web of Science Core Collection, but they
clude that Gephi provides more tools on detailed customiz- still do not provide a comprehensive outlook to the field, so
able visualization compared to Pajek [25]. According to the the search is extended to other databases.
above-mentioned analysis and for the purposes of this study,
VOSViewer, CiteSpace, and Gephi are selected to perform 3.2 Extension
the bibliometric analysis.
To enable the literature to synthesize as many as possible
• VOSviewer is devised to constitute and map bibliometric related works, Scopus, Engineering Village, and IEEE (Insti-
data [29]. tute of Electrical and Electronics) are added to the search.
• Gephi enables the researcher to carry out deeper analy- The I.S.A.R.C. (International Symposium for Automation
sis on mapped graphs and to make modifications to the and Robotics in Construction) proceedings are included
networks [30]. too, as the papers published there represent the advances,
• CiteSpace analyzes the trends developing in a specific contributions, and concerns of the researchers for all fields
domain. It also manages to visualize various network lay- of construction with great concentration on Construction
outs, detects clusters and analyzes within a given time Automation, Robotics, IT, etc. [32].
period [31].
3.3 Eligibility
The Web of Science Core Collection is selected as the
preliminary database to run keywords co-occurrence by Eligibility comprises two main steps, which are the identifi-
VOSviewer due to its flexibility to search various combina- cations of duplicates and inclusion. The former identifies and
tions of terms, its thorough journals [1] and its compatibil- subsequently removes the duplicated articles from the data-
ity with VOSviewer computing tool. Furthermore, Web of base and the latter only brings the papers which are precisely
Science Core Collection enables authors to investigate the to the point of current study into consideration, which is the
peer-reviewed, high quality scholarly articles from all over application of BIM and GIS integration in robots’ position-
the world. Other widely known databases such as Scopus, ing on construction sites. Although the number of scholarly
Engineering Village, IEEE, and I.S.A.R.C. (International papers remarkably increased in the extension phase, there
Symposium for Automation and Robotics) proceedings are are for sure many articles, which are duplicated in the dif-
included later to make this research as thorough as possible ferent databases. To tackle this problem, all the databases’
(see Sect. 3.3). information is downloaded and is converted to.csv format
in order for Excel to identify the duplicate ones and remove
A keywords network is generated by running co-occur- them. The I.S.A.R.C proceedings do not provide such export
rence type of analysis on the dataset to constitute a graph format so this procedure is done manually.
based on the keywords. The nodes of the graph indicate the The final step is to submit only those articles studying,
fields of research and the subdomains of which they consist. partially or thoroughly, the focus of the current research to
This is useful for identifying underlying concepts, adjacent qualitative analysis. To reach this objective, all the articles
topics and hidden links between themes; to illustrate the available so far are filtered based on their titles, keywords,
potentials and more importantly to determine the context and abstracts.
13
4 Findings file exported from VOSviewer. Within Gephi environment,

similar areas of studies (such as “geographic information
4.1 Bibliometric Analysis of BIM–GIS Integration system (gis)”, “geographic information system”, and “gis”)
are merged. The result is a graph comprising 30 nodes illus-
The Science mapping provided in this section is conducted trated in Fig. 7.
on 236 papers of BIM and GIS from Web of Science Core Gephi is capable of analyzing various statistics on a
Collection to provide a generic comprehensive overview of given network. The weighted degree of a node represents
the field. The first study on Building Information Modeling the weighted number relations (edges) it has [35]. In other
and Geographic Information System occurred in 2008, car- words, the higher the number of weighted degree of a rela-
ried out by Lapierre AND Cote [33], and is published by tion is, the more influential that domain is. A ‘data labora-
“URBAN AND REGIONAL DATA MANAGEMENT.” As tory’ of Gephi consisting of the metric analysis of the graph
it is illustrated in Fig. 6, from 2008 to 2019, there has been on Fig. 7 is presented in Table 1. Moreover, different layouts
an important growth of the research in the field. In 3 years, are available for different purposes in accordance with fea-
during 2014 to 2017, a sharp increase has occurred which tures of topologies [36]. The current analysis emphasizes the
implies an exceptional interest of researchers in BIM and rankings of the research areas. The visualization of the data,
GIS together. It also interesting to note that based on the therefore, is based on the ranking of the nodes.
forecasting line provided, it is predicted that this growth con- Based on the information provided by bibliometric analy-
tinues. New fields of BIM and GIS integration might emerge, sis, the following conclusions can be made:
or current solutions might be considerably improved.
1. The integration of BIM and GIS environments in con-
4.1.1 Keywords Co‑occurrence Analysis struction has attracted a great amount of attention in
recent years, but its potential applications in construc-
Following publishers’ requirements, the authors of scientific tion automation and construction robots’ positioning has
papers indicate their research focus through keywords. In not been studied. As seen from the keywords subdo-
bibliometric studies, the analysis of the keywords shows the mains table (apart from BIM and GIS, which are the
width of the research domain, and draws the boundaries in focal fields of this literature review), “citygml”, “man-
that specific domain [34]. Graphs constituted by related key- agement” and “interoperability” attracted higher atten-
words illustrate the relationships among subdomains existing tion. In addition, analysis of the node weighted degree
in the studied field [25]. Hence, an analysis is performed reveals that these three research areas have much higher
based on the keywords co-occurrence method. Every couple relative importance compared to all the others. On the
of nodes (keywords) is linked via an edge and each edge car- other hand, less attention has been paid to “cultural her-
ries a weight. The number of publications in which two key- itage”, “methodology”, and “smart city” indicating that
words occur together is represented by weight metrics [25]. researchers investigated these research areas less fre-
In this research, VOSviewer software visualizes and quently within the body of the existing literature. More
shapes the networks of subdomain studies based on the data importantly, Fig. 7 and Table 1 denote that construction
retrieved from Web of Science Core Collection. Gephi is automation is not one of the studied sub-domains and
employed in order to conduct further data analysis of the hence, requires more attention from researchers.
2. The research areas which are in the middle of Table 1
depict the potential areas in BIM and GIS integration.
“Augmented Reality”, “Facility Management”, and “lay-
60 out” are examples of such potentials. However, other
50 areas, which are not listed in Table 1 are either not inves-
Number of articles published
tigated or received much less attention, which shows the

40
research gaps in this field.
30 Other finding of this scientometric analysis reveal that
a great deal of research is directed towards the “man-
20
agement” aspect of construction projects. Table 1, how-
10 ever, denotes that many authors have contributed to the
domain by investigating “interoperability” of BIM and
0
2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 GIS, indicating one of the obstacles for the integration
of the two digital environments. On the other hand, the
Fig. 6 Distribution of papers over years (Source: Web of Science “smart city” facet of BIM and GIS integration has one
Core Collection) of the lowest weighted degrees (see Table 1) revealing
13
Fig. 7 Core research focus—

BIM and GIS
that applications of BIM and GIS integration have been common method for network of co-citation analysis [38].
more studied in “building” projects rather than in “city Figure 8 visualizes the paper clusters computed by CiteS-
planning” even though it could be utilized for both small pace, using the Log-Likelihood Ratio (LLR) algorithm. In
and large scale projects. statistical analysis, a log-likelihood ratio is a test to identify
3. As shown on Fig. 7, “ifc (Industry Foundation Classes)” a null model against an alternative model [39]. LLR algo-
and “citygml (City Geographic Markup Language)” are rithm is mainly used to calculate p-value to decide on rejec-
two data schemas being used for BIM and GIS inte- tion of a null model [39]. In this regard, after using “filter
gration. The former is the open-format standard data out small clusters”, eight clusters (out of 61) are detected
schema for BIM and the latter has been developed for as the main research areas where cluster #0 is the largest in
GIS interoperability. The bibliometric visualization of terms of size, indicating that this cluster contains the larg-
the literature database denotes that every region, where est number of publications, while cluster #8 is the smallest
several nodes are located close to each other, establishes cluster of the important ones. The labels attached to clusters
relationship within the area. For instance, “integration,” of Fig. 8 are proposed by CiteSpace. It should be mentioned
“interoperability,” “ifc”, “citygml” and “semantic web” that CiteSpace focuses on formation of clusters rather than
are located close to each other within the network (see on the underlying contents in the given clusters [37].
Fig. 7) indicating that authors used semantic web tech- Metrics evaluated by CiteSpace computing tool are Mod-
nology to enable integration of BIM and GIS with IFC ularity Q = 0.7127 and Mean Silhouette = 0.5206. The metric
and CityGML schemas. modularity (0 < Q < 1) indicates to what extent a network is
capable of being independent [40]. The amount of modular-
4.1.2 Document Co‑citation Analysis ity represents the quality of a network’s structure meaning
that modularity close to 1 indicates a network is well-struc-
The document co-citation method reveals citation patterns tured while modularity close to 0 illustrates unclear cluster
among research studies and provides information regard- boundaries within a network [37]. Table 2 illustrates the
ing the intellectual structure of the studies [37]. Creating details of clusters retrieved from CiteSpace.
a network of document co-citation analysis is a common The other metric “Silhouette” represents the uncertainty
approach for providing this kind of information via science of a given cluster and is ranged from − 1 to 1, meaning that
mapping [37]. CiteSpace is the selected software to conduct a silhouette close to 1 indicates a cluster well separated from
this analysis by creating citation clusters which is the most other clusters, whereas a silhouette close to − 1 introduces
13
Table 1 Ranking of the subdomains in relation to BIM–GIS integra-

