Exploring The Fusion Potentials of Data Visualization and Data Analytics in The Process of Mining Digitalization

Received 9 March 2023, accepted 11 April 2023, date of publication 17 April 2023, date of current version 28 April 2023.
Digital Object Identifier 10.1109/ACCESS.2023.3267813
Exploring the Fusion Potentials of Data

Visualization and Data Analytics in the
Process of Mining Digitalization
RUIYU LIANG , CHAORAN HUANG, (Member, IEEE), CHENGGUO ZHANG,
BINGHAO LI , (Senior Member, IEEE), SERKAN SAYDAM, AND ISMET CANBULAT
School of Minerals and Energy Resources Engineering, The University of New South Wales (UNSW), Sydney, NSW 2052, Australia
Corresponding author: Chengguo Zhang (chengguo.zhang@unsw.edu.au)
ABSTRACT Mining digitalisation have been receiving significant attention due to the utilisation of advanced
technologies, such as IoT, automation, and sensing. However, maximising the potential value of collected
data in the mining industry remains a challenge. Therefore, this paper aims to review timely concern topics
to facilitate the fusion implementation in mining engineering. Specifically, this review covers recent popular
topics, such as, data visualisation, data management, data analytics, data fusion, visual analytics, and mining
digital twin construction. In this paper, we aim to draw a comprehensive picture about the fusion of data
visualisation and analytics in the big data context, by examining the recent academic research related to
these topics. Therefore, this paper reviews the visualisation domain by conventional classification, including
scientific visualisation, information visualisation, and visual analytics, associated with the analysis of current
digital twin development. Next, according to the challenges and issues related to visualisation development,
this paper reviews the data management and data analytics domains as well. Incorporating with the fusion
concept, machine learning-oriented fusion applications and potential scenarios in the mining industry have
been discussed. In addition, based on the observation across various domains, this paper presents challenges
and future potentials of data fusion in mining.
INDEX TERMS Mining digitalization, data management, data analytics, data visualization, data fusion.
I. INTRODUCTION challenges in managing, visualising, and analysing raw sen-

The mining industry is currently experiencing an intense sor data and other geological data [3], [7], [8], [9], [10].
digitalisation and transformation in mobile computing, cloud Simultaneously, the integration of multiple datasets and the
storage, data analytics, advanced process control, and the development of digital twin offer new opportunities to reveal
implementation of autonomous mining equipment [1], [2], potential mechanisms of geological hazards, which requires
[3], [4]. Advanced digital technologies such as the Internet large amounts of cleaned data [11], [12]. Effective data man-
of Things (IoT), sensing, and automation have revolutionised agement is critical for various data applications as it can
industries by improving production efficiency and safety. improve efficiency and decision-making accuracy. Therefore,
However, the massive amounts of data generated by these when conducting data management, it is of utmost impor-
technologies have also created unprecedented challenges in tance to clarify data requirements across visualisation, ana-
managing, visualising, and analysing this data [2], [5], [6]. lytics, and information-sharing schemes, including quality,
To date, due to the rapid advances in information and com- quantity, attributes, and other relevant factors [13], [14], [15].
munication technology, the mining industry faces increasing According to recent publications in the mining industry,
visualisation and analytics are generally developed sepa-
The associate editor coordinating the review of this manuscript and rately, resulting in different database, data exchange mech-
approving it for publication was Eunil Park . anism, and data exchange workflow [16], [17], [18], [19].
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

40608 For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/ VOLUME 11, 2023
R. Liang et al.: Exploring the Fusion Potentials of Data Visualization and Data Analytics
From the visualisation perspective, it is generally devel- and meet the increasing social and environmental concerns
oped by utilising 3R techniques, including virtual reality among communities and authorities [2], [4], [42]. Nonethe-
(VR), augmented reality (AR), and mixed reality (MR), and less, it remains increasingly difficult for mining enterprises to
so-called extended reality (XR) [18], [20], [21]. However, decide which digital technologies are most relevant to their
due to the low data utilisation, visualisation-oriented mining needs and individual mines, especially in such a technique
education and training applications are still designed with explosion era [2].
simple interaction and are limited to solid 3D models [17], Table 1 introduces some prevalence technologies adapted
[22], [23], [24], [25], in which the solid model is manually in mining engineering. The IoT/IIoT has led more wireless
generated from complex 2D DXF/DWG drawings. Given protocol data collection, resulting in various raw data has
that, data-driven parametric modelling development receiv- been collected and stored into database. It makes data collec-
ing more attention in recent visualisation development [26], tion, even real-time data collection possible in underground
[27], [28], [29], [30], [31]. environment. Due to the specialised knowledge required for
While artificial intelligence (AI) has been increasingly underground mining processes, 3R applications have been
applied in various industry scenarios, data analytics in the developed to address the need for effective visualisation and
mining industry still faces some intrinsic issues, such as demonstration of mining information [23], [24], [43], [44],
data availability in processing, data quantity in analytics, [45]. In contrast to conventional scientific visualisation, infor-
and data quality in analytics and processing [32], [33], [34]. mation visualisation places greater emphasis on 3D spatial
These issues can also impede the development of data fusion, visualisation. It makes users easier to understand the distri-
making it challenging to achieve [2], [35], [36], [37], [38]. bution of underground mining excavation and exploration.
Given the four Vs of big data, including volume, velocity, In this regards, taxonomy of the data visualisation has two
variety, and veracity, the field of mining engineering faces branches. Scientific visualisation refers to the statistical and
significant challenges in managing and making sense of the scientific data visualisation, which typically involves visu-
vast amounts of data generated by various sources [39]. One alising physical data. Information visualisation is specifi-
solution to address these challenges is to leverage machine cally designed to present 3D and spatial-temporal data in a
learning models, which can help with tasks such as rock status virtual environment, allowing for more intuitive and inter-
monitoring through image detection, semantic model training active exploration of complex datasets [46]. Incorporating
for survey exploration, and multi-dimensional data analyt- the ever-increasing data amounts, data management receives
ics considering spatial-temporal factors. By using machine attention to facilitate the development of data analytics and
learning algorithms, mining companies can better manage visualisation. The challenges of developing an effective data
and analyse big data, improving their decision-making capa- management schema is the variety of mining data, which
bilities and gaining new insights into complex geological includes generic numerical data, drawings, images, complex
phenomena. surveys and report (text), and geological and geotechnical
To that end, this paper aims at reviewing visualisation and structures. As are result, data classification is normally the
analytics-based applications in engineering scenarios. The main task in the very beginning of a data management solu-
outcome can be referred to proceed data utilisation for knowl- tion. However, as the confidential of data across mine sites,
edge discovery and risk management in the mining industry. uniform data is normally hard to realise. The objective of data
In this regard, timely concern topics can be included, such analytics in the mining industry is to discriminate and predict
as, mining digitalisation, data management, interactive data geological hazards. Advanced machine learning models are
visualisation, digital twin, parametric model development, being developed to reveal complex geological mechanisms
data fusion, data integration, etc. such as microseismic activity, rock mass response, and geo-
logical and geotechnical events. However, the success of
these models relies on well-organised data sets, which is often
II. BACKGROUND hindered by data silos across information systems. In the
A. MINING DIGITALISATION realm of integration and information system development,
Over the past decades, the mining industry has faced sev- various timely topics are emerging in the mining industry.
eral challenges. Improving productivity to overcome natural One such topic is data fusion, which has gained considerable
factors such as decreasing ore grades, deeper deposits, and attention due to its potential for complex multi-source data
harder rock mass, combined with increasing environmental analytics when integrated with data management and analyt-
and social awareness, has boosted the industry to constantly ics. The aim of data fusion is to combine various datasets
work to enhance its processes along the whole value chain. and uncover hidden connections, making it crucial for solving
In this context, innovation plays an irreplaceable role by pro- black box issues such as those encountered in underground
viding suitable solutions to surpass these difficulties, ensur- mining activities. The topic of digital twin systems is gain-
ing continuity and sustainability in mining development [38], ing traction in many modern industries, including mining.
[40], [41]. In general, industry innovation is mostly driven It involves the real-time visualisation and interaction with a
by digital technologies, and mining engineering has been physical duplicate. Developing a digital twin system requires
trying to brace the changes to improve efficiency, reduce cost, the integration of multiple state-of-the-art techniques such
VOLUME 11, 2023 40609

TABLE 1. Advanced technologies and its development in the process of mining digitalisation.
as IoT for real-time data streaming, cloud services for data shown in Figure 1, which reveals its iteration from 2D to
exchange and information sharing, and 3D interaction for 4D (time series). This can also be recognised as scientific
bidirectional virtual-physical communication. As the devel- visualisation, information visualisation, and visual analyt-
opment of digital twin systems progresses, data management ics separately [46]. Yet, due to different focuses of vari-
and analytics become increasingly important. This includes ous visualisation schemes, all of them are playing essential
data security and data fusion to ensure that data is managed roles across different sections. For example, 2D visualisation
effectively and insights can be gained from the large amounts works for mining design, 3D visualisation is broadly devel-
of data generated by the system. oped in environment demonstration, and 4D visualisation is
emerging as a critical role in data analytics. In addition, the
B. DATA VISUALIZATION interactive visualisation, involved in 3D and 4D visualisation,
Visualisation is utilised to better understand the mining has been contributing to the development of digital twin and
operations, especially for the underground mining. With the visual analytics. In order to realise automatic update of vir-
implementation of advanced technologies, we expect to get tual environment, data-driven visualisation is receiving more
more insights into data analytics and management by data attention. Hence, Building Information Modelling (BIM) and
visualisation. For example, the patterns of complex geolog- parametric modelling are becoming two typical data-driven
ical and geotechnical structures can be revealed and involved visualisation solutions. They have been widely applied in
into data analytics through the spatial and temporal perspec- civil and building engineering [20], [31], [87], as well as
tives [84], [85]. The revolution of mining data visualisation underground engineering [26], [88], [89], [90]. In a BIM
40610 VOLUME 11, 2023

FIGURE 1. Visualisation revolution routine [86].
dimensions which poses challenge on data analytics and visu-

alisation [97], [98], [99], [100].
However, the information isolation, also known as data
silos, is emerging with the development of data manage-
ment and information system construction. This normally
presents as difficult data exchange and information shar-
ing across different applications, systems, and procedures,
which also hinders the process of data fusion. Given that,
FIGURE 2. Application procedure of big data. the research on data can help handle these issues, including
data management, data standardisation, data exchange, etc.
[19], [49], [74], [92], [98], [101] Yet, regarding recent studies,
the development of data management and the utilisation of
application, the visual model can update automatically with multi-resource data in the mining industry are far behind our
inputs. This unquestionably can be preferred over a conven- ability to collect and store data [102], [103], [104].
tional solid model that has no extensibility. Yet, considering On the other hand, the goal of data analytics is to analyse
the unforeseen of the mine workings’ framework, different the cleaned data to detect hidden patterns, which is achieved
from the standardised building parameters, the implemen- by performing data management procedures on raw data prior
tation of BIM and parametric modelling may need more to the analysis. The discovered data patterns are expected
practice in standardisation, validation, and verification. This to facilitate the applications in identifying hazard and pre-
also poses a further challenge in 3D model development for dicting future events. As an example, the analytics of rock
mining visualisation. Hence, more studies are demanded to mass response data is utilised to discriminate rock burst and
achieve a data-driven visual model development solution, correlated geological activities. In terms of various types of
which could digitalise the solid model into datasets and could data in mining operation, machine learning models receives
be real-time reconstructed with user interaction. attention on mining data analytics. In this regard, mining data
can be numerical, text-oriented surveys/reports, image/video,
drawings. Yet, different from other modern industries, data
C. DATA MANAGEMENT AND DATA ANALYTICS
quality and quantity can be challenging due to continuous
Big data refers to large growing datasets that include het-
increase and its broadly spatial variation.
erogeneous formats: structured, unstructured, and semi-
structured data [91]. Despite the ubiquity of big data in
research entities, there is currently no consensus on what
amount of data can be called big data [92], [93], [94]. D. DATA FUSION, VISUAL ANALYTICS, AND DIGITAL TWIN
To date, due to the development and adaption of advanced In the field of information technology, timely concern top-
technologies, industries are facing challenges when handling ics are including: managing and processing multi-source
the increasing data, such as collecting, integrating, storing, datasets, data fusion, visual analytics, and digital twin, which
and information sharing. covers the topic across data analytics, visualisation, and
According to recent research, the application of big data real-time data processing. While the definition of these enti-
is mainly summarised as three parts: data collection, data ties varies across different fields, there is a consensus that
structuring, and knowledge extraction, as shown in Figure 2. these approaches can help manage the constantly growing
To fine the procedure, data management emerges as a critical amount of data and improve decision-making and analysis
role, which aims to facilitate data analytics and complex capabilities [11], [105]. The ultimate goal is to make data
multidimensional data visualisation [19], [94], [95], [96]. analytics more precise and compatible with evolving techno-
For example, lots of geological surveys have more than two logical needs [16], [84], [106], [107].
VOLUME 11, 2023 40611