tion
Label Degree Weighted degree
BIM 31 52.96
GIS 31 16.38
CityGML 21 15.33
Management 23 12.98
Interoperability 19 12.21
Model 29 11.93
Construction 23 11.48
IFC 15 11.11
Facility management 19 10.32
Support 20 9.49
System 21 9.37
Framework 18 9.00
CAD 9 7.00
Information 17 6.66
Semantic web 19 6.50
Design 17 6.40
Integration 17 6.35
Ontology 14 6.35
Performance 9 5.82
Technology 12 5.52
Building 10 5.21
Visualization 13 4.96
Indoor 16 4.77 Fig. 8 Clustering structure of BIM and GIS integration
3D GIS 8 4.49
Layout 13 4.34
Table 2 Core clusters of document co-citation analysis if BIM and
Augmented reality 14 4.26 GIS
Optimization 9 4.05
Cluster ID Size Silhouette Mean year Focus of cluster
Cultural heritage 7 4.00
Methodology 10 3.09 0 89 0.705 2014 Geographic information
Smart city 8 2.76 system
1 29 0.918 2006 Construction safety
2 29 0.838 2008 3D GIS
3 28 0.845 2013 Facility management
heterogeneity of members within a given cluster [41]. By
4 28 0.800 2011 Key factors for BIM
interpreting aforementioned information of the study (Fig. 8; adoption
Table 2), the following results can be formulated: 5 27 0.954 2008 Conservation
6 22 0.870 2010 Flood damage assessment
4. By applying document co-citation analysis, the number 7 15 0.857 2014 3D modelling
of articles considered for the clustering are more than
the number of articles in the database indicating the fact
that there are some that appear in more than one clus- 5. An overview on the Mean Year indicates on what sub-
ter. This indicates that the studies have high integrity domain researchers focused during the years. Document
and research endeavors took benefit from the previous co-citation analysis of BIM–GIS integration reveals that
ones meaning that the research in BIM–GIS integra- the recent attempts are mainly slanted towards Geo-
tion is built on the studies conducted before. Results of graphic Information System, 3D Modelling and Facility
studies conducted in the field of BIM and GIS reflect Management. Also, construction safety appears as the
that researchers exchange their ideas and focus on the one of the earliest applications of BIM–GIS integration
field. However, investigators have not studied potentials studied by researchers. Other research focuses are also
of BIM and GIS integration in relation to construction complied in the list on Table 2. Citation patterns of BIM
automation, which needs to be investigated. and GIS integration (see Fig. 8) reveal that authors of
13
existing literatures have not investigate the construction where to publish their studies in order to reach more poten-
automation relations with BIM–GIS integration domain tial readers in their field [43].
and there exists a lack of exchange of ideas between Among Gephi’s statistics, HITS (Hyperlink-Induced
these two domains. Topic Search) algorithm is capable of rating web pages
6. Given the structure of the clusters demonstrated in [44]. The HITS metric calculates two values for a web page;
Fig. 8, BIM–GIS studies indicate a relatively high struc- authority and hub value. By estimating “authority” Gephi
tural integrity. Silhouette values indicated in Table 2 provides each node (journal) a score indicating the value of
show that clusters of the visualized network are con- the content within the journal. It also provides a “hub” value
nected through citations inside and outside of their clus- for outgoing links of each node (journal) indicating the value
ters. As Hicks [42] argues, such structure of clusters of links [45]. Figure 9 illustrates 10 prominent journals in
occurs when authors cite studies from other clusters, the field of BIM and GIS integration ranked based on the
which creates a well-formed citation pattern of a given score calculated by Gephi using the HITS algorithm.
field. Therefore, as BIM–GIS forms well-structured Table 3 also illustrates the top journals of the field accom-
clusters, this draws a promising future of the field. panied by their rank, “authority” value, “hub” score, and
the main research areas of each one. The most important
outlet of the field is by far “Automation in Construction”
4.1.3 Direct Citation of Sources with highest hub score (0.772) and highest weighted out
degree (71.0). To provide better insight of each journal’s
Direct citation of sources indicates prominent journals in a research areas, their subdomains are also listed in the table.
field of study [25]. Identification of prominent journals is This helps authors and readers to refine their choice of jour-
beneficial to readers, authors, and editors. It enables readers nals to either publish their work or to read about their field
to select which journals are focused on their field of study of research. The HITS analysis in Gephi shows the following
in order to find creditable articles, and it indicates to authors results:
Fig. 9 Graph of prominent

journals in BIM & GIS
Table 3 Prominent journals of BIM and GIS

Rank Journal Weighted Authority Hub Research areas
out degree
1 Automation in construction 71 0.043792 0.772793

Construction building technology, engineering
2 ISPRS International Journal of Geo-information 22 0.486019 0.379324
Physical geography, remote sensing
3 Building and Environment 9 0.222176 0.358016
Construction Building Technology, Engineering
4 Computers in Industry 8 0.325105 0.267654
Computer Science, Engineering
5 5th International Conference on 3D Geo-informa- 8 0 0.152319
Image Science Photography Technology, Physi-
tion cal Geography, Remote Sensing
6 eWork and eBusiness in Architecture, Engineering 6 0.222176 0.139729 Computer Science, Construction Building Tech-
and Construction nology, Engineering
7 Journal of Computing in Civil Engineering 1 0.434159 0.127924 Construction Building Technology
8 Buildings 0 0.222176 0 Computer Science, Engineering
9 Journal of Information Technology in Construction 0 0.444958 0 Engineering
10 Journal of Spatial Science 0 0.33123 0 Physical Geography, Remote Sensing
11 Urban and Regional Data Management 0 0.109054 0 Remote Sensing, Engineering
13
The analysis of the network in Fig. 9 can be interpreted novel approaches, or recently developed ideas. Luwel [47]
as follows: argues that co-authorship analysis provides thorough inves-
tigation of scientific collaboration which ultimately leads
7. The majority of the articles in BIM and GIS integra- to higher productivity of the work, higher citation, and
tion are published in two journals namely, Automation attraction of attention. Based on what is discussed above,
in Construction and ISPRS International Journal of the following part of the literature analysis is divided into
Geo-information. It is worthy to note that comparison 3 sections namely, pioneer researchers, pioneer institutions,
between “weighted out degree” values of these journals pioneer countries with regard to BIM and GIS integration.
shows that flow of information begins from Automation
in Construction rather than ISPRS International Journal 4.1.4.1 Pioneer Researchers As Fig. 10 illustrates, there are
of Geo-information with a high difference (see Table 3). two major clusters of collaborating researchers in the refer-
8. The investigation of BIM and GIS integration with ence field. Each of the clusters introduces, directly or indi-
regard to construction automation is a focus of study of rectly, prominent authors in BIM and GIS integration. Direct
none of the journals. This fact corroborates the findings indication of co-authorship refers to papers the authors pub-
of current literature in previous sections that potentials lished in collaboration, while indirect co-authorship refers
of BIM and GIS integration from construction automa- to having mutual co-authors. Additionally, HITS algorithm
tion view has not been studied sufficiently. This fact is run to rank the prominent authors through applying
confirms that more attention needs to be paid for appli- authority scores [45]. Quality of the connected nodes with
cations of BIM–GIS in construction automation. link to other influential nodes of the graph constitutes the
authority score [48]. Gephi ranks the nodes with authority
4.1.4 Co‑authorship Analysis score to visualize the prominent researchers in terms of their
influence to the field. Thus, the following interpretation can
Conducting a co-authorship analysis enables researchers be made of the results shown on Fig. 10:
to explore and investigate the collaboration networks of
pioneer researchers, institutions, and countries to acquire 9. The majority of the BIM and GIS integration researchers
more profound knowledge of the field, develop expertise, collaborate. However, an integral network of collabora-
increase productivity, and decrease isolation [46]. Addition- tion is far to be present. Some isolated authors should
ally, it would be worthy to the scientist who are carrying out identify the collaboration networks in order to be able
research in a specific field to identify the prominent research- to enhance their productivity. Nearly 40% of the authors
ers, institutions, and countries to keep track of innovations, have established a strong relationship working in BIM
Fig. 10 Prominent researchers in BIM and GIS
13
and GIS, which provides prospects for improving pro- (as shown on Fig. 12). Co-authorship analysis of countries
ductivity of the field in future. Nevertheless, there are a can contribute to redefine strategies and to establish policies
few authors who do not belong to any cluster reflecting to improve productivity. Based on the statistical analysis and
the fact that those authors of the network (Fig. 10) are the graph illustrated in Fig. 12, the results can be interpreted
carrying on the research in isolation. as follows:
11. Taiwan is the only country, which does not work in

4.1.4.2 Pioneer Institutions Similar to individuals’ col- collaboration with other prominent countries, while
laboration in the BIM and GIS domain, a network of insti- the United States connects with all countries. In addi-
tutional collaboration can be created to identify the promi- tion, the flow of information correlates with the USA,
nent universities and institutes around the world. As Fig. 11 which is the most prominent country in BIM and GIS
illustrates, the HITS algorithm ranks the institutions based integration.
on the “hub” score to show the influence of the nodes’ actors 12. According to the average degree values calculated by
[45]. The size of the nodes represents the “hub” score show- Gephi, Canada can be categorized as a country where
ing the influence of the institution on others. Interpretation BIM and GIS integration has not been greatly studied.
of the HITS algorithm analysis along with the network dem- About 5% (13 out of 236) of the investigated articles
onstrated in Fig. 11, shows that: are developed in Canada. The names and affiliations
10. Apart from the four isolated ones, the majority of the
institutions working on BIM and GIS collaborate.
However, this collaboration does not establish a strong
relationship among them (visualized by the low num-
ber of connections). This indicates one of the problems
in the field. Solving this lack of collaboration has the
potential to result in significant progress and produc-
tivity improvement of the BIM and GIS integration
field.
4.1.4.3 Pioneer Countries Following a procedure similar to

the one identifying prominent researchers and pioneer insti-
tutions, Gephi reveals the influential countries in the BIM
and GIS integration domain. Directed and undirected edges
map the flow of information and closeness among countries Fig. 12 Pioneer countries in research on BIM and GIS integration
Fig. 11 Pioneer institutions of BIM and GIS
13
of the authors with more than 2 published papers are framework for digital twins of factories improves the IFC-
listed in Table 4: format imported files by initially importing the ifcXML file
to IFC Engine [52] and subsequently submitting the file into
The data on Table 4 implies that BIM and GIS integration VEROSIM [49]. VEROSIM supports IFC and CityGML
field has not been broadly studied in Canada, so more inves- files to run simulation for spatial analysis [53]. Delbrüg-
tigations needs to be carried out in this domain. ger et al. [49] classify building components as navigable
surfaces and objects as either obstacle or agent. Obstacles
4.2 Qualitative Analysis can be either static or dynamic. They have also developed
a scene content containing all the possible components in
Section 3.1 of the current literature review adopted biblio- factories and mapped each element to navigation criteria.
metric analysis to investigate the integration between BIM Ibrahim et al. [54] have developed interactive model-
and GIS. The systematic literature review identified the based path planning for Unmanned Aerial Vehicles (UAV)
most relevant articles to the research topic to study in depth to capture data on construction sites. They use a semi-
the contributions to the field. In what follows, these papers automated approach with a drone to capture visual data
are classified into three categories for explicit distinction on construction sites. Interoperability problems have been
between different subdomains, namely: BIM and robotics, eliminated using a web platform technology. With a mobile
GIS and robotics, and BIM and GIS in relation to robots’ application, used to plan the flight, the proposed system
positioning and navigation on construction sites. integrates web platform with visual model derived from the
UAV’s camera to compare the construction progress with the
4.2.1 BIM and Robotics schedule. A BIM model is used to plan the aerial trajectory
in order to inspect the related sectors of the construction
The purpose of the current literature review is to investi- sites.
gate the potentials of BIM–GIS integration in relation to Darwish et al. [55] have created a framework in which
construction automation especially in robot’s navigation on RGB and depth (RGB-D) sensors are used to visualize
construction sites. After having a comprehensive bibliomet- indoor environments, taking structural constraints into
ric analysis of the BIM and GIS integration domain, this account. They propose two main purposes of the RGB-D
section is to qualitatively investigate, in depth, the previ- sensors namely, robot collision avoidance within indoor
ous contributions of researchers regarding the applications environment [56] and 3D model reconstruction [57]. Dar-
of BIM in robots’ navigation. The application of GIS is wish et al. [55] focus on the latter application in their pro-
investigated afterwards. This section describes the previ- posed framework, which considers all the features in RGB
ous research attempts followed by the interpretations of the and depth images to reconstruct an indoor environment.
studied papers. Nahangi et al. [58] present a method with which UAV
Delbrügger et al. [49] have developed a BIM-based navi- localization and navigation can be tackled in GPS-denied
gation framework for digital twins of factories. It comprises indoor construction environments. The proposed method
of building information model and factory equipment classi- uses connected coordinates of BIM model with AprilTags.
fied as fixed and dynamic. In addition to BIM, the proposed With UAV’s camera, the data of tags are captured and are
framework incorporates pathfinding technology. The authors transformed so that the UAV can localize. The authors con-
have developed a navigation core utilizing two approaches firm that the Global Positioning System (GPS) is accurate
namely, corridor map method and navigation mesh. The for- and reliable for outdoor environments, but inefficient for the
mer refers to convex polygons covering navigable surfaces, indoor ones. AprilTags is a visual fiducial system in which
which are mainly triangles [50] and the latter represents the tags can be ordinarily printed. The coordinates of the
edges as navigation corridors with a free space for colli- tags correspond with those in a BIM model so that the UAV
sion-avoidance provided by a sphere [51]. The navigation is able to localize itself in global coordinate system [58]. In
Table 4 Prominent researchers Author Affiliation Number of publica- Publication year

of Canada in BIM and GIS tions
integration
Hammad. A Concordia University 2 2016, 2017
Salimzadeh. N Concordia University 2 2016, 2017
Pottinger. R University of British Columbia 2 2017, 2018
Staub-French. S University of British Columbia 2 2017, 2019
Zadeh. PA University of British Columbia 2 2017, 2019
13
order to localize the UAV, an on-board camera is employed pose estimation accuracy. In this method, the authors employ
to detect the AprilTags. BIM to identify coordinates of fiducial markers with a UAV
Lin et al. [59] have developed a method for automatic equipped with camera to identify the relative pose. All the
generation of indoor environment employing BIM and GIS information is ultimately put together to calculate the coor-
at geometry level. With integration of BIM, GIS and i-GIT dinates in the global coordinate system. They use EKF to
algorithm, they generate several possible routes for naviga- consider uncertainty, which is the characteristic of construc-
tion purposes. To accomplish the automatic indoor navi- tion sites. BIM, in this approach, is specifically used to help
gation, they have developed a collective algorithm named the drone to identify the 3D coordinates of the AprilTags.
Intelligent Generation of Indoor Topology (i-GIT) which Neges et al. [64] have developed a system based on the
supports IFC schema and automatically generates space Bluetooth Low Energy (BLE) to improve the quality of
boundaries for vertical and horizontal navigation. A set of indoor location tracking on construction sites and for facil-
algorithms are employed to generate floor-level paths and ity management purposes. The authors classify the work
non-planar paths and to reduce the complexity and redun- spaces systematically using a building information model
dancy of path nodes. The former refers to horizontal naviga- and, then, place BLE beacons on different locations. The
tion while the latter responds to vertical navigation needs. result of the study shows that the functionality of the pro-
ESRI ArcScene is utilized to identify space boundaries of posed method is limited. The experiment also indicates that
the IFC file and the algorithms were run in that environment. the signal strength and method robustness is greatly affected
Siemiątkowska et al. [60] adopt a semantic approach by the dynamic nature of construction sites and facilities.
based on a BIM model for robot delivery indoor naviga- Palacz et al. [65] propose a method to navigate indoor
tion. They have developed a hierarchical action planning mobile robots using the IFC schema of BIM and a graph,
to incorporate time-optimized robot navigation including combined with artificial intelligence. They argue that the
two technologies namely, object detection based on point structure of buildings and the semantics of building com-
cloud and object detection based on image. Both horizontal ponents have great influence on the possible routes between
and vertical navigation are taken into account and are tested two points. This graph-based navigation approach assigns
within dynamic environment. They employ BIM to extract attributes to graph nodes and graph edges. The former con-
semantic information and indoor topology in order to be able tains semantics of building elements and the latter stores the
to run the hierarchical action planning. cost of navigation between spaces dependent on different
Hamieh et al. [61] have developed BIM-based indoor variables namely, opening width, lift existence, space dis-
path planning method named BiMov using four definition tance, and door types. The authors argue that the informa-
phases. The first phase identifies all possible paths within a tion derived from IFC schema only includes construction
space using algorithms run on an IFC file. The second phase elements, while other elements such as chairs, anything left
reduces the number of paths by discriminating between a on the floor, etc. should be considered. Hence, an additional
mobile object, a person, or a bulky equipment. The possible algorithm could the robot navigation for passing obstacles.
paths within a space are further refined in third phase based This contribution assumes the robot has such algorithm built
on the content of the path, which can be influenced by the in [65].
presence of machinery or restricted areas. Phase 4 considers Kim et al. [66] have provided a method in which a mobile
number of paths affected by real-time situation or building’s robot collects spatial data specifically developed for con-
passages. Although the authors presented four-stage plan- struction sites with many uncertainties. The proposed sys-
ning regarding robot navigation within indoor environments, tem uses Simultaneous Localization and Mapping (SLAM)
they put emphasis only on the first two phases. techniques to build a map of construction site for navigation
Quintana et al. [62] have developed a method using BIM, through point clouds. However, the SLAM technique does
3D laser scanner, and a color camera to detect 3 positions not provide obstacle-awareness for a mobile robot so that
of a door within indoor environment. To attain this goal, kinematic modelling of the robot is analyzed. The authors,
they integrate geometry and color information obtained then, develop an algorithm based on fuzzy control to navi-
from the environment to detect the angle of a given door, gate the mobile robot in unknown environments with obsta-
and identify it as open, semi-open, and closed. Their sys- cles. In this method, real-time 3D environment reconstruc-
tem also provides an accurate position of the door in the tion based on laser scanning is used instead of a building
world-coordinate-system. information model [66].
Kayhani et al. [63] assess the Extended Kalman Filter Based on the literatures reviewed the following findings
(EKF) to improve indoor localization using AprilTags. They are presented:
use that improvement to navigate Unmanned Aerial Vehicle
(UAV) within an indoor environment. They adopt a proba- (I) Many indoor navigation and localization methods
bilistic approach towards data fusion in order to improve are introduced and investigated especially with
13
integration to BIM. Indoor localization methods 4.2.2 GIS and Robotics