TABLE 2. Some examples of various digital twin constructions across different industries.
Yet, conventional data fusion and analytics have lim- The development of digital twin has got more applications
itation in understanding data correlations. For example, in manufacturing. Some studies [117], [118], [119] have
the spatial-temporal connections are not as simple as focused on framework design and the integration of state-
distance-oriented in underground mining. Visualisation and of-the-art techniques, for instance, the blockchain for data
recognition can be a critical part in understanding data dis- security. Some scholars [116] proposed a digital twin frame-
tribution. Hence, visual analytics emerges as an efficient work for production life-cycle management, which consists
tool in enhancing the recognition of data and complex data of three parts, physical space, virtual space, and information-
correlations. Visual analytics is aiming to generate knowl- processing layer. In the context of digital twins, some typical
edge and discover hidden opportunities from massive and frameworks are concluded, all adapted frameworks in recent
complex data, which integrates the human effort to make research are developed from them.
the analytics more reliable [46], [102], [108], [109]. It high- • Digital Twin Reference Architecture: This framework
lights the visualisation and interactive operation in ana- was developed by the Industrial Internet Consortium
lytics, aiming to facilitate data analytics and knowledge (IIC) to provide a common language and understanding
discovery through customise visualisation and interaction. for digital twins. It consists of four layers: the physical
As underground mining is much more like a black-box issue layer, the virtual layer, the data and analytics layer, and
[110], [111], visualisation and custom interaction is an impor- the application layer [120], [121].
tant tool in operation, which can help extract more underlying • Cyber-Physical System (CPS) Reference Architecture:
information to analytics as well. This framework was developed by the National Insti-
To extend visual analytics in a real-time mode, the digital tute of Standards and Technology (NIST) to provide a
twin concept has been prevailing in various industries, which common language and understanding for cyber-physical
is expected to be an integration platform for all available systems, which include digital twins. It consists of five
advanced technologies [14], [112], [113], such as data visual- layers: the physical layer, the communication layer, the
isation, data analytics, data fusion, and even visual analytics. computation layer, the control layer, and the application
The basic idea of the digital twin is to link physical and layer [122], [123], [124].
digital objects in an accurate and real-time manner [114]. • Smart Manufacturing Systems (SMS) Reference Archi-
In a narrow sense, the digital twin is about information tecture: This framework was developed by the National
that fully describes a potential or actual physical production Institute of Standards and Technology (NIST) to sup-
from the micro atomic level to the macro geometrical level port the development of smart manufacturing systems,
[115], [116]. Some industry examples are listed in Table 2. which can include digital twins. It consists of five layers:
40612 VOLUME 11, 2023

to visualise geological data such as seismic data, geolog-

ical models, and mineralogy. This helps geologists and
mining engineers to better understand the subsurface envi-
ronment and make informed decisions about resource extrac-
tion. Information visualisation is used to represent data
in a graphical or pictorial format. In mining, this can be
used to visualise production data, equipment performance
data, and other operational data. Information visualisation
helps mining engineers and operators to quickly identify
trends and anomalies in the data, and make more informed
decisions.
In the development of visualisation, visual model is the
virtual replica of the physical world. The visualisation is
working with all the visual model and conveying information
to users. According to current visualisation solutions, there
are some generic methods in visual replica generation.
• Generating 3D model from 2D drawings (CAD) by
model development software, such as 3Ds Max, blender,
FIGURE 3. A typical five-component framework of digital twin and SolidWorks. The generated 3D model is normally
system [116].
recognised as 3D solid model. This method is often used
to create detailed representations of mining equipment
and infrastructure, allowing engineers and operators to
the physical layer, the connectivity layer, the informa- visualise the components and their relationships in a 3D
tion layer, the functional layer, and the enterprise layer environment. For example, Figure 4 (a) is a VR-based
[125], [126]. interactive undeground mining scenario.
Figure 3 shows one of the well-accepted digital twin frame- • Directly visualising the 2D structures according to the
works, also known as the five components framework (phys- datasets generated from 2D drawings. This method
ical space, virtual space, connection between them, data and involves visualising the 2D drawings and datasets on
service) [40], [116], [117], [122], [127], [128], [129], [130]. a computer screen, without converting them to a 3D
Besides the five components framework, some researchers model. This can be useful for quickly reviewing and
[54] proposed a closed-loop digital twin framework under the analysing the data, without the need for a complex 3D
integration of BIM, IoT, and data mining (DM) techniques. model.
The proposed framework takes advantage of BIM and IoT • Interpreting numerical data and generating visual rep-
techniques to realise the 4D visualisation and task-centred resentation by default 3D models, such as sphere, box,
or worker-centred process model, which is the visual model and cylinder. For example, Figure 4 (b) is a scenario to
to simulate both the tasks execution and worker coopera- dynamic present mining production data. This method
tion [128], [131], [132], [133]. involves using default 3D models to represent numerical
data, such as the volume or concentration of minerals in a
III. DATA VISUALIZATION mining area. These models can help engineers and oper-
This section provides an overview of mining visualisation ators to quickly visualise and analyse the data, without
application by generic classifications, including scientific the need for complex 3D modelling software.
visualisation, information visualisation, and visual analytics. As the complexity of underground mining operation, engi-
Visual analytics is a combination of conventional visuali- neers and operators prefers to visualise 3D models and inter-
sations. Driven by the development of visual analytics, the act with the visual model. Yet, the 3D solid model is an
automatic generation of the visual environment is emerging object-oriented geometry model, which is hard to update over
as a vital branch of the development of data visualisation, time. This can impede the application in long-term operation
for instance, parametric modelling and BIM. In addition, of the mining industry. To address this challenge, some stud-
as the digital twin is a platform integrating visualisation and ies are exploring parametric modelling and information mod-
analytics, this section will also cover applications of mining elling solutions in the development of an information system
digital twin system. construction [90]. It aims to digitalise all general components
into standard parameters, which can be reconstructed into
A. CONVENTIONAL VISUALIZATION 3D model following the program commands. However, while
Scientific and information visualisation are two conventional these solutions can help to create more adaptable models, they
visualisations. Scientific visualisation is used to explore and still face limitations in terms of interactivity and real-time
understand complex scientific data. In mining, it is used updates.
VOLUME 11, 2023 40613

applications in handling complex non-standarded structures

in underground mining scenarios.
According to recent study, some scholar proposed a para-
metric entity and mesh fused modelling solution. The solution
emphasised the spatial parameters in 3D solid model develop-
ment. The proposed multi-factor spatial-temporal data struc-
ture has several advantages, one of which is the potential
to create a parametric underground environment and realise
auto-extension [148]. As for the BIM in mining engineering,
someone [31] proposed a BIM-oriented tunnel design inte-
grated with geological-geotechnical information and ground
conditions. The proposed implementation of the geological
and geotechnical model allowed:
• To manage and update soil data during the entire site
investigation process;
• To share the results of site and laboratory tests in a
common data environment in order to properly calibrate
the geotechnical model for the specific problem, that in
turn re-enters in the BIM flow; and
FIGURE 4. Examples of the use of an immersive VR theatre for mining • To possibly generate 2D stratigraphic profiles, calculate
data visualisation [146]. volumes and finally export results into other interopera-
ble BIM platforms.
In addition, Huang et al. [147] integrated machine learn-
B. PARAMETRIC MODELING AND ing and computer vision techniques, consolidating the tun-
INFORMATION SHARING nelling BIM into a powerful tool for data analysis. They
The objectives of parametric modelling and information also introduced an interactive platform by case study, which
sharing include digitalising 3D visual models into datasets considered the comprehensive and collaborative integration
and binding more data information into visual model. The of GIS, 3D geological modelling, construction methods, and
bound information can help realising various visualisation sensing technologies into the BIM in order to form a reli-
and analysis purposes, such as multi-source visualisation able basis for decisions and management in the lifetime of
comparison and integrating visualisation data with analytics the underground project. Li [90] proposed a 3D information
procedures. To achieve these objectives, CAD systems have modelling technique for linear mine workings. Localised
been improved by recognising the connectivity of shapes Industry Foundation Class (IFC) standard was designed for
through sharing parameters and building links. Currently, the sinking and driving engineering. The proposed information
parametric modelling system is typically classified into three model includes 3D mine geological modelling and mine data
branches [76], [87], [89], [90], [147]. information construction. However, this prototype of mining
BIM needs more concern on the balance of construction
• Solid modelling, which defines the complex shapes
and information. Compared to the application in tunnelling,
or assemblies using key parameters, such as roadway
computer science techniques are required in order to inte-
section, transport composition, etc.;
grate and ensure the prototype could fuse with data ana-
• Assembly modelling, it allows the creation of assemblies
lytics and decision making, rather than another information
of individual objects; and
island.
• Topology-based modelling. It uses mathematical topol-
The BIM applications show the insights into near real-
ogy to define shapes and structures, often allowing for
time modelling, data-driven modelling, and analytics fusion.
more complex and flexible modelling than traditional
Based on this, 3D Simultaneous Localisation and Mapping
solid modelling techniques.
(SLAM) and related point cloud data model development
However, the interaction with visualisation models is chal- are emerging and attracting mining managers [149]. Ren
lenging the parametric modelling, which drives the devel- et al. [55], [56] studied utilising high precise SLAM and UAV
opment of BIM. BIM integrated two forms of 3D solid devices to model underground workings. Its application could
modelling techniques, namely the boundary representation improve underground production safety and contribute to risk
(Brep) and the Constructive Solid Geometry (CSG), to realise management and evaluation. However, given the tremendous
functions of editing, visualising, measuring, clash detection data and high requirements in computation, the applications
as well as other non-editing use [86], [87]. This development are relatively isolated from others, and general issues, such
has strengthen the interaction across visualisation objectives, as data management and structure integration, are broadly
complex datasets, and users. But, it is still challenging BIM existing.
40614 VOLUME 11, 2023

C. MINING DIGITAL TWIN DEVELOPMENT service system, physical entity, virtual model, information
The development of mining digital twin has been focusing on connection, and twin data. The research, to some extent,
fields below: realises digital collaboration through attribute information
and associated information. Additionally, some Scholars have
• Eliminate data silos, such as incorporating real-time
integrated 3R applications into the digital twin construction as
data from various information systems, including under-
well. For example, a VR-based emergency scenario has been
ground mobile equipment, ventilation, and electrical
developed by [145]. It focused on the education and training
systems, to provide a comprehensive view of the mine’s
fields that contributed to the development of an efficient
operation.
fusion of 3R and digital twin in the potential operations in
• Integrate multi-source data. This focus will incorporate
the mining industry.
data from various sources, including geological sur-
As a result, there is a growing focus on the development of
veys, drilling data, and equipment telemetry, to create
digital systems in the mining industry. These systems aim to
a detailed model of the target mine [5].
optimise mining production, facilitate planning and manage-
• Optimise specific procedure. For example, optimising
ment, and enhance the collaboration of advanced technolo-
time-based maintenance policy in the mining indus-
gies. One popular framework that has been adapted is the
try [150], [151], [152], the simulation and decision mak-
five-component framework, which is inherited and evaluated
ing for raise boring method.
to suit specific mining operations. However, given the objec-
• Specific experiment field. This is a research-oriented
tives and involved technologies, the final framework can be
field. It can be a real-time virtual synchronisation of an
varied based on discussion of the framework classification in
experiment sample, for example, a rock sample in rock
Section II-D.
mass properties test [142].
A digital twin-based decision analysis framework for the IV. DATA MANAGEMENT AND ANALYTICS
operation and maintenance of tunnels has been developed to Efficient data management can provide a reasonable data
integrate the life cycle spatio-temporal data of tunnels [153]. structure for data analytics, which can reduce the pre-process
The framework includes a four-layer decision analysis pro- work and improve efficiency. This symbiotic relationship
cess, which consists of the twin data acquisition layer, twin becomes even more important as industries increasingly seek
data fusion layer, model building layer, and service layer. to utilise data effectively. This section will review the data
To facilitate the application of BIM in tunnels, the COBie related fields, such as data management and correlated analyt-
(Construction Operations Building Information Exchange) ics applications. Finally, incorporating data fusion and visual
standard has been extended to tunnel objectives. This exten- analytics, the development of data integration will also be
sion enables effective fusion of spatio-temporal data of the reviewed in this section.
digital twin tunnel. The application of this framework high-
lights the importance of data standardisation and compatibil- A. DATA MANAGEMENT AND INFORMATION SYSTEMS
ity in the development of a digital twin system. Data management is an essential aspect of developing and
In terms of risk estimation, one digital twin system for oil maintaining information systems, particularly in the min-
and gas industry has been proposed using prognostic and ML ing industry [154].The challenges related to data manage-
techniques [152]. It focuses on the integration of sensor data ment in comprehensive mining information systems arise
to analyse hazard likelihood pace of oil pipelines and propose due to several factors, such as the absence of standardisa-
a virtual intelligent integrated automated control system to tion and a robust data structure, the intricate nature of the
predict the risk rate. This case provides reasonable refer- data generated, and the high volume of data [19], [154],
ence in data fusion and ML applications in risk estimation. [155]. These challenges can cause system isolation, which
To optimise and improve the accuracy of decision-making in can hinder the development of fusion analytics in most
the process of raise boring method for underground infras- scenarios.
tructure, some study focuses on the development of a digi- In particular, mining operations use different systems and
tal twin-based decision making paradigm [151]. It proposed processes to collect, store, and analyse data, which makes it
a five-dimension framework that contained physical entity, difficult to integrate data from different sources and ensure
digital representation, service entity, cross-systems entity and data quality and consistency. Moreover, the structure of the
connection entity. Incorporating the five-dimension frame- data generated can be complex and varied, making it chal-
work, four models were developed, including the hybrid mod- lenging to manage and analyse the data, especially when
elling of data-based model, visual geometric models, domain different data types need to be integrated [156]. Additionally,
knowledge-based model and physics-based model. the massive volume of data generated in the mining industry
To facilitate the underground space planning and man- can lead to data redundancy and inconsistencies, which can
agement, some study has developed a underground space affect data quality and accuracy.
visualisation-oriented digital twin system to guarantee the To address these challenges, it is crucial to develop efficient
planning of underground development and planning. It also data management systems, standardise data collection and
referred a five-component digital twin framework, including management, and implement data quality control measures.
VOLUME 11, 2023 40615

FIGURE 6. The workflow and an earth model data structure [16].
FIGURE 5. Data sources in the mining industry.
mining equipment fleet are vital in mining production,

and its maintenance becomes a key issue simultaneously.
These measures can help ensure that the data is accurate, reli-
Teng et al. [150], [159], [160] have proposed device repair
able, and consistent and support effective decision-making in
and maintenance systems merit a highly efficient cloud
the mining industry. The mining data management objectives
database and collaborative data sharing framework. In addi-
have been summarised through a literature review and pre-
tion, Agioutantis [160] also considered the compatibility of
sented in Figure 5.
the designed system, providing access to a different platform/
To handle the ever-increasing data volumes and complex-
system. To some extent, it releases the isolation of infor-
ity, some study focuses on the Hadoop Distributed Files
mation systems in mining digitalisation, and it is the trend
System (HDFS) and Not only SQL (NoSQL) database.
of developing all data management and related information
Regarding the advantages of HDFS, it can be concluded as:
systems.
• Scalability: HDFS is designed to scale out horizontally In order to utilise the time series data in visualisation and
across commodity hardware, allowing for the storage analytics, which is so-called 4D data, some scholar proposed
and processing of massive amounts of data. mining 4D data structure (Figure 6), incorporating with 4D
• Fault-tolerance: HDFS is built to be fault-tolerant, with data management and data analytics [16]. The proposed 4D
data replication and automatic failover to ensure that data is based on a dynamic-static classification of under-
data remains available even in the event of hardware ground mining data, in which the dynamic data can have time
failures. attribute and vary over the mining operation. From the data
• Cost-effective: HDFS can be run on inexpensive com- management perspective, the data classification can improve
modity hardware, making it a cost-effective solution for the processing efficiency in handling the ever-increasing
storing large amounts of data. data.
• High throughput: HDFS is optimised for high- In order to realise data storage, preliminary analysis, data
throughput access to large files, making it well-suited cleaning, and even initial data visualisation, information sys-
for batch processing of big data. tems are developed through the integration of correlated data
Additionally, the NoSQL database has arisen as a solid procedures [156], [158], [161]. Considering various require-
alternative [73], [157]. In contrast to a relational database, ments of the system, different platforms are developed, such
a NoSQL database is one that is less structured/confined in as gas monitoring and dynamic alarming system, air qual-
format, and thus, allows for more flexibility and adaptability. ity pollution monitoring, deformation and risk monitoring,
Nonetheless, due to the non-relational structure of NoSQL etc. [70], [162], [163], [164] To achieve more flexible and
database, the application requires uniform data diagram to security data storage, cloud databases and web services pro-
realise data sharing. vide more possibilities and emerge as an important role
In this regard, Li [158] and Qi [156] studied and con- in information system construction [89], [154]. The most
structed Hadoop-based mining data analytics platforms. They popular and well-accepted framework in information system
concentrated on the production safety of coal mines and construction is summarised as five layers [34], [156], [158]:
challenges from increased monitoring data. In this regard, data layer, collection layer, storage layer, analytics layer, and
they proposed an optimised Hadoop strategy for efficient application layer (Figure 7). As a result, Table 3 shows some
data management. Different from traditional information sys- examples of mining information systems incorporating cloud
tems, the proposed cloud platform merits tremendous data database and web service platform. Currently, cloud service
processing from cloud computing, which provides more mainly contributes to real-time data exchange and statistical
potential possibilities for further data analytics and resource routine data demonstration. Most Cloud computing-based
scheduling. risk analytics and decision-making are limited to data fusion
On the other hand, the development of automation strategy and data processing outcomes [165], [166], [167],
increases the number of mechanical devices. These bulky [168], [169], [170].
40616 VOLUME 11, 2023

TABLE 3. Some examples of mining information systems.
text, speech, and images are increasingly being used to gain

insights into geological and geotechnical activities. To extract
meaning and insights from these unstructured data sources,
techniques such as semantic analytics and correlated machine
learning models are being developed, often with the support
of Natural Language Processing (NLP) [80], [195], [196],
[197], [198].
With the development of data management, data analytics
is getting more opportunities to extract potential connec-
tions among various datasets, which drives the emergence
of data fusion as well. Emerging AI algorithms are devel-
oped in multiple situations, such as rockburst prediction,
seismic data processing, fall of ground hazard discrimination
and prediction, etc. [194], [199], [200], [201], [202], [203],
[204], [205]
This section has also reviewed some general algorithms
in mining data analytics. Various machine learning (ML)
models, such as SVM, k-NN, RF, and XGBoost, have widely
applied in mining event discrimination and prediction. Some
FIGURE 7. Well-accepted information system framework [34], [156], [158].
study shows that it has contributed to the mining hazards
discrimination and classification through a good quality pro-
duction data [206]. In summary, machine learning (ML) can
B. DATA ANALYTICS provide the following advantages in engineering applications
Conventional data analytics in mining engineering can be [198], [207], [208], [209], [210], [211]:
classified into two types: single-factor analytics and multi- • ML can deal with high-dimensional data. As varied
factor analytics. Single-factor analytics are typically per- sensors and integration of information systems generate
formed in isolation, from data collection to analysis. This a huge amount of data of high dimensions, it is difficult
often involves the development of a separate information to understand or find a relationship by traditional data-
system to process the data and facilitate further analy- mining methods.
sis. However, as mining operations become more complex, • ML has the ability to learn and adapt. As an infor-
single-factor analytics may not be sufficient to interpret mation system faces external unpredictable events and
and present geological and geotechnical activities accurately. internal complexity, the predefined modelset by humans
As a result, multi-factor analytics, which takes into account becomes powerless in dealing with uncertain events.
multiple variables simultaneously, are becoming more pop- So ML, which can learn from the data by itself, pro-
ular. In addition, considering traditional data sources, such vides a feasible way to dynamically respond to such
as surveys and reports, unstructured data sources such as challenges.
VOLUME 11, 2023 40617

• ML is able to derive knowledge. While data avail- data and static rock mass data for hazard prediction. Wang
able explodes in information systems, knowledge et al. [217] concluded the workflow of data fusion tech-
for decision-making does not grow synchronously. nology for multisource collaborative monitoring in under-
By learning from big data, knowledge can be derived ground engineering. Data fusion proposed to be processed in
from manufacturing systems that were data-rich but data level and feature level by algorithms such as principal
knowledge-sparse. component analysis (PCA), Wavelet transform, independent
component analysis, etc. After fusion, standardisation and
In addition to the aforementioned cases, data analytics in
normalisation of data would contribute to standardising the
mining can be summarised into three aspects: hazard predic-
data for further model trning. Thirdly, data analytics model,
tion and evaluation, production management, and time-series
for example prediction model, will be trained by KNN, ANN
(4D) data analytics. Compared to the typical 3D concept, 4D
or DL algorithms. Finally, a safety state identification and
data and its analytics play an essential role in the development
prediction model could be finalised after mode analysis.
of information systems as it incorporates the time dimension,
While data fusion can result in a higher accuracy and
which improves the accuracy of analytics and predictions.
reliable outcome, it is challenging the data quality and quan-
For hazard prediction and forecasting, time-based predic-
tity in practice. To facilitate data fusion application, AI-
tion is crucial to understanding, assessing, and acting upon
oriented algorithms and ML models are receiving attention
mining geomechanical risks [78]. Researchers have studied
in raw data processing and further analysis. Based on the
4D data analytics for hazard prediction in the mining industry
traditional random fusion methods such as the weighted
and proposed a system that covers the 4D data management
average method, Kalman filter method, and multi-Bayesian
module and the Weight of Evidence (WoE) algorithm-based
estimation method, multivariate data fusion is the trend
data analytics module [16], [79], [212].
combined with AI algorithms such as fuzzy logic theory,
neural network, rough set theory, and learning technology
C. DATA FUSION [11], [105], [218].
One of the most accepted definitions of data fusion was
provided by the Joint Directors of Laboratories (JDL),
which stands for ‘‘multi-level, multifaceted process han- V. DISCUSSION
dling the automatic detection, association, correlation, esti- A. DATA VISUALIZATION
mation, and combination of data and information from several Visualisation is an essential tool to facilitate the understand-
sources’’ [213]. Leading in data fusion in mining engineering ing and recognition of complex underground procedures. Yet,
is an emerging research area that focuses on integrating and according to recent research, the limitations to the develop-
analysing multiple sources of data to improve the under- ment of underground mining visualisation can be summarised
standing of mining processes and optimise mining operations. below:
According to recent research in the mining industry, data
fusion applications are focusing more on safety concerns and • The flexibility and adaptability to long-term operation
risk assessment. Conventional visualisation is currently relying on the
In terms of the safety concerns in underground coal mines, solid model, which lacks the capability of expansion
some scholars have focused on underground fire detection with long-term operation. When a solid model is used
based on a multi-dimensional data fusion solution [214]. to represent a complex system, such as a mining oper-
They involved various entertainment datasets, such as ation, changes or updates to the system may require
temperature, gas and radon concentration. Incorporating the significant modifications to the solid model. This can be
borehole temperature measurement and gas measurement time-consuming and costly, especially if the model was
methods, they proposed the index of fire zone delineation not designed with future changes in mind. Furthermore,
based on multidimensional data fusion, resulting in a high as a mining operation progresses and evolves over time,
accuracy model for the detection of underground shallow cola the solid model may become outdated and no longer
seam fire ares. As for risk assessments, some study focuses accurately represent the current state of the system. This
on utilising data fusion to evaluate the risk of water inrush can lead to errors and inefficiencies in decision-making
from the coal seam floor [187]. The data covers grouting and planning based on the model.
quantity, the loss of drilling fluid, gamma value, water tem- To address these limitations, researchers and practi-
perature, average water absorption, distance between grout- tioners in visual modelling are exploring alternative
ing loss points, water pressure on coal seam floor, etc. The approaches that allow for more flexibility and adaptabil-
outcome of this fusion showed that the analysis and risk ity, such as using parametric models or generative design
prediction can be consistent with the area of higher risk techniques. These approaches enable the visual model
scores. Rockburst risk assessment research has also integrated to be updated or modified more easily as the system
various experiment data for prediction, such as UCS, geolog- changes over time, which can lead to more accurate and
ical parameters, tunnel shape, etc. [215], [216] McGaughey efficient decision-making and planning.
[16], [80] proposed data fusion analytics with geological • The interaction with visual model
40618 VOLUME 11, 2023

With the development of 3R applications, the interaction specific aspects of the model that are relevant to their
with the visual environment has received more attention needs.
in the underground mining industry. These 3R technolo- Despite its potential benefits, digitalisation of visual
gies provide a more immersive and intuitive way for models is still an area of limited focus. More research
users to interact with the mining environment, which and development is needed to fully realise the potential
can greatly enhance understanding of mining operations of this approach and to develop effective workflows for
and facilitate the design of long-term mining plans. digital twin and data fusion applications.
However, the current interaction design is limited by • The fusion with conventional analytics
the adaption of solid models, which are often static and Multi-modality analytics has gained attention in data
lack the flexibility to adapt to changes in the mining fusion, given its ability to combine data from multi-
environment. As a result, the current state of interaction ple sources. In underground mining, visualisation data
design is not sufficient to fully realise the potential of can provide critical information about geological and
3R technologies. Future research should focus on devel- geotechnical structures, as well as complex rock mass
oping more dynamic and adaptable interaction design features. Incorporating visualisation data into analyt-
solutions that can keep pace with the evolving needs of ics procedures can enhance the accuracy of ML mod-
the mining industry. els. However, this requires effective transformation and
• The association of visual model and routine data digitalisation procedures, which can decompose visual
The visual environment plays a crucial role in data rep- models into datasets for analysis and interaction. Fur-
resentation and interpretation. To facilitate routine data thermore, robust data management and workflow design
presentation, information visualisation systems have are essential for the success of multi-modality analytics
been developed to present various monitoring data by in mining engineering.
spatial distribution. However, these systems are limited • The limitation of visualisation applications
in terms of compatibility with specific datasets, which According to limited visualisation applications, visuali-
limits their applicability in different contexts. Addition- sation is still playing an information demonstration tool
ally, due to the lack of up-to-date visual models, the in mining engineering. While the concept of digital twin
accuracy of routine data visualisation cannot be guaran- and visual analytics have boosted the development of
teed. As a result, there is a need to develop visualisation visualisation platform construction, the objectives are
systems that can accommodate a variety of datasets and still limited to demonstration. There are some potential
incorporate updated visual models to ensure accurate scopes in visualisation application.
and efficient data representation. This would allow for
– Dynamic path finding-oriented underground space
better decision-making, improved safety, and increased
navigation. Due to the complexity of underground
productivity in mining operations.
space, workers and automation equipment are hard
• The transformation between drawings and visual
to make quick decision in an emergency scenarios.
environment
In this regard, a real-time data visualisation plat-
One possible solution for automatic visual model gener-
form, which is adapted with data-driven model gen-
ation is to transform CAD drawings into a visual envi-
eration solution, can facilitate dynamic path finding
ronment. However, this requires efficient data diagrams
visualisation and provide dynamic safety routine
and reasonable data exchange workflows to support the
plan in real-time.
conversion process. Currently, there is limited research
– Interaction-oriented underground mining design.
and case study support for data management that can
Incorporating with the data-driven model genera-
meet the demands of data-driven visual model genera-
tion, visualisation platform can play role in design
tion and update procedures.
validation. In this scenario, users can plan regular
• The digitalisation of visual model
operation in a visualisation system, then the embed-
With the rise of digital twin and data fusion applications,
ded analytics model can facilitate the validation and
there is a growing need for targeted interaction with
proceed feasibility verification, so that to enhance
specific components of visual models. This requirement
efficiency and safety in long-term operation.
can be met through the digitalisation of visual mod-
els, which involves a more detailed decomposition of
the model into datasets that can be used for analytics B. DATA MANAGEMENTN AND ANALYTICS
and interaction purposes.Unlike the transformation of In the big data context, data management can facili-
drawings, which primarily involves converting a visual tate the utilisation across different applications, such as
representation into a digital format, digitalisation goes decision-making, data-driven visualisation, multi-modality
further by breaking down the visual model into dis- data analytics. Yet, according to current mining applications
crete components that can be analysed and manipu- and academic research, the development of mining data man-
lated. This approach enables more targeted and efficient agement is hindered by lacking uniform data diagram. This
interaction with the model, allowing users to focus on causes the low efficiency in raw data processing and cleaning,
VOLUME 11, 2023 40619