can be classified in three main categories [67]:
This section of the literature review studies qualitatively
• Wave-based propagation: comprising devices receiv- the applications of GIS in robots’ navigation on construc-
ing waves of two frequency ranges, namely: ultra- tion sites independent from BIM. In this part, the papers
sonic and sound waves. Different types of receivers identified in the methodology section are investigated
are also presented—such as radio frequency (RF), in depth to have a comprehensive overview of GIS and
ultra-wideband (UWB), and wireless local area robots’ positioning on construction sites. In what follows,
network (WLAN) [68–70]. Nonetheless, research- the papers are described and later, discussed.
ers studying the abovementioned techniques have Mangiameli et al. [72] have developed a method based
reported several limitations with regard to accuracy. on GIS for generation of raster maps showing obstacles
Infrared accuracy is reported at room-level (i.e. its in urban areas. The authors develop this method to enable
accuracy is limited and it would not be a functional flight planning of an Unmanned Aerial Vehicle (UAV).
option for larger spaces) and its performance is dis- Their approach first represents the building data as a vec-
rupted by sunlight. WLAN and RFID accuracy is tor data shapefile, and then, converts it to raster to be able
insufficient and varies from 4 to 9 m. UWB, in con- to use GIS. The authors use Spline algorithm to extract
trast, provides 9 cm accuracy but it is expensive and buildings’ height for the raster map, identification of pos-
requires complicated deployment of transmitters, sible path within urban environment, and conversion of the
which makes it inefficient. path identified to waypoints for navigation of UAV. The
• Image-based localization: relies on computer vision possible obstacles are determined and subsequently are
techniques and image matching. Computer vision georeferenced in order to be avoided.
techniques, itself, are categorized into two methods: Zaki and Dunnigan [73] identify three challenges to
Global and Local. The former refers to detecting navigate autonomous robots, namely: representation and
edges and recognizing features, whereas the latter schemes, planning algorithm, and the integration architec-
detects landmarks with the help of tags and images. ture of both. They combine GIS modelling and descrip-
The reported studies indicate that these methods suf- tion logic for representation and schemes, modify and fuse
fer from lack of precision and, more importantly, are algorithms, and ultimately introduce a navigation archi-
not appropriate for dynamic environments such as tecture. GIS and ontology are used to constitute digital
construction sites. representation of dynamic data. The proposed framework
• Inertia-based localization: this method uses an ini- does not consider neither vertical navigation nor moving
tial location and navigates through accelerometers, obstacles.
inertia measurement units (IMU), and other motion Yang et al. [74] propose a GIS platform in which the GIS
detectors. Ibrahim and Moselhi [67] have developed database is modified. The authors introduce properties such
a localization technique, which combines IMU and as road width, lane number, lane info, and if-traffic to be
Kalman Filter. Their method produce a higher accu- able to describe the environment of the Unmanned Ground
racy compared to ultrasonic and sound waves, but Vehicle (UGV). In addition, they redefine road models and
it remains yet inefficient due to the very demanding turning strategy to generate a cost map navigating system for
computational calculations. UGV navigation in urban environments.
Fernández-Caramés et al. [75] introduce a method for
(II) Researchers use algorithm-based approaches to integration of indoor localization approach and GIS for the
navigate mobile robots. Taneja et al. [71] catego- purpose of real-time navigation. They employ GIS to ana-
rize them into three major classes namely: center- lyze indoor spatial data and develop a method that detects
line based, metric-based, and visibility-based. a door within indoor environment with data fusion of laser
Center-line based algorithms select the medial axis and vision sensors. Extended Kalman Filter is used for path
of an indoor space, metric-based algorithms move finding.
along the navigable boundaries of a given space, Mirats Tur et al. [76] have developed a map-based naviga-
and visibility-based algorithms are comprised of tion system in urban environments using GIS. The proposed
nodes and edges representing the end points of a system enables a robot or a team of robots to navigate within
path and the visible lines of a given indoor environ- urban environments with prior assumption that an under-
ment respectively. The limitation associated with standable navigation map is available. In this system, robots
these algorithms is that they do not consider the can connect to the map and navigate based on it. The authors
dynamic objects and furniture of an indoor environ- highlighte communication protocols and cooperation issues
ment to which construction sites are subject. as the important aspects of the work.
13
Sun et al. [77] have proposed GLANS (GIS Based algorithms is that they are not mutually exclusive,
Large-Scale Autonomous Navigation System) for robot thus combinable.
navigation in urban settings. They argue that current
simultaneous localization and mapping (SLAM) tech- Moreover, path planning algorithms are divided into five
niques cannot be utilized for large-scale environments. In categories namely, (1) “sampling-based algorithms,” (2)
this method, a GIS database suggests a topological path “node-based optimal-based algorithms,” (3) “mathematical
on which the mobile robot can navigate, detect obstacles model based algorithms,” (4) “bio-inspired algorithms,” and
and consequently modify the path. Moreover, the adjust- (5) “multi-fusion based algorithms” [73].
ment results can be shared with other mobile robots so
that the navigation and localization process is optimized. 4.2.3 BIM and GIS (In Respect to Construction Automation)
Their method is independent of the Global Positioning
System (GPS). The main objective of this section is to identify the require-
Park et al. [78] have developed a GIS-based method to ments of BIM and GIS integration for the purpose of con-
analyze trafficability of terrain for autonomous robot navi- struction robot navigation and positioning. To attain this
gation. In this method, GIS is employed to analyze the pos- goal, first, the tools and methods for integration of BIM
sibility of having a piece of terrain under traffic of unmanned and GIS are presented. Then, a comparison of the differ-
ground vehicles by generating grid maps. The GIS database ent methods and tools will determine whether one of them
analyzes the spatial data of a given environment and assigns is appropriate for the future purposes of this research, or a
a cost to each grid. Once all the grids are assigned with a novel approach should be developed to assist construction
cost value, a path can be generated to navigate the mobile robot navigation and positioning.
robot. Hwang et al. [80] have created a roadmap to develop a
Rackliffe et al. [79] have developed a GIS-based approach prototype of interoperable framework to facilitate BIM and
with which Unmanned Aerial Vehicle (UAV) can be landed GIS integration. In this direction, they employ IFC format of
and Unmanned Ground Vehicle (UGV) can be navigated. a BIM model to integrate it into GIS environment.
In this method, they integrate GIS with sensor data of the Liu et al. [81] explore BIM as a technology facilitating
vehicle in urban settings. collaborations between construction stakeholders and man-
The literature review conducted on robot navigation using agement of building components data. They bring IFC into
GIS technology indicated that algorithms and the implemen- consideration as interoperable data format of BIM contain-
tation play a key role in this regard. ing spatial information of building components. They also
perceive GIS as a platform to provide further spatial analy-
(III) The majority of the research on robot navigation sis on the information provided by IFC. The authors focus
with Geo-referenced locations focuses on algo- on identifying the requirements for “a generic 3D indoor
rithms and computational issues. However, there framework” and identify four of them to be relevant for
are some uncertainties associated with a given indoor spatial analysis, namely [81]: “Generation of a vec-
space that should be considered. Construction sites, tor map,” “Management of data,” “Analysis of environment,”
for instance, are dynamic and are associated with and “Management of safety”.
many uncertainties. Dealing with uncertainties, Irizary and Karan [82] employ GIS to conduct spatial
is one issue that cannot be addressed by predeter- data analysis to find the best location and number of cranes
mined algorithms so that other approaches should on construction sites. To do this, they need semantic infor-
be included. mation with regard to the building elements, and they find
(IV) Zaki and Dunnigan [73] argue that algorithms BIM as a response to this need. To overcome the challenges
applicable to path planning are different from the of the integration of BIM and GIS, they combine an “opti-
ones for motion planning. They define “path plan- mal algorithm”, GIS, and BIM to create a model, optimizing
ning algorithms” as “seeking the most appropri- the location and the number of tower cranes [82].
ate path to a given point”, and “motion planning Zhu et al. [83] have developed an open-source approach
algorithms” as “robot’s actual movement.” Thus, (OSA) to integrate BIM and GIS using IFC and shapefile
they classify motion-planning algorithms into eight format respectively. They utilize IFC-Tree as the spatial
categories namely, (1) “Bug Algorithms,” “Road- structure of IFC to export data into shapefile format through
map,” (2) “Cell Decomposition,” (3) “Potential developing and implementing Automatic Multipatch Gen-
Fields,” (4) “Sampling-based motion planning,” eration (AMG) algorithm. Their work needs to be improved
(5) “Kalman filtering,” (6) “Heuristic Approaches” in terms of efficiency so their next contribution is built upon.
and, (7) “Mathematical programming.” The com- Zhu et al. [84] introduce an enhanced open-source
mon characteristics of all the above-mentioned approach (E-OSA) to integrate geometric data derived
13
from IFC into shapefile in order to contribute to BIM Based on the scientific works mentioned in this section,
and GIS integration. The authors improve the efficiency the following conclusions can be drawn:
of their previous contribution which is open-source
approach (OSA). In this enhanced approach, Brep, swept (V) BIM and GIS integration occurs at different lev-