as well as the data exchange and data interpretation. Con- capture a wide range of data points, resulting in complex
sidering the data fusion and the development of information and high-dimensional time-series data. The management of
systems, such as digital twin system, the multi-modality data such data requires efficient storage, retrieval, processing, and
becomes the key role in data management [34], [35], [114], analysis techniques to extract valuable insights and support
[126], [210], [211]. In this regard, the following issues should decision-making.
get more efforts in the development of data management and
analytics: 4) MULTI-MODALITY DATA
Multi-modality data refers to the integration of data from
1) DATA VARIETY AND DATA QUANTITY different sources and types, such as text, images, audio, and
This situation poses significant challenges for data manage- numerical data. In the context of mining engineering, multi-
ment in the mining industry. The diverse classification of modality data can include geological and geotechnical data,
data types and properties leads to difficulties in integrating sensor data, visual data from cameras and drones, and digi-
and storing data, resulting in data silos and data fragmenta- talised visualisation data. The challenge with multi-modality
tion. This hinders the ability to effectively analyse and make data is to effectively integrate and process the data in
informed decisions from the available data. Additionally, the a meaningful way that can provide valuable insights for
increasing volume of data generated by new technologies decision-making. This requires advanced data management
can exceed the capacity of traditional data storage and man- and analysis techniques, such as data fusion, machine learn-
agement systems, leading to issues with data processing and ing, and NLP.
retrieval. Therefore, efficient and effective data management
solutions are crucial for successful and sustainable mining
operations. 5) DATA UNIFICATION
The challenges of data unification arise due to the various
2) DATA EXCHANGE AND INFORMATION-SHARING classifications and types of data generated in mining opera-
Data exchange and information-sharing are crucial aspects of tions. These different data types may require different storage
data management, as they allow different managers within a and processing methods, which can lead to fragmentation and
mining operation to access and use data for various purposes. difficulty in integrating the data. Additionally, data may be
However, data exchange and information-sharing can also be generated from different sources or systems, which can also
challenging due to issues such as data format incompatibil- lead to inconsistencies and difficulties in unifying the data.
ity, data security concerns, and differing data access rights The lack of standardisation in data formats and structures
among managers. To address these challenges, standardisa- further exacerbates the challenge of data unification. Further-
tion of data formats and protocols can be helpful to ensure more, data quality and completeness can also be an issue,
interoperability between different systems and stakeholders. as the data may be incomplete, inconsistent, or inaccurate.
In addition, clear guidelines for data access rights and sharing The sheer volume of data generated also poses a challenge for
agreements can help ensure that sensitive data is protected data unification, as the process of collecting, cleaning, and
while still enabling information-sharing among relevant par- integrating large amounts of data can be resource-intensive
ties. The utilisation of cloud database can also facilitate data and time-consuming.
exchange and information-sharing by providing a central
location for storing and accessing data from different sources. 6) DATA SECURITY
Finally, the use of advanced data analytics techniques such Data security have received more attention due to the
as machine learning and data fusion can enable more effec- data applications in various sections. In recent research,
tive use of shared data by providing insights and predic- blockchain offers a robust and resilient mechanism for dis-
tions that can inform decision-making across the mining tributing and storing record history over the internet. The
operation. chain structure links data blocks sequentially in chronolog-
ical order and thereby ensures that this distributed ledger
3) TIME SERIES DATA AND ITS EVER-INCREASING FEATURES cannot be tampered with or forged cryptographically [219],
Time-series data refers to a sequence of data points that [220], [221], [222], [223]. With the high requirements of
are measured at regular intervals over time. In mining data security and fidelity, blockchain has become one of
engineering, examples of time-series data include sensor the most prevalent technologies to ensure transparency, trust,
readings from equipment, temperature and humidity mea- and security. There are at least nine potential use cases for
surements in underground tunnels, and geological monitoring blockchain in mining, which include (1) digital identity for
data. The ever-increasing features of time-series data refer assets and people, (2) data integrity, (3) provenance, (4) cra-
to the increasing number of variables or parameters that dle to grave blockchain for assets, (5) workflow automa-
are measured and recorded over time. With the advance- tion in combination with IoT, (6) supply chain optimisation,
ments in technology and the rise of IoT, more and more (7) tokenised mines, (8) workforce health recording, and
sensors are being deployed across mining operations to (9) human resources management.
40620 VOLUME 11, 2023

C. DATA FUSION analytics to these diverse data sources, it is possible to

1) DATA FUSION SCENARIOS gain a more comprehensive understanding of the mining
The objectives of data fusion is integrating reasonable data operation as a whole, and to identify opportunities for
sources and reasoning potential insights and hidden mech- process optimisation and cost reduction.
anism of target event, such as geological activities, rock • Graphy Neural Networks (GNNs). Graph Neural Net-
mass response, etc. Yet, regarding complex scenarios and works (GNNs) are a type of neural network that is
corresponding datasets, data fusion schemes may be different. well-suited for non-Euclidean data because they operate
Through the review, this study concludes typical scenarios of directly on graphs or other non-Euclidean structures.
data fusion in mining engineering: By leveraging GNNs, mining engineers can analyse and
model the relationships between different elements of a
• Geological Exploration and Characterisation: Data
mining system, such as the geological structures, mining
fusion can integrate geological data from various
equipment, and production processes. This can help to
sources, such as drilling, sampling, geophysical surveys,
optimise mining operations, improve safety, and reduce
and geological maps to create a comprehensive geologi-
costs. For example, GNNs can be used to predict equip-
cal model. This model can then be used for various appli-
ment failures, optimise ore processing, and model the
cations, including resource estimation, mine planning,
flow of materials through the mining system.
and risk assessment.
• Mineral Processing: Data fusion can combine data from
different stages of mineral processing, including ore 2) FUSION CHALLENGES
characterisation, comminution, flotation, and dewater- Considering all the potential scenarios and applications in
ing. The integrated data can provide a more detailed mining engineering, some typical challenges are discussed:
understanding of the processing plant’s performance and
• Data quality and heterogeneity: One of the main chal-
improve the overall recovery rate.
lenges in data fusion is dealing with the heterogeneity
• Mine Safety and Risk Management: Data fusion can
of the data sources, including different data formats,
integrate various types of data related to mine safety,
structures, and quality. It is important to ensure that the
including geotechnical data, ventilation data, and envi-
data is of good quality and that any biases or errors in
ronmental monitoring data. This integrated data can help
the data are identified and corrected.
in risk assessment and decision-making related to mine
• Data unification: With the various classifications of data,
safety.
unifying different data sources into a single format can
• Asset Management: Data fusion can integrate data
be difficult, especially when dealing with multi-modal
related to equipment condition, maintenance history,
data. This requires a consistent and systematic approach
and operational data. This integrated data can provide
to data management, including data pre-processing and
insights into equipment performance, predict equipment
feature extraction.
failures, and optimise maintenance schedules.
• System compatibility: When integrating different data
• Environmental Management: Data fusion can inte-
sources, it is crucial to ensure that the different systems
grate various types of environmental data, including
and technologies are compatible with each other. This
air quality, water quality, and noise pollution data.
includes issues related to data storage, data transfer, and
This integrated data can help in environmental impact
data processing.
assessment, compliance monitoring, and remediation
• Semantic interoperability: Ensuring that the data from
planning.
different sources are semantically compatible can be a
In terms of data types in data fusion applications, this study major challenge in data fusion. This requires the use of
summarised potential fields of data fusion applications: common vocabularies, ontologies, and other semantic
• Numerical data analytics. It can be single factor ana- models to ensure that the data can be understood and
lytics and multi-factor integrated analytics. This is the used in a consistent manner.
initial stage of data fusion in mining. For example, the • Uncertainty and variability: Dealing with uncertainty
discrimination of seismic activities. and variability in the data can be difficult, especially
• Multiple data sources analytics. This can be a high when integrating data from different sources with dif-
dimension data analysis scenario. It can help improve ferent levels of uncertainty. This requires the use of
accuracy in decision-making, risk assessment, and appropriate statistical methods and modeling techniques
resource optimisation. to account for uncertainty and variability in the data.
• Semantics analytics. As geological surveys and reports • Computational complexity: As the volume of data and
are broadly adapted in mining engineering, the integra- the complexity of models increase, the computational
tion of semantics analytics can help to identify patterns complexity of data fusion algorithms can become a sig-
and relationships between different types of data, such nificant challenge. This requires the development of effi-
as geological and geotechnical data, equipment perfor- cient algorithms and computational methods to handle
mance data, and operational data. By applying semantics the large-scale data and complex models.
VOLUME 11, 2023 40621

• Human factors: The adoption and integration of data has reviewed the timely concern topics in the process of
fusion technologies in the mining industry depend on mining digitalisation, including data visualisation, data man-
various human factors such as the skills and expertise of agement, data analytics, data fusion applications, and digital
the workforce, cultural and organisational barriers, and twin construction as well. In order to fuse more data into
resistance to change. visual environment and facilitate the interaction with visual
model, conventional data visualisation is limited to the flex-
D. FUTURE POTENTIALS ibility and scalability. Hence, the parametric modelling and
In spite of all challenges, data fusion technologies will be BIM become one solid alternative. However, considering the
rapidly adapted in the mining industry, numerous opportuni- unforeseen of the complexity of mining workings and geo-
ties will be led in many fields as well. Several future potentials logical and geotechnical structures, an effective data-driven
are discussed in this section. visual model development solution can be the trend in the
To facilitate the development of high efficiency data man- fusion of visualisation and analytics. While it could pose chal-
agement, a uniform data diagram can be proposed. The data lenges to the data management domain. On the other hand,
diagram can fuse the routine data and visual model data so the lack of uniform data diagram/structure emerges as the
that to realise data-driven visualisation and analytics in one most essential issue of the development of digitalisation. The
integrated platform. Incorporating the uniform data diagram, data management is facing various challenges, including data
complex data exchange workflows will be designed as well. variety and quality, the efficiency of data exchange, the solu-
This will provide a more efficient and accurate way to manage tion to handle data variety over time, visualisation-oriented
and process the complex and ever-increasing data in long- multi-modality fusion, and also data security. Therefore, one
terms operation. main domain of future trend of mining digitalisation is data
With the support of the standardised data diagram, a more unification-oriented data management. Nonetheless, as the
standardised and comprehensive approach of visualisation data fusion emerging as the trend of data analytics and ML
can be realised. It can be a data-driven mode for all geo- model development, the fusion scenarios are discussed in the
logical and geotechincal structures, mining development, and mining industry. It can be adapted in various fields, such as
interoperation data visualisation. Benefiting from the data geological exploration and characterisation, mineral process-
exchange workflow design, the interaction can be realised ing, mine safety and risk management, asset management,
across user-visual space and visual-physical design. It can and environmental management. In this regards, this paper
provide a more accurate and real-time demonstration of the concluded the challenges as well. Incorporating with the data
mining operation, which will improve the efficiency and and visualisation domains, the challenges are including data
accuracy of decision-making and visualisation. unification, the solution of data-driven visual development,
Due to the low quality and quantity of collected interoperation-based data fusion design, and the pipeline sys-
data in the very beginning, data fusion can be diffi- tem development. Finally, the overview and future potentials
cult to apply and present a reasonable outcome. In this in the mining industry are proposed according to the fusion
regard, the interpolation-based data fusion can facilitate the of visualisation and analytics.
decision-making and ML model training. It can help to
REFERENCES
address some of the challenges associated with data integra- [1] S. Saydam, B. Hebblewhite, M. Karmis, M. Hitch, F. Cawood,
tion and compatibility. By using advanced interpolation tech- K. de Jager, D. Drinkwater, L. Eagle, S. Joyce, B. Klein, P. Knights,
niques, such as kriging, data sets can be effectively integrated D. Laurence, B. G. Lottermoser, N. M. J. Palarski, K. Spitz, I. Thompson,
J. Trudinger, T. Vargas, B. Watzman, and H. Wotruba, ‘‘Mines
and analysed to provide more accurate insights into mining of the future,’’ Soc. Mining Professors, Morgantown, WV, USA,
operations. Tech. Rep. Version 1.0 (Major Findings), 2019.
Finally, a visual analytics pipeline system can be designed [2] L. Barnewold and B. G. Lottermoser, ‘‘Identification of digital technolo-
gies and digitalisation trends in the mining industry,’’ Int. J. Mining Sci.
and developed aiming to realise interoperation and real-time Technol., vol. 30, no. 6, pp. 747–757, Nov. 2020.
visual-physical communication, which can also be a digital [3] J. Duarte, M. F. Rodrigues, and J. S. Baptista, ‘‘Data digitalisation in the
twin prototype of mining digital twin system. It can provide open-pit mining: Preliminary results,’’ in Occupational and Environmen-
tal Safety and Health II, vol. 277. Cham, Switzerland: Springer, 2020,
a powerful tool for mining engineers to better understand and pp. 715–723.
manage complex mining operations. By integrating real-time [4] J. Duarte, M. F. Rodrigues, and J. S. Baptista, ‘‘Data digitalisation in the
data and analytics into a digital twin prototype, engineers can open-pit mining industry: A scoping review,’’ Arch. Comput. Methods
Eng., vol. 28, no. 4, pp. 3167–3181, Jun. 2021.
simulate and analyse different scenarios to optimise mining [5] P. Kalinowski, O. Dlugosz, and P. Kaminski, Digital Twin of the Mining
operations and improve safety. Additionally, these systems Shaft and Hoisting System as an Opportunity to Improve the Manage-
can provide a platform for ongoing monitoring and predictive ment Processes of Shaft Infrastructure Diagnostics (Don’t short) and
Monitoring, Book Section Digital Twin of the Mining Shaft and Hoisting
maintenance of mining equipment and processes. System as an Opportunity to Improve the Management Processes of Shaft
Infrastructure Diagnostics (Don’t Short) and Monitoring. London, U.K.:
VI. CONCLUSION IntechOpen, 2021.
The fusion of advanced technologies is a fast-growing topic [6] S. V. Lukichev, ‘‘Digital past, present, and future of mining industry,’’
Russian Mining Ind., vol. 2021, no. 4, pp. 73–79, 2021.
in modern industries. It is expected to extract more poten- [7] O. Kalenov and S. Kukushkin, ‘‘Digital transformation of mining enter-
tials to provide informed decisions in operation. This paper prises,’’ in Proc. E3S Web Conf., vol. 278, 2021, p. 01015.
40622 VOLUME 11, 2023

[8] G. Flores-Gonzalez, ‘‘Digital transformation in cave mining,’’ in Proc. [29] Z. Hou, E. Hou, Z. Zhao, N. Deng, and R. Jia, ‘‘A new 3D tun-
8th Int. Conf. Exhib. Mass Mining (MassMin). Santiago, Chile: Univ. nel modelling method—Symmetrical modelling,’’ Metal Mine, vol. 5,
Chile, 2020, pp. 10–22. pp. 107–111, Jun. 2009.
[9] Z. Dragicevic and S. Bosnjak, ‘‘Digital transformation in the mining [30] Z. Liang, K. Zhou, and K. Gao, ‘‘Development of virtual reality seri-
enterprise: The empirical study,’’ Mining Metall. Eng. Bor, vols. 1–2, ous game for underground rock-related hazards safety training,’’ IEEE
pp. 73–90, Jan. 2019. Access, vol. 7, pp. 118639–118649, 2019.
[10] Y. Lazarenko, O. Garafonova, V. Marhasova, and N. Tkalenko, ‘‘Digital [31] S. Fabozzi, S. A. Biancardo, R. Veropalumbo, and E. Bilotta, ‘‘I-BIM
transformation in the mining sector: Exploring global technology trends based approach for geotechnical and numerical modelling of a con-
and managerial issues,’’ in Proc. E3S Web Conf., vol. 315, 2021, p. 04006. ventional tunnel excavation,’’ Tunnelling Underground Space Technol.,
[11] J. Bleiholder and F. Naumann, ‘‘Data fusion,’’ ACM Comput. Surv., vol. 108, Feb. 2021, Art. no. 103723.
vol. 41, no. 1, pp. 1–41, Dec. 2009. [32] I. Damnjanovic and K. Reinschmidt, Data Analytics for Engineering and
[12] W. Ding, X. Jing, Z. Yan, and L. T. Yang, ‘‘A survey on data fusion in Construction Project Risk Management. Cham, Switzerland: Springer,
Internet of Things: Towards secure and privacy-preserving fusion,’’ Inf. 2020.
Fusion, vol. 51, pp. 129–144, Nov. 2019. [33] Y. Pan and L. Zhang, ‘‘Roles of artificial intelligence in construction engi-
[13] D. Burnett, J. Thorp, D. Richards, K. Gorkovenko, and D. Murray-Rust, neering and management: A critical review and future trends,’’ Autom.
‘‘Digital twins as a resource for design research,’’ in Proc. 8th ACM Int. Construct., vol. 122, Feb. 2021, Art. no. 103517.
Symp. Pervasive Displays, Jun. 2019, pp. 1–2. [34] V. S. Agneeswaran, Big Data Analytics Beyond Hadoop: Realtime
[14] J.-F. Uhlenkamp, K. Hribernik, S. Wellsandt, and K.-D. Thoben, ‘‘Digital Applications With Storm, Spark, and More Hadoop Alternatives.
twin applications: A first systemization of their dimensions,’’ in Proc. Upper Saddle River, NJ, USA: FT Press, 2014.
IEEE Int. Conf. Eng., Technol. Innov. (ICE/ITMC), Jun. 2019, pp. 1–8. [35] R. Wang, W. Ji, M. Liu, X. Wang, J. Weng, S. Deng, S. Gao, and
[15] A. Rasheed, O. San, and T. Kvamsdal, ‘‘Digital twin: Values, challenges C.-A. Yuan, ‘‘Review on mining data from multiple data sources,’’ Pat-
and enablers,’’ 2019, arXiv:1910.01719. tern Recognit. Lett., vol. 109, pp. 120–128, Jul. 2018.
[16] W. McGaughey, ‘‘4D data management and modelling in the assessment [36] A. Morshedlou, H. Dehghani, and H. Hoseinie, ‘‘A data driven decision
of deep underground mining hazard,’’ in Proc. Int. Conf. Deep High Stress making approach for long-wall mining production enhancement,’’ Mining
Mining, 2014, pp. 93–106. Sci., vol. 26, pp. 7–20, Jun. 2019.
[17] M. Janiszewski, L. Uotinen, M. Szydlowska, H. Munukka, J. Dong, and [37] U. Awan, S. Shamim, Z. Khan, N. U. Zia, S. M. Shariq, and M. N. Khan,
M. Rinne, ‘‘Visualization of 3D rock mass properties in underground ‘‘Big data analytics capability and decision-making: The role of data-
tunnels using extended reality,’’ in Proc. IOP Conf. Earth Environ. Sci., driven insight on circular economy performance,’’ Technol. Forecasting
vol. 703. Bristol, U.K.: IOP Publishing, 2021, Art. no. 012046. Social Change, vol. 168, Jul. 2021, Art. no. 120766.
[18] J. Tibbett, F. T. Suorineni, and B. Hebblewhite, ‘‘The application of virtual [38] F. Sánchez and P. Hartlieb, ‘‘Innovation in the mining industry: Techno-
reality technology and scientific visualisation to the understanding of logical trends and a case study of the challenges of disruptive innovation,’’
block cave mining systems,’’ in Proc. 3rd Australas. Ground Control Mining, Metall. Exp., vol. 37, pp. 1385–1399, Jul. 2020.
Mining Conf., 2014, pp. 195–200. [39] M. Reichstein, G. Camps-Valls, B. Stevens, M. Jung, J. Denzler,
[19] C. Qi, ‘‘Big data management in the mining industry,’’ Int. J. Minerals, N. Carvalhais, and M. Prabhat, ‘‘Deep learning and process understand-
Metall. Mater., vol. 2, no. 27, pp. 131–139, 2020. ing for data-driven Earth system science,’’ Nature, vol. 566, pp. 195–204,
[20] J.-Y. Wong, C.-C. Yip, S.-T. Yong, A. Chan, S.-T. Kok, T.-L. Lau, Feb. 2019.
M. T. Ali, and E. Gouda, ‘‘BIM-VR framework for building information [40] D. Jones, C. Snider, A. Nassehi, J. Yon, and B. Hicks, ‘‘Characterising
modelling in engineering education,’’ Int. J. Interact. Mobile Technol. the digital twin: A systematic literature review,’’ CIRP J. Manuf. Sci.
(iJIM), vol. 14, no. 6, p. 15, Apr. 2020. Technol., vol. 29, pp. 36–52, May 2020.
[21] G. Schroeder, C. Steinmetz, C. E. Pereira, I. Müller, N. Garcia, [41] W. Liang, S. Luo, G. Zhao, and H. Wu, ‘‘Predicting hard rock pillar sta-
D. Espindola, and R. Rodrigues, ‘‘Visualising the digital twin using web bility using GBDT, XGBoost, and LightGBM algorithms,’’ Mathematics,
services and augmented reality,’’ in Proc. IEEE 14th Int. Conf. Ind. vol. 8, no. 5, p. 765, May 2020.
Informat. (INDIN), Jul. 2016, pp. 522–527. [42] D. A. Shibanov, S. L. Ivanov, and P. V. Shishkin, ‘‘Digital technologies
[22] M. Molokac, G. Alexandrova, M. Kobylanska, B. Hlavnova, P. Hroncek, in modeling and design of mining excavators,’’ J. Phys., Conf., vol. 1753,
and D. Tometzova, ‘‘Virtual mine—Educational model for wider soci- no. 1, Feb. 2021, Art. no. 012052.
ety,’’ in Proc. 15th Int. Conf. Emerg. eLearn. Technol. Appl. (ICETA), [43] Y. Ito, M. Takeuchi, S. Mikami, and Y. Kawamura, ‘‘Development and
Oct. 2017, pp. 1–5. validation of virtual reality application in mining education,’’ J. MMIJ,
[23] H. Bertignoll, M. Labrador Ortega, S. Feiel, and P. Moser, vol. 136, no. 5, pp. 33–39, 2020.
‘‘MiReBooks—Mixed reality lehrbücher für das bergbau-studium,’’ [44] L. M. Daling, S. Khodaei, S. Thurner, A. Abdelrazeq, and I. Isenhardt,
BHM Berg-und Hüttenmännische Monatshefte, vol. 164, no. 4, ‘‘A decision matrix for implementing AR, 360◦ and VR experiences
pp. 178–182, Apr. 2019. into mining engineering education,’’ in HCI International 2021—Posters.
[24] L. Daling, C. Kommetter, A. Abdelrazeq, M. Ebner, and M. Ebner, Cham, Switzerland: Springer, pp. 225–232, doi: 10.1007/978-3-030-
‘‘Mixed reality books: Applying augmented and virtual reality in min- 78642-7_30.
ing engineering education,’’ in Augmented Reality in Education. Cham, [45] M. Janiszewski, L. Uotinen, J. Merkel, J. Leveinen, and M. Rinne,
Switzerland: Springer, 2020, pp. 185–195, doi: 10.1007/978-3-030- ‘‘Virtual reality learning environments for rock engineering, geology and
42156-4_10. mining education,’’ in Proc. 54th U.S. Rock Mech./Geomechanics Symp.
[25] D. Kalkofen, S. Mori, T. Ladinig, L. Daling, A. Abdelrazeq, M. Ebner, Richardson, TX, USA: OnePetro, 2020, pp. 1–8.
M. Ortega, S. Feiel, S. Gabl, T. Shepel, J. Tibbett, T. H. Laine, M. Hitch, [46] H. Ltifi, C. Kolski, and M. B. Ayed, ‘‘Survey on visualization and visual
C. Drebenstedt, and P. Moser, ‘‘Tools for teaching mining students in analytics pipeline-based models: Conceptual aspects, comparative studies
virtual reality based on 360◦ video experiences,’’ in Proc. IEEE Conf. and challenges,’’ Comput. Sci. Rev., vol. 36, May 2020, Art. no. 100245.
Virtual Reality 3D User Interfaces Abstr. Workshops (VRW), Mar. 2020, [47] A. Aziz, O. Schelén, and U. Bodin, ‘‘A study on industrial IoT for the
pp. 455–459. mining industry: Synthesized architecture and open research directions,’’
[26] H. Li, W.-Z. Chen, X.-J. Tan, and E.-Y. Chen, ‘‘Digital design and stability IoT, vol. 1, no. 2, pp. 529–550, Dec. 2020.
simulation for large underground powerhouse caverns with parametric [48] F. Chen, P. Deng, J. Wan, D. Zhang, A. V. Vasilakos, and X. Rong, ‘‘Data
model based on BIM-based framework,’’ Tunnelling Underground Space mining for the Internet of Things: Literature review and challenges,’’ Int.
Technol., vol. 123, May 2022, Art. no. 104375. J. Distrib. Sensor Netw., vol. 11, no. 8, 2015, Art. no. 431047.
[27] K. Suzuki Morales and F. Suorineni, ‘‘Using numerical modelling to rep- [49] P. Gackowiec and M. Podobińska-Staniec, ‘‘IoT platforms for the mining
resent parameters affecting cave mining,’’ in Proc. 1st Int. Conf. Under- industry: An overview,’’ Inżynieria Mineralna, vol. 1, no. 1, pp. 267–272,
ground Mining Technol., Crawley, WA, Australia, 2017, pp. 295–307. Apr. 2021.
[28] X. Feng and Y. Wang, ‘‘An intergratedintelligent modelling on roek [50] R. Laskier, ‘‘Modernizing the mining industry with the Internet of
mechanics,’’ J. Northeastern Univ. Natural Sci., vol. 16, no. 1, pp. 1–5, Things,’’ in Internet of Things and Data Analytics Handbook. Hoboken,
1995. NJ, USA: Wiley, 2017, pp. 521–543.
VOLUME 11, 2023 40623