solid, mapped representation and clipping are successfully els. Researchers defined several levels of integra-
transformed into Brep within a shapefile format using an tion with regard to BIM and GIS integration so
algorithm. It is also discussed that CityGML and Shapefile a common definition with consensus on it is not
as the most prominent data exchange formats with their available. BIM and GIS could integrate mainly
pros and cons [84]. on two levels, which interrelate fundamental level
Wang et al. [12] consider BIM as the digital representa- and application level [89]. Fundamental level refers
tion of a shared database of construction projects to enable to data exchange and interoperability of BIM and
construction practitioners collaborate throughout the project GIS, while application level refers to developing
lifecycle. The authors take GIS into account as geographi- new software tools to benefit from BIM and GIS
cal, cartographical, and remote sensing technology, which advantages. Another classification comprises 5 cat-
comprises spatial data and classify key applications of BIM egories namely, “schema-based,” “service-based,”
and GIS integration into (1) integration of data, (2) pro- “ontology-based,” “processes-based,” and “sys-
jects’ lifecycle applications (3) management of energy, and tem-based” [90]. A third classification comprises
(4) management of urban environments. Additionally, data of three levels namely, data, process, and applica-
integration is identified as the fundamental and the most tion [91]. Data level incorporates extending cur-
challenging step in this regard. rent data schemas or modifying data formats to fit
Hong et al. [85] have studied the correlation of IFC and other software. Process level refers to cooperation
CityGML as the most prominent data format with regard of data schemas, while at the application level, new
to BIM and GIS respectively. They identify features of the software is developed to incorporate BIM and GIS
two, prior to mapping the IFC to CityGML at various level privileges. Although the aforementioned classifica-
of details (LoD), from LoD0 to LoD4. The authors con- tions define different levels of integration, much
sider their contribution as the foundation of BIM and GIS of the research attempts are being carried out on
of indoor and outdoor environment [85]. data level. In this direction, Zhu et al. [87] have
Adouane et al. [86] have developed a model-based extended the data level into two sub-levels, which
approach to facilitate IFC data conversion into CityGML. are geometry level and semantic level. The for-
They encounter semantic and geometry as the main chal- mer focuses on geometry transformation of data,
lenges in this regards. In this direction, they have also devel- whereas the latter concentrates on full attribute data
oped a series of additive algorithms to overcome the issues translation.
occurred in the project. Their work indicates that the seman- (VI) Many researchers have identified various data
tical and geometrical issues occurring when converting IFC exchange formats for both BIM and GIS. The for-
into CityGML, could be handled by a set of algorithms [86]. mer comprises less formats in terms of quantity
Zhu et al. [87] assess integration of BIM and GIS at data compared to the latter. There is a consensus among
level. They conduct literature review on scholarly papers to the researchers, however, that IFC is the promising
investigate data models in terms of relevance and features, data schema representing BIM [12, 81–87]. Build-
examine other potential data models for BIM and GIS inte- ingSMART (formerly the International Alliance for
gration, and provide roadmap for future works. BIM and GIS Interoperability) developed IFC as an EXPRESS-
are considered as well-developed technologies where BIM is based tool [92]. IFC uses three types of geometri-
employed throughout a building lifecycle, while GIS mostly cal definitions to represent 3D models: boundary-
correlates with location issues and spatial data analysis in representation (b-rep), constructive solid geometry
various domains. They have identified the challenges and the (CSG), and sweep volumes [93]. B-rep uses the
methods to integrate BIM and GIS. object’s boundary surfaces to represent a 3D com-
Isikdag et al. [88] have developed BO-IDM based on plex object [94], CSG applies a set of Boolean
building information for indoor navigation purposes. They operators namely, union, intersection, and differ-
determine the requirements for BIM and GIS integration and ence on primitive shapes such as spheres, cones,
they attain this goal through simplifying BIM models [88]. pyramids, or cylinders [95], and sweep volumes
Even though the proposed framework is practical, it shows uses a path to extrude 2D objects in order to create
important limitations such as removing a void for the sake solid shapes [87]. American Institute of Architects
of simplicity, thereby making it insufficient for the purposes (AIA) defined IFC Levels of Development (LOD)
of automation in construction. from lowest to highest amount of information they
13
contain. The five levels are LOD100, LOD200, (VIII) CityGML is defined based on the Levels of Detail
LOD300, LOD400, and LOD500. The BIMForum (LoD) provided in a 3D model from LoD0 to LoD4
have developed LOD350 in addition to the afore- [87]. CityGML also uses boundary representation
mentioned levels as there was a need for a Level (b-rep) to visualize 3D models and it allows users
of development between LOD300 and LOD400 to extend it through application domain extension
in order to detect/avoid clashes, layout, etc. [87]. (ADE).
BuildingSMART have also developed other IFC
schemas such as XML-based IFC standard and Figure 13 illustrates various levels of detail on a residential
ifcXML in addition to EXPRESS-based IFC stand- house. LoD0 is just the footprints of the house in 2 dimen-
ard [96] which can be used for BIM–GIS integra- sional environments, while LoD1 represents in solid shapes
tion. with a flat roof. LoD2 becomes more advanced in terms
(VII) Contrary to the case with BIM, researchers have of showing details compared to LoD1. LoD3 and LoD4,
not reached a consensus regarding GIS data both, demonstrate the openings of the building but LoD4
exchange format. City Geographic Markup Lan- incorporates interior spaces and components such as interior
guage (CityGML) and Shapefile are two primary walls, and doors.
formats in terms of data exchange schema in GIS.
The Environmental System Research Institute
(ESRI) has developed Shapefile as an open data 5 Discussion
schema containing attributes and spatial features
[97]. On the other hand, CityGML is an XML- The current study is a systematic literature review (SLR)
based standard. The Open Geospatial Consor- combining scientometric analysis and qualitative analysis.
tium (OGC) has approved it as the standard open The former is used to investigate a large dataset of articles on
data schema representing 3D models of cities and BIM and GIS integration, which is difficult to conduct with
landscapes [19]. Although Shapefile data schema conventional methods, and the latter is utilized to deeply
is the native format of GIS and can be exported explore the field with relation to robot navigation for con-
to non-GIS software tools such as Collaborative struction sites. The literature review methodology adopts
Design Activity (COLLADA), SketchUp, and 3D a systematic approach in order to be able to investigate the
Studio Max [98], CityGML is more suitable for field comprehensively. It extends earlier review works and
BIM and GIS integration. This is because Shapefile examines the domain of BIM–GIS integration from a new—
is a non-semantic data model while CityGML is. automation in construction—perspective in order to address
Moreover, CityGML can provide bidirectional data the existing limitations and to reduce complexity.
transformation for BIM and GIS integration while This SLR testifies that the current solutions relying on
Shapefile only allows transforming data from BIM BIM with developed localization methods show many limi-
to GIS [84]. tations such as lack of vertical navigation (i.e. from one floor
Fig. 13 Levels of details on a residential house (Source: [99])
13
Fig. 14 Requirements of
BIM–GIS integration for robot
navigation
to another floor), inaccuracy, not considering the dynamic opportunities it provides. Moreover, other methods should
nature of construction sites, etc. Therefore, more research be incorporated to detect objects on construction sites, and
needs to be performed in this regard or new approaches react to its dynamic context.
needs to be developed. The current study, which is part of a Figure 14 illustrates the identified requirements for a digi-
larger research project aimed to provide digital framework tal framework for robots’ positioning on construction sites.
for robot navigation on construction sites, investigates the Having identified the requirements, our future research will
BIM–GIS domain to find its potential for improving robot seek to propose a novel approach to construction robots’
navigation. navigation, integrated with BIM and GIS to cover the limi-
GIS, on the other hand, enables researchers to develop tations of previous attempts and to decrease the complexity
methods for robot navigation both for indoor and outdoor substantially.
applications through applying various algorithms. The
reported contributions are associated with high complex-
ity and are unsatisfactory, as they do not consider construc- 6 Conclusion and Future Work
tion sites uncertainties such as constant changes. Additional
complexity comes from the analysis of data obtained from Studies have indicated that construction industry is suffering
robot sensors. In this direction, complexity could be reduced from low productivity, compared to other industries; also,
exponentially through defining navigable surfaces in which scarceness of qualified workforce is foreseen in near future.
building components are excluded. Construction automation is introduced as one possible solu-
BIM and GIS technologies are becoming omnipresent in tion to these challenges. It is comprised of many aspects and
construction projects and provide great benefits to the pro- practices but one of its functionalities is Digital Fabrica-
ject stakeholders. However, due to their intrinsic differences, tion (Dfab). To enable construction robots to accomplish
specifically in terms of focus, the integration of BIM and the assigned tasks perfectly, they need to be precisely posi-
GIS is somewhat challenging and still under investigation. A tioned on the intended place. BIM and GIS have indicated
number of research attempts are carried out to tackle naviga- great potential in this regard. Since BIM and GIS are already
tion issues with either BIM or GIS for indoor environments, being used for other purposes in construction projects, rely-
but they are still incompatible with construction sites’ char- ing on them for robots’ navigation would reduce the com-