[51] C. Maheswari, E. B. Priyanka, S. Thangavel, S. V. R. Vignesh, and [72] W. P. Rogers, M. G. Nelson, A. Richins, and A. Hodgson, ‘‘Data manage-
C. Poongodi, ‘‘Multiple regression analysis for the prediction of extrac- ment best practices of complex socio-technical systems: A review of U.S.
tion efficiency in mining industry with industrial IoT,’’ Prod. Eng., mining safety and health management,’’ Geo-Resources Environ. Eng.,
vol. 14, no. 4, pp. 457–471, Oct. 2020. vol. 2, pp. 83–88, Jun. 2017.
[52] M. Mardonova and Y. Choi, ‘‘Review of wearable device technology and [73] A. Corbellini, C. Mateos, A. Zunino, D. Godoy, and S. Schiaffino, ‘‘Per-
its applications to the mining industry,’’ Energies, vol. 11, no. 3, p. 547, sisting big-data: The NoSQL landscape,’’ Inf. Syst., vol. 63, pp. 1–23,
Mar. 2018. Jan. 2017.
[53] F. Molaei, E. Rahimi, H. Siavoshi, S. G. Afrouz, and V. Tenorio, ‘‘A [74] R. Saeid Dindarloo and E. Siami-Irdemoosa, ‘‘Data mining in mining
comprehensive review on Internet of Things (IoT) and its implications in engineering: Results of classification and clustering of shovels failures
the mining industry,’’ Amer. J. Eng. Appl. Sci., vol. 13, no. 3, pp. 499–515, data,’’ Int. J. Mining, Reclamation Environ., vol. 31, no. 2, pp. 105–118,
Mar. 2020. 2017.
[54] Y. Pan and L. Zhang, ‘‘A BIM-data mining integrated digital twin frame- [75] A. Young and P. Rogers, ‘‘A review of digital transformation in mining,’’
work for advanced project management,’’ Autom. Construction, vol. 124, Mining, Metall. Explor., vol. 36, no. 4, pp. 683–699, 2019.
Apr. 2021, Art. no. 103564. [76] Z. Deng, L. Wang, W. Liu, Z. Wang, and E. Qiule, ‘‘Mathematical
[55] Z. L. Ren, L. G. Wang, and L. Bi, ‘‘Robust GICP-based 3D LiDAR SLAM modeling and fuzzy approach for disaster analysis on geo-spatial rock
for underground mining environment,’’ Sensors, vol. 19, no. 13, p. 2915, mass in open-pit mining,’’ Comput. Commun., vol. 150, pp. 384–392,
2019. Jan. 2020.
[56] E. Jones, J. Sofonia, C. Canales, S. Hrabar, and F. Kendoul, ‘‘Applications [77] C. Ordonez, Z. Chen, A. Cuzzocrea, and J. Garcia-Garcia, ‘‘An intelligent
for the hovermap autonomous drone system in underground mining oper- visual big data analytics framework for supporting interactive exploration
ations,’’ J. Southern Afr. Inst. Mining Metall., vol. 120, no. 1, pp. 49–56, and visualization of big OLAP cubes,’’ in Proc. 24th Int. Conf. Inf.
2020. Visualisation (IV), 2020, pp. 421–427.
[57] Z. Wu, H. Wang, W. Yu, J. Xi, W. Lei, and T. Tang, ‘‘3D high- [78] W. McGaughey, ‘‘Data-driven geotechnical hazard assessment: Practice
efficiency and high-precision model-driven modelling for power trans- and pitfalls,’’ in Proc. 1st Int. Conf. Mining Geomechanical Risk. Crawley,
mission tower,’’ Int. Arch. Photogramm., Remote Sens. Spatial Inf. Sci., WA, Australia: Australian Centre for Geomechanics, 2019, pp. 219–232.
vol. 44, pp. 421–426, Nov. 2020. [79] W. McGaughey, V. Laflèche, C. Howlett, J. Sydor, D. Campos, J. Pur-
[58] C. Rausch and C. Haas, ‘‘Automated shape and pose updating of building chase, and S. Huynh, ‘‘Automated, real-time geohazard assessment in
information model elements from 3D point clouds,’’ Autom. Construc- deep underground mines,’’ in Proc. 8th Int. Conf. Deep High Stress
tion, vol. 124, Apr. 2021, Art. no. 103561. Mining. Crawley, WA, Australia: Australian Centre for Geomechanics,
[59] F. Noardo, K. Arroyo Ohori, F. Biljecki, T. Krijnen, C. Ellul, L. Harrie, 2017, pp. 521–528.
and J. Stoter, ‘‘Geobim benchmark 2019: Design and initial results,’’ [80] J. McGaughey, ‘‘Artificial intelligence and big data analytics in mining
in International Archives of the Photogrammetry, Remote Sensing and geomechanics,’’ J. Southern Afr. Inst. Mining Metall., vol. 120, no. 1,
Spatial Information Sciences. Göttingen, Germany: Copernicus GmbH, pp. 15–21, 2020.
2019, pp. 1339–1346. [81] G. Aouad, M. Kagioglou, R. Cooper, J. Hinks, and M. Sexton, ‘‘Technol-
[60] Y. Chen and G. Medioni, ‘‘Object modelling by registration of multi- ogy management of IT in construction: A driver or an enabler?’’ Logistics
ple range images,’’ Image Vis. Comput., vol. 10, no. 3, pp. 145–155, Inf. Manage., vol. 12, no. 1–2, pp. 130–137, 1999.
Apr. 1992. [82] Insight Editor, ‘‘Insight report—IoT digital twins and AI in mining,’’
[61] W. Song and S. Dyke, ‘‘Development of a cyber-physical experimental Insight, Sydney, NSW, Australia, Tech. Rep. 1, 2020. [Online]. Available:
platform for real-time dynamic model updating,’’ Mech. Syst. Signal https://au.insight.com/en_AU/content-and-resources/2020/iot-digital-
Process., vol. 37, nos. 1–2, pp. 388–402, May 2013. twins-and-ai-in-mining.html
[62] C. Tommasi, C. Achille, and F. Fassi, ‘‘From point cloud to BIM: A mod- [83] S. Kosenkov, V. Y. Turchaninov, I. S. Korovin, and D. Y. Ivanov, ‘‘Digital
elling challenge in the cultural heritage field,’’ Int. Arch. Photogramm., twin of the oil well, based on data mining technologies,’’ in Proc. 2nd Int.
Remote Sens. Spatial Inf. Sci., vol. 41, pp. 429–436, Jun. 2016. Conf. Modeling, Simulation Optim. Technol. Appl. (MSOTA). Association
[63] Z. Shang and Z. Shen, ‘‘Real-time 3D reconstruction on construction site of Russian Oil and Gas Field Services Providers (Soyuzneftegazservice),
using visual SLAM and UAV,’’ 2017, arXiv:1712.07122. 2018, pp. 233–238, doi: 10.12783/dtcse/msota2018/27534.
[64] M. Buyukdemircioglu and S. Kocaman, ‘‘Reconstruction and efficient [84] V. Buck, F. Stäbler, E. González, and J. Greinert, ‘‘Digital Earth viewer:
visualization of heterogeneous 3D city models,’’ Remote Sens., vol. 12, A 4D visualisation platform for geoscience datasets,’’ in Proc. Workshop
no. 13, p. 2128, Jul. 2020. Visualisation Environ. Sci. (EnvirVis). The Eurographics Association,
[65] M. Kritzler, M. Funk, F. Michahelles, and W. Rohde, ‘‘The vir- 2021, pp. 1–5.
tual twin: Controlling smart factories using a spatially-correct aug- [85] H. Hamledari, S. Sajedi, B. McCabe, and M. Fischer, ‘‘Automation of
mented reality representation,’’ in Proc. 7th Int. Conf. Internet Things, inspection mission planning using 4D BIMs and in support of unmanned
Linz, Austria: Association for Computing Machinery, Oct. 2017, aerial vehicle–based data collection,’’ J. Construction Eng. Manage.,
pp. 1–2. vol. 147, no. 3, Mar. 2021, Art. no. 04020179.
[66] A. Abdelrazeq, L. Daling, R. Suppes, Y. Feldmann, and F. Hees, ‘‘A [86] WSP, ‘‘Building information modelling—BIM,’’ Montréal, QC, Canada,
virtual reality educational tool in the context of mining engineering-the Tech. Rep., 2022. [Online]. Available: https://www.wsp.com/en-GL/
virtual reality mine,’’ in Proc. 13th Int. Technol., Educ. Develop. Conf., services/building-information-modelling-bim
2019, pp. 8067–8073. [87] R. Sacks, C. Eastman, G. Lee, and P. Teicholz, BIM Handbook: A Guide
[67] C. Zhang, Y. Zhao, T. Teng, J. Chen, and L. Zhang, ‘‘Application of mine to Building Information Modeling for Owners, Designers, Engineers,
VR teaching system in the course of mining introduction,’’ J. Higher Contractors, and Facility Managers. Hoboken, NJ, USA: Wiley, 2018.
Educ., vol. 4, pp. 99–101, Jun. 2020. [88] Y. Hai, ‘‘Modeling and application study of open pit ore-rock transport
[68] R. J. Anthony de Belen, H. Nguyen, D. Filonik, D. D. Favero, and system based on BIM,’’ M.S. thesis, School Manag., Xi’an Univ. Archit.
T. Bednarz, ‘‘A systematic review of the current state of collaborative Technol., Xi’an, China, 2015.
mixed reality technologies: 2013–2018,’’ AIMS Electron. Electr. Eng., [89] A. Khalili, ‘‘An XML-based approach for geo-semantic data exchange
vol. 3, no. 2, pp. 181–223, 2019. from BIM to VR applications,’’ Autom. Construct., vol. 121, Jan. 2021,
[69] S. R. Kadir, A. Lilja, N. Gunn, C. Strong, R. T. Hughes, B. J. Bailey, Art. no. 103425.
J. Rae, R. G. Parton, and J. McGhee, ‘‘Nanoscape, a data-driven 3D [90] W. Li, S. Li, Z. Lin, and Q. Li, ‘‘Information modeling of mine working
real-time interactive virtual cell environment,’’ eLife, vol. 10, Jun. 2021, based on BIM technology,’’ Tunnelling Underground Space Technol.,
Art. no. e64047. vol. 115, Sep. 2021, Art. no. 103978.
[70] J. Xie, X. Wang, S. Hao, and Z. Yang, ‘‘Virtual monitoring method for [91] S. Lohr, ‘‘How big data became so big,’’ New York Times, New York,
hydraulic supports based on digital twin theory,’’ Mining Technol. Trans. NY, USA, Tech. Rep., 11:BU3, 2012.
Inst. Mining Metall., vol. 128, no. 2, pp. 77–87, 2019. [92] B. Furht and F. Villanustre, ‘‘Introduction to big data,’’ in Big Data Tech-
[71] M. Aparicio and C. J. Costa, ‘‘Data visualization,’’ Commun. Des. Quart. nologies and Applications. Cham, Switzerland: Springer, 2016, pp. 3–11,
Rev., vol. 3, no. 1, pp. 7–11, 2015. doi: 10.1007/978-3-319-44550-2_1.
40624 VOLUME 11, 2023