acteristics. BIM and GIS integration shows great potentials plexity and the amount of time spent to implement other
to be employed for robot navigation purposes as several methods. However, BIM–GIS integration is challenging due
research studies confirm it. GIS can be utilized to identify to their different intrinsic focus. Hence, the current study
optimal path so that construction robots would be able to adopts a Systematic Literature Review (SLR) to thoroughly
navigate and localize properly. BIM can also be used in this review the research in the domain. In addition, scientometric
regard. BIM can provide a priori obstacle detection to robot analysis is used to investigate 236 articles. To deeper under-
through geometry and semantics of 3D models. Integration stand the challenges of the BIM–GIS integration in respect
of BIM and GIS has the potential to considerably reduce the to robot navigation, qualitative analysis is carried out on the
complexity of conventional navigation methods beside other topics derived from keywords’ co-occurrence method. Based
13
on the qualitative analysis, challenges, gaps, and limitations 10. Froese TM, Rankin J (2009) Strategic roadmaps for construction
of current solutions are investigated and the requirements to innovation: assessing the state of research. J Inf Technol Constr
14:400–411. https://doi.org/10.14288/1.0048556
address limitations are determined. More importantly, this 11. Ma Z, Ren Y (2017) Integrated application of BIM and GIS: an
research aims to propose a novel approach using BIM and overview. Proc Eng 196:1072–1079
GIS integration for construction robots’ navigation. Future 12. Wang H, Pan Y, Luo X (2019) Integration of BIM and GIS in
work can also incorporate non-scholar sources such as texts sustainable built environment: a review and bibliometric analy-
sis. Autom Constr 103:41–52. https://doi.org/10.1016/j.autco
and articles on websites in order to be investigated. n.2019.03.005
13. Doumbouya L, Guan CS, Gao G, Pan Y (2017) Application
Acknowledgements The authors are grateful to Natural Sciences and of BIM technology in design and construction: a case study of
Engineering Research Council of Canada for the financial support pharmaceutical industrial base of amino acid building project.
through its CRD program 543867-2019 as well as the industrial part- In: 16th international scientific conference on engineering for
ners of the ETS Industrial Chair on the Integration of Digital Technol- rural development, Latvia University of Agriculture, Faculty of
ogy in Construction. Engineering, Jelgava, Latvia, pp 1495–1502
14. Longley PA, Goodchild MF, Maguire DJ, Rhind DW (2005)
Availability of Data and Materials The authors declare that they made Geographic information systems and science. Wiley, London
sure that all data and materials as well as software application support 15. Li Z, Quan SJ, Yang PP-J (2016) Energy performance simula-
their published claims and comply with field standards. tion for planning a low carbon neighborhood urban district: a
case study in the city of Macau. Habit Int 53:206–214. https://
Code Availability All the software files used in the literature are doi.org/10.1016/j.habitatint.2015.11.010
available. 16. Yamamura S, Fan L, Suzuki Y (2017) Assessment of urban
energy performance through integration of BIM and GIS for
smart city planning. Proc Eng 180:1462–1472. https: //doi.
Compliance with Ethical Standards org/10.1016/j.proeng.2017.04.309
17. Tashakkori H, Rajabifard A, Kalantari M (2015) A new
Conflict of interest The authors declare that they have no conflict of 3D indoor/outdoor spatial model for indoor emergency
interest. response facilitation. Build Environ 89:170–182. https://doi.
org/10.1016/j.buildenv.2015.02.036
18. Brundu FG, Patti E, Osello A et al (2017) IoT software infra-
structure for energy management and simulation in smart cit-
ies. IEEE Trans Ind Inf 13:832–840. https://doi.org/10.1109/
References TII.2016.2627479
19. Deng Y, Cheng JCP, Anumba C (2016) A framework for 3D
1. Chen Q, García de Soto B, Adey BT (2018) Construction auto- traffic noise mapping using data from BIM and GIS inte-
mation: research areas, industry concerns and suggestions for gration. Struct Infrastruct Eng 12:1267–1280. https: //doi.
advancement. Autom Constr 94:22–38. https://doi.org/10.1016/j. org/10.1080/15732479.2015.1110603
autcon.2018.05.028 20. Afkhamiaghda M, Mahdaviparsa A, Afsari K, McCuen T (2019)
2. The Scape Group (2016) Sustainability in the supply chain. https Occupants behavior-based design study using BIM–GIS integra-
://www.scapegroup.co.uk/uploads/research/Supply-Chain-Repor tion: an alternative design approach for architects. In: Mutis I,
t_Website.pdf. Accessed 02 Dec 2019 Hartmann T (eds) Advances in informatics and computing in
3. Bock T (2015) The future of construction automation: techno- civil and construction engineering. Springer, Cham, pp 765–772
logical disruption and the upcoming ubiquity of robotics. Autom 21. Amirebrahimi S, Rajabifard A, Mendis P, Ngo T (2016) A
Constr 59:113–121. https://doi.org/10.1016/j.autcon.2015.07.022 framework for a microscale flood damage assessment and visu-
4. García de Soto B, Agustí-Juan I, Hunhevicz J et al (2018) Produc- alization for a building using BIM–GIS integration. Int J Digital
tivity of digital fabrication in construction: cost and time analysis Earth 9:363–386. https://doi.org/10.1080/17538947.2015.10342
of a robotically built wall. Autom Constr 92:297–311. https://doi. 01
org/10.1016/j.autcon.2018.04.004 22. Morris B (2003) The components of the wired spanning forest are
5. Barbosa F, Woetzel J, Mischke J et al (2017) Reinventing con- recurrent. Probab Theory Relat Fields 125:259–265. https://doi.
struction: a route to higher productivity. McKinsey Global org/10.1007/s00440-002-0236-0
Institute 23. Kitchenham B (2004) Procedures for performing systematic
6. Streule T, Miserini N, Bartlomé O et al (2016) Implementation of reviews. Keele UK Keele Univ 33:1–26
scrum in the construction industry. Proc Eng 164:269–276. https 24. Moher D, Shamseer L, Clarke M et al (2015) Preferred report-
://doi.org/10.1016/j.proeng.2016.11.619 ing items for systematic review and meta-analysis protocols
7. Agarwal R, Chandrasekaran S, Sridhar M (2016) Imagining con- (PRISMA-P) 2015 statement. Syst Rev 4:1
struction’s digital future. https://www.mckinsey.com/industries/ 25. van Eck NJ, Waltman L (2014) Visualizing Bibliometric Net-
capital-projects-and-infrastructure/our-insights/imagining-const works. In: Ding Y, Rousseau R, Wolfram D (eds) Measuring
ructions-digital-future#. Accessed 19 Sept 2019 scholarly impact: methods and practice. Springer, Cham, pp
8. Rich BD (2014) Principles of future proofing: a broader under- 285–320
standing of resiliency in the historic built environment. Preserv 26. Cobo MJ, López-Herrera AG, Herrera-Viedma E, Herrera F
Educ Res 7:31–49 (2011) Science mapping software tools: review, analysis, and
9. Bowmaster J, Rankin J (2019) A research roadmap for off-site cooperative study among tools. J Am Soc Inform Sci Technol
construction: automation and robotics. In: Modular and offsite 62:1382–1402. https://doi.org/10.1002/asi.21525
construction (MOC) summit proceedings, pp 173–180. https:// 27. De Nooy W, Mrvar A, Batagelj V, Granovetter M (2005) Explora-
doi.org/10.29173/mocs91 tory social network analysis with Pajek Cambridge University
Press, Cambridge, p 334
13
28. Bastian M, Heymann S, Jacomy M (2009) Gephi: an open relationships. Scientometrics 81:499. https://doi.org/10.1007/
source software for exploring and manipulating networks. In: s11192-008-2173-x
3rd international AAAI conference on weblogs and social media 49. Delbrügger T, Lenz LT, Losch D, Roßmann J (2017) A navigation
29. van Eck NJ, Waltman L (2010) Software survey: VOSviewer, framework for digital twins of factories based on building infor-
a computer program for bibliometric mapping. Scientometrics mation modeling. In: 2017 22nd IEEE international conference
84:523–538 on emerging technologies and factory automation (ETFA). IEEE,
30. Cherven K (2015) Mastering Gephi network visualization. Packt New York, pp 1–4
Publishing Ltd, London 50. Arkin RC (1987) Path planning for a vision-based autonomous
31. Chaomei C (2014) The CiteSpace manual. http://cluster.ischo robot. In: Mobile robots I. International Society for Optics and
ol.drexel.edu/~cchen/citespace/CiteSpaceManual.pdf Photonics, pp 240–250
32. ISARC Proceedings—The international association for automa- 51. Geraerts R, Overmars MH (2007) The corridor map method: a
tion and robotics in construction. https://www.iaarc.org/publi general framework for real-time high-quality path planning. Com-
cations. Accessed 19 Nov 2019 put Anim Virt Worlds 18:107–119
33. Lapierre A, Cote P (2007) Using open web services for urban 52. http://rdf.bg/product-list/ifc-engine/. Accessed 27 Nov 2019
data management: a testbed resulting from an OGC initiative 53. 3D Simulation Software. In: VEROSIM Solutions. https://www.
for offering standard CAD/GIS/BIM services. In: Urban and verosim-solutions.com/en/. Accessed 27 Nov 2019
regional data management. Annual Symposium of the Urban 54. Ibrahim A, Roberts D, Golparvar-Fard M, Bretl T (2017) An inter-
Data Management Society, pp 381–393 active model-driven path planning and data capture system for
34. Su H-N, Lee P-C (2010) Mapping knowledge structure by key- camera-equipped aerial robots on construction sites. Comput Civ
word co-occurrence: a first look at journal papers in Technology Eng 2017:117–124
Foresight. Scientometrics 85:65–79. https://doi.org/10.1007/ 55. Darwish W, Li W, Tang S et al (2019) An RGB-D Data process-
s11192-010-0259-8 ing framework based on environment constraints for mapping
35. Opsahl T, Agneessens F, Skvoretz J (2010) Node centrality in indoor environments. In: Vosselman G, Oude Elberink SJ, Yang
weighted networks: generalizing degree and shortest paths. Soc MY (eds) ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci.
Net 32:245–251. https://doi.org/10.1016/j.socnet.2010.03.006 Copernicus GmbH, London, pp 263–270
36. https://gephi.org/tutorials/gephi-tutorial-layouts.pdf. Accessed 56. Endres F, Hess J, Sturm J et al (2014) 3-D mapping with an
22 Aug 2019 RGB-D camera. IEEE Trans Robot 30:177–187. https://doi.
37. Chen C, Ibekwe-SanJuan F, Hou J (2010) The structure and org/10.1109/TRO.2013.2279412
dynamics of cocitation clusters: a multiple-perspective cocita- 57. Tsai G-J, Chiang K-W, Chu C-H et al (2015) The performance
tion analysis. J Am Soc Inform Sci Technol 61:1386–1409. https analysis of an indoor mobile mapping system with RGB-D Sensor.
://doi.org/10.1002/asi.21309 In: ISPRS—international archives of the photogrammetry, remote
38. Hosseini MR, Martek I, Zavadskas EK et al (2018) Critical sensing and spatial information sciences XL-1/W4:183–188. https
evaluation of off-site construction research: a Scientometric ://doi.org/10.5194/isprsarchives-XL-1-W4-183-2015