[93] A. Oussous, F. Z. Benjelloun, A. A. Lahcen, and S. Belfkih, ‘‘Big data [116] Y. Zheng, S. Yang, and H. Cheng, ‘‘An application framework of digital
technologies: A survey,’’ J. King Saud Univ. Comput. Inf. Sci., vol. 30, twin and its case study,’’ J. Ambient Intell. Hum. Comput., vol. 10, no. 3,
no. 4, pp. 431–448, Oct. 2018. pp. 1141–1153, Jan. 2019.
[94] K. Hsu, ‘‘Big data analysis and optimization and platform components,’’ [117] F. Tao, F. Sui, A. Liu, Q. Qi, M. Zhang, B. Song, Z. Guo, S. C.-Y. Lu, and
J. King Saud Univ. Sci., vol. 34, no. 4, Jun. 2022, Art. no. 101945. A. Nee, ‘‘Digital twin-driven product design framework,’’ Int. J. Prod.
[95] C. Zhao, L. Ren, Z. Zhang, and Z. Meng, ‘‘Master data management for Res., vol. 57, no. 12, pp. 3935–3953, 2019.
manufacturing big data: A method of evaluation for data network,’’ World [118] F. Tao, J. Cheng, Q. Qi, M. Zhang, H. Zhang, and F. Sui, ‘‘Digital twin-
Wide Web, vol. 23, no. 2, pp. 1407–1421, Mar. 2020. driven product design, manufacturing and service with big data,’’ Int. J.
[96] B. Salimi, B. Howe, and D. Suciu, ‘‘Data management for causal algo- Adv. Manuf. Technol., vol. 94, nos. 9–12, pp. 3563–3576, Feb. 2018.
rithmic fairness,’’ 2019, arXiv:1908.07924. [119] F. Tao, Y. Zhang, Y. Cheng, J. Ren, D. Wang, Q. Qi, and P. Li, ‘‘Digital
[97] I. Atif, F. T. Cawood, and M. A. Mahboob, ‘‘Development of interactive twin and blockchain enhanced smart manufacturing service collaboration
dashboards and intelligent data analytics for visual decision-making in and management,’’ J. Manuf. Syst., vol. 62, pp. 903–914, Jan. 2022.
the underground mining environment: The sterkfontein cave case study,’’
[120] E. Negri, L. Fumagalli, and M. Macchi, ‘‘A review of the roles of
in Proc. Int. Bhurban Conf. Appl. Sci. Technol. (IBCAST), Jan. 2021.
digital twin in CPS-based production systems,’’ Proc. Manuf., vol. 11,
[98] A. Soofastaei, Data Analytics Applied to the Mining Industry. pp. 939–948, Jan. 2017.
Boca Raton, FL, USA: CRC Press, 2020.
[121] C. Liu, L. Le Roux, C. Körner, O. Tabaste, F. Lacan, and S. Bigot,
[99] M. Ghasemaghaei, S. Ebrahimi, and K. Hassanein, ‘‘Data analytics com-
‘‘Digital twin-enabled collaborative data management for metal additive
petency for improving firm decision making performance,’’ J. Strategic
manufacturing systems,’’ J. Manuf. Syst., vol. 62, pp. 857–874, Jan. 2022.
Inf. Syst., vol. 27, no. 1, pp. 101–113, Mar. 2018.
[100] W. Brown, L. Zhang, D. K. Sharma, Y. Jin, I. Dabipi, W. Zhu, and [122] A. Saad, S. Faddel, and O. Mohammed, ‘‘IoT-based digital twin for
E. Lawrence, ‘‘The integration of data analytics to assess multi-complex energy cyber-physical systems: Design and implementation,’’ Energies,
environments of research to practices in engineering education,’’ in Proc. vol. 13, no. 18, p. 4762, Sep. 2020.
IEEE Frontiers Educ. Conf. (FIE), Oct. 2018, pp. 1–6. [123] T. Gabor, L. Belzner, M. Kiermeier, M. T. Beck, and A. Neitz,
[101] E. Tylecková and D. Noskievicová, ‘‘The role of big data in industry 4.0 ‘‘A simulation-based architecture for smart cyber-physical systems,’’
in mining industry in Serbia,’’ Syst. Saf., Hum. Tech. Facility Environ., in Proc. IEEE Int. Conf. Autonomic Comput. (ICAC), Jul. 2016,
vol. 2, no. 1, pp. 166–173, Mar. 2020. pp. 374–379.
[102] M. Vuckovic and J. Schmidt, ‘‘Visual analytics approach to compre- [124] J. Lee, M. Azamfar, J. Singh, and S. Siahpour, ‘‘Integration of digital
hensive meteorological time-series analysis,’’ Data, vol. 5, no. 4, p. 94, twin and deep learning in cyber-physical systems: Towards smart man-
Sep. 2020. ufacturing,’’ IET Collaborative Intell. Manuf., vol. 2, no. 1, pp. 34–36,
[103] Z. Gui, Y. Wang, F. Li, S. Tian, D. Peng, and Z. Cui, ‘‘High performance Mar. 2020.
spatiotemporal visual analytics technologies and its applications in big [125] C. Zhang, G. Zhou, H. Li, and Y. Cao, ‘‘Manufacturing blockchain
socioeconomic data analysis,’’ in Spatial Synthesis. Cham, Switzerland: of things for the configuration of a data- and knowledge-driven digi-
Springer, 2020, pp. 221–255, doi: 10.1007/978-3-030-52734-1_15. tal twin manufacturing cell,’’ IEEE Internet Things J., vol. 7, no. 12,
[104] I. Triguero, D. García-Gil, J. Maillo, J. Luengo, S. García, and F. Herrera, pp. 11884–11894, Dec. 2020.
‘‘Transforming big data into smart data: An insight on the use of the [126] C. Mandolla, A. M. Petruzzelli, G. Percoco, and A. Urbinati, ‘‘Build-
k-nearest neighbors algorithm to obtain quality data,’’ WIREs Data Min- ing a digital twin for additive manufacturing through the exploitation
ing Knowl. Discovery, vol. 9, no. 2, pp. 1–24, Mar. 2019. of blockchain: A case analysis of the aircraft industry,’’ Comput. Ind.,
[105] R. E. Kondo, E. D. De Lima, E. De Freitas Rocha Loures, vol. 109, pp. 134–152, Aug. 2019.
E. A. P. Dos Santos, and F. Deschamps, ‘‘Data fusion for industry 4.0: [127] R. T. Wanasinghe, L. Wroblewski, K. B. Petersen, G. R. Gosine,
General concepts and applications,’’ in Proc. 25th Int. Joint Conf. Ind. L. A. James, O. D. Silva, K. I. G. Mann, and J. P. Warrian, ‘‘Digital twin
Eng. Oper. Manag. (IJCIEOM). Cham, Switzerland: Springer, 2020, for the oil and gas industry: Overview, research trends, opportunities, and
pp. 362–373, doi: 10.1007/978-3-030-43616-2_38. challenges,’’ IEEE Access, vol. 8, pp. 104175–104197, 2020.
[106] S. Xiang, ‘‘Research for multi-source data management system of digital [128] Q. Min, Y. Lu, Z. Liu, C. Su, and B. Wang, ‘‘Machine learning based
mine,’’ in Proc. 6th Int. Conf. Appl. Sci., Eng. Technol. Amsterdam, digital twin framework for production optimization in petrochemical
The Netherlands: Atlantis Press, 2016, pp. 237–240. industry,’’ Int. J. Inf. Manage., vol. 49, pp. 502–519, Dec. 2019.
[107] H. M. M. J. van Rensburg, A. G. S. Gous, J. C. Vosloo, and [129] S. Ke, F. Xiang, Z. Zhang, and Y. Zuo, ‘‘A enhanced interaction frame-
M. Van Heerden, ‘‘Improving data management for environmental work based on VR, AR and MR in digital twin,’’ Proc. CIRP, vol. 83,
reporting in the gold mining industry,’’ South Afr. J. Ind. Eng., vol. 30, pp. 753–758, Jan. 2019.
no. 3, pp. 163–173, Nov. 2019.
[130] E. Carlos Garcia, D. M. Prett, and M. Morari, ‘‘Model predictive control:
[108] D. Keim, G. Andrienko, J.-D. Fekete, C. Görg, J. Kohlhammer, and
Theory and practice—A survey,’’ Automatica, vol. 25, no. 3, pp. 335–348,
G. Melançon, Visual Analytics: Definition, Process, and Challenges.
May 1989.
Berlin, Germany: Springer, 2008, pp. 154–175.
[131] P. Raj and C. Surianarayanan, Digital Twin: The Industry Use Cases,
[109] W. Cui, ‘‘Visual analytics: A comprehensive overview,’’ IEEE Access,
Volume 117 of the Digital Twin Paradigm for Smarter Systems and
vol. 7, pp. 81555–81573, 2019.
Environments: The Industry Use Cases. Amsterdam, The Netherlands:
[110] P. Małkowski and Z. Niedbalski, ‘‘A comprehensive geomechanical
Elsevier, 2020, pp. 285–320.
method for the assessment of rockburst hazards in underground mining,’’
Int. J. Mining Sci. Technol., vol. 30, no. 3, pp. 345–355, May 2020. [132] M. R. Enders and N. Hoßbach, ‘‘Dimensions of digital twin
[111] R. Liang, S. Xu, P. Hou, and C. Zhu, ‘‘Research on 3D modeling tech- applications—A literature review,’’ in Proc. 25th Americas Conf.
nology of mining method for underground mining of metallic deposits,’’ Inf. Syst., 2019, pp. 1–10.
China Mining Mag., vol. 28, no. 3, pp. 73–77, 2019. [133] Q. Qi, F. Tao, T. Hu, N. Anwer, A. Liu, Y. Wei, L. Wang, and A. Y. C. Nee,
[112] M. Zborowski, ‘‘Finding meaning, application for the much-discussed ‘‘Enabling technologies and tools for digital twin,’’ J. Manuf. Syst.,
‘digital twin,’’’ J. Petroleum Technol., vol. 70, no. 6, pp. 26–32, Jun. 2018. vol. 58, pp. 3–21, Jan. 2021.
[113] P. A. Kobryn, ‘‘The digital twin concept,’’ in Proc. Frontiers Eng., Rep. [134] F. Tao, H. Zhang, A. Liu, and A. Y. C. Nee, ‘‘Digital twin in
Lead.-Edge Eng. Symp. Washington, DC, USA: National Academies, industry: State-of-the-art,’’ IEEE Trans. Ind. Informat., vol. 15, no. 4,
2020, pp. 64–66. pp. 2405–2415, Apr. 2019.
[114] M. Liu, S. Fang, H. Dong, and C. Xu, ‘‘Review of digital twin about con- [135] G. N. Schroeder, C. Steinmetz, C. E. Pereira, and D. B. Espindola,
cepts, technologies, and industrial applications,’’ J. Manuf. Syst., vol. 58, ‘‘Digital twin data modeling with automationml and a communication
pp. 346–361, Jan. 2021. methodology for data exchange,’’ IFAC-PapersOnLine, vol. 49, no. 30,
[115] M. Grieves and J. Vickers, ‘‘Digital twin: Mitigating unpredictable, pp. 12–17, 2016.
undesirable emergent behavior in complex systems,’’ in Transdisci- [136] S. Rabah, A. Assila, E. Khouri, F. Maier, F. Ababsa, V. Bourny, P. Maier,
plinary Perspectives on Complex Systems: New Findings and Approaches, and F. Mérienne, ‘‘Towards improving the future of manufacturing
F.-J. Kahlen, S. Flumerfelt, and A. Alves, Eds. Cham, Switzerland: through digital twin and augmented reality technologies,’’ Proc. Manuf.,
Springer, 2017, pp. 85–113. vol. 17, pp. 460–467, Jan. 2018.
VOLUME 11, 2023 40625