analysis. Autom Constr 87:235–247. https://doi.org/10.1016/j. 58. Nahangi M, Heins A, McCabe B, Schoellig A (2018) Automated
autcon.2017.12.002 localization of UAVs in GPS-denied indoor construction envi-
39. Dunning T (1993) Accurate methods for the statistics of surprise ronments using fiducial markers. In: ISARC—Int. Symp. Autom.
and coincidence. Comput Linguist 19:61–74 Robot. in Constr. Int. AEC/FM Hackathon: the future of build.
40. Shibata N, Kajikawa Y, Takeda Y, Matsushima K (2008) Detect- Things. International Association for Automation and Robotics
ing emerging research fronts based on topological measures in Construction I.A.A.R.C)
in citation networks of scientific publications. Technovation 59. Lin WY, Lin PH, Tserng HP (2017) Automating the generation
28:758–775. https://doi.org/10.1016/j.technovation.2008.03.009 of indoor space topology for 3D route planning using BIM and
41. Rousseeuw PJ (1987) Silhouettes: a graphical aid to the inter- 3D-GIS techniques. In: ISARC—Proc. Int. Symp. Autom. Robot.
pretation and validation of cluster analysis. J Comput Appl Constr. International Association for Automation and Robotics in
Math 20:53–65. https://doi.org/10.1016/0377-0427(87)90125-7 Construction I.A.A.R.C), pp 437–444
42. Hicks D (1999) The difficulty of achieving full coverage of 60. Siemiątkowska B, Harasymowicz-Boggio B, Przybylski M et al
international social science literature and the bibliometric con- (2013) BIM based indoor navigation system of Hermes mobile
sequences. Scientometrics 44:193–215. https://doi.org/10.1007/ robot. In: Padois V, Bidaud P, Khatib O (eds) Romansy 19—robot
BF02457380 design, dynamics and control. Springer, Vienna, pp 375–382
43. Guidry JA, Guidry Hollier BN, Johnson L et al (2004) Surveying 61. Hamieh A, Deneux D, Tahon C (2017) BiMov: BIM-based indoor
the cites: a ranking of marketing journals using citation analy- path planning. In: Eynard B, Nigrelli V, Oliveri SM et al (eds)
sis. Mark Educ Rev 14:45–59. https://doi.org/10.1080/10528 Advances on mechanics, design engineering and manufacturing:
008.2004.11488853 proceedings of the international joint conference on mechanics,
44. gephi/gephi. In: GitHub. https : //githu b .com/gephi / gephi . design engineering and advanced manufacturing (JCM 2016),
Accessed 24 Oct 2019 14–16 September, 2016, Catania, Italy. Springer, Cham, pp
45. Khokhar D (2015) Gephi cookbook. Packt Publishing Ltd, 889–899
London 62. Quintana B, Prieto SA, Adán A, Bosché F (2018) Door detec-
46. Ding Y (2011) Scientific collaboration and endorsement: network tion in 3D coloured point clouds of indoor environments. Autom
analysis of coauthorship and citation networks. J Inf 5:187–203. Constr 85:146–166. https://doi.org/10.1016/j.autcon.2017.10.016
https://doi.org/10.1016/j.joi.2010.10.008 63. Kayhani N, Heins A, Zhao WD et al (2019) Improved tag-based
47. Luwel M (2005) The use of input data in the performance analysis indoor localization of UAVs using extended Kalman filter. In: Al-
of R&D systems. In: Moed HF, Glänzel W, Schmoch U (eds) Hussein M (ed) Proc. Int. Symp. Autom. Robot. Constr., ISARC.
Handbook of quantitative science and technology research: the International Association for Automation and Robotics in Con-
use of publication and patent statistics in studies of S&T systems. struction I.A.A.R.C), pp 624–631
Springer, Dordrecht, pp 315–338 64. Neges M, Wolf M, Propach M et al (2017) Improving indoor loca-
48. Lu H, Feng Y (2009) A measure of authors’ centrality in co- tion tracking quality for construction and facility management. In:
authorship networks based on the distribution of collaborative ISARC—Proc. Int. Symp. Autom. Robot. Constr. International
13
Association for Automation and Robotics in Construction 83. Zhu J, Wang X, Wang P et al (2019) Integration of BIM and
I.A.A.R.C), pp 88–95 GIS: Geometry from IFC to shapefile using open-source technol-
65. Palacz W, Ślusarczyk G, Strug B, Grabska E (2019) Indoor robot ogy. Autom Constr 102:105–119. https://doi.org/10.1016/j.autco
navigation using graph models based on BIM/IFC. In: Rutkowski n.2019.02.014
L, Scherer R, Korytkowski M et al (eds) Artificial intelligence and 84. Zhu J, Wang X, Chen M et al (2019) Integration of BIM and
soft computing. Springer, Cham, pp 654–665 GIS: IFC geometry transformation to shapefile using enhanced
66. Kim P, Chen J, Kim J, Cho YK (2018) Slam-driven intelligent open-source approach. Autom Constr 106:102859. https://doi.
autonomous mobile robot navigation for construction applications. org/10.1016/j.autcon.2019.102859
In: Workshop of the European group for intelligent computing in 85. Hong C-H, Hwang J-R, Kang H-Y (2012) A study on the correlation
engineering. Springer, London, pp 254–269 analysis for connection between IFC and CityGML. In: Proceedings
67. Ibrahima M, Moselhib O (2015) IMU-based indoor localization of the 4th ACM SIGSPATIAL international workshop on indoor
for construction applications. In: ISARC. Proceedings of the inter- spatial awareness. ACM, New York, pp 9–12
national symposium on automation and robotics in construction. 86. Adouane K, Stouffs R, Janssen P, Domer B (2019) A model-based
IAARC Publications, London, p 1 approach to convert a building BIM-IFC data set model into Cit-
68. Caldas CH, Torrent DG, Haas CT (2006) Using global position- yGML. J Spat Sci 2019:1–24
ing system to improve materials-locating processes on industrial 87. Zhu J, Wright G, Wang J, Wang X (2018) A critical review of the
projects. J Constr Eng Manag 132:741–749 integration of geographic information system and building informa-
69. Goodrum PM, McLaren MA, Durfee A (2006) The application of tion modelling at the data level. ISPRS Int J Geoinf 7:66
active radio frequency identification technology for tool tracking on 88. Isikdag U, Zlatanova S, Underwood J (2013) A BIM-oriented model
construction job sites. Autom Constr 15:292–302 for supporting indoor navigation requirements. Comput Environ
70. Jang W-S, Skibniewski MJ (2008) A wireless network system for Urban Syst 41:112–123. https://doi.org/10.1016/j.compenvurb
automated tracking of construction materials on project sites. J Civ sys.2013.05.001
Eng Manag 14:11–19 89. Irizarry J, Karan EP, Jalaei F (2013) Integrating BIM and GIS to
71. Taneja S, Akinci B, Garrett JH Jr, Soibelman L (2016) Algorithms improve the visual monitoring of construction supply chain man-
for automated generation of navigation models from building infor- agement. Autom Constr 31:241–254. https://doi.org/10.1016/j.autco
mation models to support indoor map-matching. Autom Constr n.2012.12.005
61:24–41 90. Kang TW, Hong CH (2015) A study on software architecture
72. Mangiameli M, Muscato G, Mussumeci G, Milazzo C (2013) A GIS for effective BIM/GIS-based facility management data integra-
application for UAV flight planning. IFAC Proc Vol 46:147–151. tion. Autom Constr 54:25–38. https://doi.org/10.1016/j.autco
https://doi.org/10.3182/20131120-3-FR-4045.00025 n.2015.03.019
73. Zaki O, Dunnigan M (2017) A navigation strategy for an autono- 91. Amirebrahimi S, Rajabifard A, Mendis P, Ngo T (2016) A BIM–GIS
mous patrol vehicle based on multi-fusion planning algorithms and integration method in support of the assessment and 3D visualisation
multi-paradigm representation schemes. Robot Auton Syst 96:133– of flood damage to a building. J Spat Sci 61:317–350. https://doi.
142. https://doi.org/10.1016/j.robot.2017.07.012 org/10.1080/14498596.2016.1189365
74. Yang Q, Wang M, Kwan M-P, Yang Y (2015) A novel GIS platform 92. Mignard C, Nicolle C (2014) Merging BIM and GIS using ontolo-
for UGV application in the unknown environment. In: 2015 23rd gies application to urban facility management in ACTIVe3D.
international conference on geoinformatics, pp 1–6 Comput Ind 65:1276–1290. https://doi.org/10.1016/j.compi
75. Fernández-Caramés C, Serrano FJ, Moreno V et al (2016) A real- nd.2014.07.008
time indoor localization approach integrated with a geographic 93. Donkers S, Ledoux H, Zhao J, Stoter J (2016) Automatic conversion
information system (GIS). Robot Auton Syst 75:475–489. https:// of IFC datasets to geometrically and semantically correct CityGML
doi.org/10.1016/j.robot.2015.08.005 LOD3 buildings. Trans GIS 20:547–569. https://doi.org/10.1111/
76. Mirats Tur JM, Zinggerling C, Corominas Murtra A (2009) tgis.12162
Geographical information systems for map based navigation in 94. Wu I, Hsieh S (2007) Transformation from IFC data model to GML
urban environments. Robot Auton Syst 57:922–930. https://doi. data model: methodology and tool development. J Chin Inst Eng
org/10.1016/j.robot.2009.06.003 30:1085–1090. https://doi.org/10.1080/02533839.2007.9671335
77. Sun M, Yang S, Liu H (2018) GLANS: GIS based large-scale auton- 95. Wyvill B, Guy A, Galin E (1999) Extending the CSG Tree. Warp-
omous navigation system. In: Tan Y, Shi Y, Tang Q (eds) Advances ing, blending and boolean operations in an implicit surface mod-
in swarm intelligence. Springer, Cham, pp 142–150 eling system. Comput Graph Forum 18:149–158. https://doi.
78. Park W-I, Kim D-J, Lee H-J (2013) Terrain trafficability analysis for org/10.1111/1467-8659.00365
autonomous navigation: a GIS-based approach. Int J Control Autom 96. Deng Y, Cheng JCP, Anumba C (2016) Mapping between BIM
Syst 11:354–361 and 3D GIS in different levels of detail using schema mediation
79. Rackliffe N, Yanco HA, Casper J (2011) Using geographic informa- and instance comparison. Autom Constr 67:1–21. https://doi.
tion systems (GIS) for UAV landings and UGV navigation. In: 2011 org/10.1016/j.autcon.2016.03.006
IEEE conference on technologies for practical robot applications. 97. Environmental Systems Research Institute, Inc (1997) ESRI Shape-
IEEE, London, pp 145–150 file technical description. https://www.esri.com/library/whitepaper
80. Hwang J-R, Hong C-H, Choi H-S (2013) Implementation of pro- s/pdfs/shapefile.pdf. Accessed 26 Nov 2019
totype for interoperability between BIM and GIS: Demonstration 98. Environmental Systems Research Institute, Inc (2008) The Multi-
paper. In: IEEE 7th international conference on research challenges patch geometry type. https://support.esri.com/en/white-paper/1483.
in information science (RCIS). IEEE, London, pp 1–2 Accessed 26 Nov 2019
81. Liu L, Li B, Zlatanova S, Liu H (2018) The path from BIM to A 99. Gröger G, Kolbe TH, Nagel C, Häfele K-H (2012) OGC city geog-
3D indoor framework—a requirement analysis. Int Arch Photo- raphy markup language (CityGML) encoding standard version 2.0;
gramm Remote Sens Spatial Inf Sci XLII–4:373–378. https://doi. OGC Doc; Open Geospatial Consortium: Wayland, MA, USA
org/10.5194/isprs-archives-XLII-4-373-2018
82. Irizary J, Karan E (2012) Optimizing location of tower cranes on Publisher’s Note Springer Nature remains neutral with regard to
construction sites through GIS and BIM integration. Electron J Inf jurisdictional claims in published maps and institutional affiliations.
Technol Constr 17:351–366
13

Int BIMSIGSLR

Uploaded by

Copyright:

Available Formats

Int BIMSIGSLR

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Int BIMSIGSLR

Uploaded by

Copyright:

Available Formats

Archives of Computational Methods in Engineering

Integration of BIM and GIS for Construction Automation, a Systematic

Received: 15 June 2020 / Accepted: 10 January 2021

1 Introduction employing traditional methods, lack of implementing indus-

Fig. 1 Overall Relationship

Fig. 2 Context determination

3 Keywords Co‑occurrence Analysis identified and subsequently a qualitative analysis is con-

Fig. 3 Keywords network of

Fig. 4 Keywords network of

Fig. 5 Keywords network of

4 Findings file exported from VOSviewer. Within Gephi environment,

tigated or received much less attention, which shows the

Fig. 7 Core research focus—

Table 1 Ranking of the subdomains in relation to BIM–GIS integra-

Fig. 9 Graph of prominent

Table 3 Prominent journals of BIM and GIS

1 Automation in construction 71 0.043792 0.772793

Fig. 10 Prominent researchers in BIM and GIS

11. Taiwan is the only country, which does not work in

4.1.4.3 Pioneer Countries Following a procedure similar to

Fig. 11 Pioneer institutions of BIM and GIS

Table 4 Prominent researchers Author Affiliation Number of publica- Publication year

integration to BIM. Indoor localization methods 4.2.2 GIS and Robotics

Fig. 13 Levels of details on a residential house (Source: [99])

You might also like

3 Keywords Co‑occurrence Analysis identified and subsequently a qualitative analysis is con-

4 Findings file exported from VOSviewer. Within Gephi environment,

integration to BIM. Indoor localization methods 4.2.2 GIS and Robotics