[137] Q. Qi and F. Tao, ‘‘Digital twin and big data towards smart manufac- [159] X. Teng, H. Quan, J. Qi, Y. Sun, L. Bai, and Z. Wang, ‘‘Research on
turing and Industry 4.0: 360 degree comparison,’’ IEEE Access, vol. 6, data integration and management and control system of mining equipment
pp. 3585–3593, 2018. maintenance,’’ Mining Technol., vol. 21, no. 5, pp. 180–183, Sep. 2021.
[138] M. Grieves, ‘‘Digital twin: Manufacturing excellence through virtual [160] Z. Agioutantis, S. Delmadorou, N. Steiakakis, C. Steiakakis, and
factory replication,’’ White Paper, vol. 1, no. 2014, pp. 1–7, 2014. S. Papaterpos, ‘‘A real-time event-driven database productivity and main-
[139] K. Y. H. Lim, P. Zheng, and C.-H. Chen, ‘‘A state-of-the-art survey of digi- tenance planning tool for continuous surface mining operations,’’ Int. J.
tal twin: Techniques, engineering product lifecycle management and busi- Mining Mineral Eng., vol. 10, nos. 2–4, pp. 177–188, 2019.
ness innovation perspectives,’’ J. Intell. Manuf., vol. 31, pp. 1313–1337, [161] J. Wang, L. Bi, L. Wang, M. Jia, and D. Mao, ‘‘A mining technology col-
Nov. 2019. laboration platform theory and its product development and application
[140] B. D. Cameron, A. Waaler, and M. T. Komulainen, ‘‘Oil and gas digital to support China’s digital mine construction,’’ Appl. Sci., vol. 9, no. 24,
twins after twenty years. How can they be made sustainable, maintainable p. 5373, 2019.
and useful?’’ Tech. Rep., 2018, pp. 1650–3686. [162] H. Zhou, C.-K. Qu, D.-W. Hu, C.-Q. Zhang, M. U. Azhar, Z. Shen, and
[141] Q. Lu, L. Chen, S. Li, and M. Pitt, ‘‘Semi-automatic geometric digital J. Chen, ‘‘In situ monitoring of tunnel deformation evolutions from aux-
twinning for existing buildings based on images and CAD drawings,’’ iliary tunnel in deep mine,’’ Eng. Geol., vol. 221, pp. 10–15, Jun. 2017.
Autom. Construct., vol. 115, Jul. 2020, Art. no. 103183. [163] A. Choiri, M. N. Mohammed, S. Al-Zubaidi, O. I. Al-Sanjary, and
[142] D. Elmo and D. Stead, ‘‘Disrupting rock engineering concepts: Is there E. Yusuf, ‘‘Real time monitoring approach for underground mine air qual-
such a thing as a rock mass digital twin and are machines capable of ity pollution monitoring system based on IoT technology,’’ in Proc. IEEE
learning rock mechanics?’’ in Proc. Int. Symp. Slope Stability Open Pit Int. Conf. Autom. Control Intell. Syst. (I2CACIS), Jun. 2021, pp. 364–368.
Mining Civil Eng. (Slope Stability). Perth, WA, Australia: Australian [164] X.-S. Zhou, J. Yang, and Y.-T. Zhu, ‘‘Image recognition techniques
Centre for Geomechanics, 2020, pp. 565–576. applied to intelligent system for monitor alarm,’’ J.-Shanghai Jiaotong
[143] H. Xiao, L. He, J. Li, C. Zou, and C. Shao, ‘‘Permeability prediction Univ.-Chin. Ed., vol. 36, no. 4, pp. 498–501, 2002.
for porous sandstone using digital twin modeling technology and lattice [165] A. Novak, D. Bennett, and T. Kliestik, ‘‘Product decision-making infor-
Boltzmann method,’’ Int. J. Rock Mech. Mining Sci., vol. 142, Jun. 2021, mation systems, real-time sensor networks, and artificial intelligence-
Art. no. 104695. driven big data analytics in sustainable industry 4.0,’’ Econ., Manage.
[144] M. Bing, G. Shirong, G. Yinan, Z. Jiaxin, and J. Ersong, ‘‘Construction of Financial Markets, vol. 16, no. 2, pp. 62–72, 2021.
digital twin system for intelligent mining in coal mines,’’ J. Mining Sci. [166] M. Hosseini, A. Shahri, K. Phalp, and R. Ali, ‘‘Engineering transparency
Technol., vol. 7, no. 2, pp. 143–153, 2022. requirements: A modelling and analysis framework,’’ Inf. Syst., vol. 74,
[145] T. Kaarlela, S. Pieskä, and T. Pitkäaho, ‘‘Digital twin and virtual reality pp. 3–22, May 2018.
for safety training,’’ in Proc. 11th IEEE Int. Conf. Cognit. Infocommun. [167] F. Graeme Bonham-Carter, Geographic Information Systems for Geo-
(CogInfoCom), Sep. 2020, pp. 000115–000120. scientists: Modelling With GIS. Amsterdam, The Netherlands: Elsevier,
2014.
[146] J. Tibbett, ‘‘A systems approach to understanding block caving geome-
[168] S. Huang, H. Wang, J. Yu, X. Feng, X. Wang, Y. Tang, and H. Lang,
chanics,’’ M.S. thesis, School Mining Eng., UNSW, Sydney, NSW,
‘‘Construction of intelligent mine information system based on BIM and
Australia, 2015.
GIS,’’ Value Engineer, vol. 38, no. 11, pp. 184–186, 2019.
[147] M. Huang, J. Ninic, and Q. Zhang, ‘‘BIM, machine learning and com-
[169] W. Wang, C. Sun, and G. Su, ‘‘Design of coal mine safety intelligent
puter vision techniques in underground construction: Current status and
control information platform based on cloud platform,’’ COAL Eng.,
future perspectives,’’ Tunnelling Underground Space Technol., vol. 108,
vol. 6, pp. 52–56, 2019.
Feb. 2021, Art. no. 103677.
[170] P. Yang, N. Xiong, and J. Ren, ‘‘Data security and privacy protection for
[148] S. Xiong, ‘‘Study on key technology of underground mine production 3D
cloud storage: A survey,’’ IEEE Access, vol. 8, pp. 131723–131740, 2020.
visual management and control system,’’ M.S. thesis, School Resour. Saf.
[171] B. Ziȩtek, A. Banasiewicz, R. Zimroz, J. Szrek, and S. Gola, ‘‘A portable
Eng., Central South Univ., Changsha, China, 2012.
environmental data-monitoring system for air hazard evaluation in deep
[149] S. Saydam, B. Liu, B. Li, W. Zhang, S. K. Singh, and S. Raval, ‘‘A coarse-
underground mines,’’ Energies, vol. 13, no. 23, p. 6331, Nov. 2020.
to-fine approach for rock bolt detection from 3D point clouds,’’ IEEE
[172] L. Muduli, D. P. Mishra, and P. K. Jana, ‘‘Application of wireless sensor
Access, vol. 9, pp. 148873–148883, 2021.
network for environmental monitoring in underground coal mines: A sys-
[150] J. Savolainen and M. Urbani, ‘‘Maintenance optimization for a multi-unit tematic review,’’ J. Netw. Comput. Appl., vol. 106, pp. 48–67, Mar. 2018.
system with digital twin simulation: Example from the mining industry,’’ [173] D. Suresh and K. Yarrakula, ‘‘Subsidence monitoring techniques in coal
J. Intell. Manuf., vol. 32, no. 7, pp. 1953–1973, 2021. mining: Indian scenario,’’ Indian J. Geo-Marine Sci., vol. 47, no. 10,
[151] F. Hu, X. Qiu, G. Jing, J. Tang, and Y. Zhu, ‘‘Digital twin-based decision pp. 1918–1933, 2018.
making paradigm of raise boring method,’’ J. Intell. Manuf., vol. 33-4, [174] R. Lu, F. Ma, J. Guo, and H. Zhao, ‘‘Analysis and monitoring of roadway
2022, doi: 10.1007/s10845-022-01941-0. deformation mechanisms in nickel mine, China,’’ Concurrency Comput.,
[152] E. B. Priyanka, S. Thangavel, X.-Z. Gao, and N. S. Sivakumar, ‘‘Digital Pract. Exp., vol. 31, no. 10, pp. e4832, 2019.
twin for oil pipeline risk estimation using prognostic and machine learn- [175] M. Scaioni, L. Barazzetti, A. Giussani, M. Previtali, F. Roncoroni, and
ing techniques,’’ J. Ind. Inf. Integr., vol. 26, May 2022, Art. no. 100272. M. I. Alba, ‘‘Photogrammetric techniques for monitoring tunnel defor-
[153] G. Yu, Y. Wang, Z. Mao, M. Hu, V. Sugumaran, and Y. K. Wang, ‘‘A mation,’’ Earth Sci. Informat., vol. 7, no. 2, pp. 83–95, 2014.
digital twin-based decision analysis framework for operation and main- [176] Z. Libing, W. Weidong, M. Aoshu, and L. Bing, ‘‘Construction of the
tenance of tunnels,’’ Tunnelling Underground Space Technol., vol. 116, safety risk monitoring of tailings based on big data of information plat-
Apr. 2021, Art. no. 104125. form,’’ Appl. Mech. Mater., vol. 724, pp. 368–372, Sep. 2015.
[154] S. Singh, E. Shehab, N. Higgins, K. Fowler, D. Reynolds, [177] J. Satria, I. Irawan, and N. Setiawan, ‘‘Rock mass classification for design
A. J. Erkoyuncu, and P. Gadd, ‘‘Data management for developing of excavation method and support system of tunnel 1 Sigli–Aceh toll road,
digital twin ontology model,’’ Proc. Inst. Mech. Eng., B, J. Eng. Manuf., Indonesia,’’ in Proc. IOP Conf. Earth Environ. Sci., vol. 871. Bristol,
vol. 235, no. 14, pp. 2323–2337, 2021. U.K.: IOP Publishing, 2021, Art. no. 012055.
[155] G. Manogaran, P. M. Shakeel, S. Baskar, C.-H. Hsu, S. N. Kadry, [178] S. Yang, X. Zhou, and J. Bi, ‘‘Design of heterogeneous data integration
R. Sundarasekar, P. M. Kumar, and B. A. Muthu, ‘‘FDM: Fuzzy- platform of smart mine,’’ Ind. Mine Autom., vol. 41, no. 5, pp. 23–26,
optimized data management technique for improving big data analytics,’’ 2015.
IEEE Trans. Fuzzy Syst., vol. 29, no. 1, pp. 177–185, 2021. [179] M. Rylnikova, D. Radchenko, and D. Klebanov, ‘‘Intelligent mining
[156] H. Qi, ‘‘Research on task scheduling optimization methods of mine safety engineering systems in the structure of industry 4.0,’’ in Proc. 2nd Int.
supervision cloud platform based on Hadoop,’’ M.S. thesis, School Inf. Innov. Mining Symp., vol. 21, 2017, p. 01032.
Control Eng., China Univ. Mining Technol., Xuzhou, China, 2019. [180] Z. Zhou, C. Zhao, and Y. Huang, ‘‘An optimization method for the
[157] V. N. Gudivada, D. Rao, and V. V. Raghavan, ‘‘NoSQL systems for big station layout of a microseismic monitoring system in underground mine
data management,’’ in Proc. IEEE World Congr. Services, Jun. 2014, engineering,’’ Sensors, vol. 22, no. 13, p. 4775, 2022.
pp. 190–197. [181] C. Wang, G. Si, C. Zhang, A. Cao, and I. Canbulat, ‘‘A statistical method
[158] Y. Li, ‘‘Research and design of mine big data analysis platform based on to assess the data integrity and reliability of seismic monitoring systems
Hadoop,’’ M.S. thesis, School Control Comput. Eng., North China Electr. in underground mines,’’ Rock Mech. Rock Eng., vol. 54, pp. 5885–5901,
Power Univ., Beijing, China, 2021. Nov. 2021.
40626 VOLUME 11, 2023

[182] C. Ma, T. Li, and H. Zhang, ‘‘Microseismic and precursor analysis of [201] K. M. Asim, S. S. Moustafa, I. A. Niaz, E. A. Elawadi, T. Iqbal, and
high-stress hazards in tunnels: A case comparison of rockburst and fall of F. Martínez-Álvarez, ‘‘Seismicity analysis and machine learning models
ground,’’ Eng. Geol., vol. 265, Nov. 2020, Art. no. 105435. for short-term low magnitude seismic activity predictions in Cyprus,’’ Soil
[183] Y. Maeda, Y. Yamanaka, T. Ito, and S. Horikawa, ‘‘Machine learning Dyn. Earthq. Eng., vol. 130, Mar. 2020, Art. no. 105932.
based detection of volcano seismicity using the spatial pattern of ampli- [202] F. Liu, ‘‘Rockburst risk discrimination and forecast based on microseis-
tudes,’’ Geophys. J. Int., vol. 593, pp. 416–444, Apr. 2020. mic monitoring,’’ M.S. thesis, School Water Conservancy Hydropower,
[184] A. Mignan, D. Landtwing, P. Kästli, B. Mena, and S. Wiemer, ‘‘Induced Xi’an Univ. Technol., Xi’an, China, 2020.
seismicity risk analysis of the 2006 basel, switzerland, enhanced geother- [203] R. S. Faradonbeh and A. Taheri, ‘‘Long-term prediction of rockburst
mal system project: Influence of uncertainties on risk mitigation,’’ hazard in deep underground openings using three robust data mining
Geothermics, vol. 53, pp. 133–146, Jan. 2015. techniques,’’ Eng. Comput., vol. 35, no. 2, pp. 659–675, 2019.
[185] Y. Hou, X. Jiang, and W. Quan, ‘‘Construction of virtual simulation [204] Y. Chen, ‘‘Automatic microseismic event picking via unsupervised
experiment teaching system for mining and safety,’’ Educ. Modernization, machine learning,’’ Geophys. J. Int., vol. 222, no. 1, pp. 1750–1764, 2020.
vol. 6, no. 46, pp. 40–43, 2019. [205] Y. Zhou, Z. Zhao, C. Liu, X. Jiang, and D. Ma, ‘‘Inversion analysis
[186] J. Xie, Z. Yang, Y. Wang, and X. Wang, ‘‘A remote VR operation system of crustal stress distribution law in gully geomorphic mining area,’’
for a fully mechanised coal-mining face using real-time data and collabo- Geotechnical Geological Eng., vol. 37, no. 5, pp. 4075–4087, 2019.
rative network technology,’’ Mining Technol. Trans. Inst. Mining Metall., [206] H. Al-Sahaf, Y. Bi, Q. Chen, A. Lensen, Y. Mei, Y. Sun, B. Tran, B. Xue,
vol. 127, no. 4, pp. 230–240, 2018. and M. Zhang, ‘‘A survey on evolutionary machine learning,’’ J. Roy. Soc.
[187] Y. Guo, S. Dong, Y. Hao, Z. Liu, T.-C. J. Yeh, W. Wang, Y. Gao, P. Li, New Zealand, vol. 49, no. 2, pp. 205–228, 2019.
and M. Zhang, ‘‘Risk assessments of water inrush from coal seam floor [207] P. Wozniakowska and D. W. Eaton, ‘‘Machine learning-based analy-
during deep mining using a data fusion approach based on grey system sis of geological susceptibility to induced seismicity in the Montney
theory,’’ Complexity, vol. 2020, pp. 1–12, May 2020. Formation, Canada,’’ Geophys. Res. Lett., vol. 47, no. 22, Nov. 2020,
[188] H. Gao, ‘‘Coal mine geology digitization management system develop- Art. no. e2020GL089651.
ment,’’ in Proc. IOP Conf. Earth Environ. Sci., vol. 565. Bristol, U.K.: [208] S. Qu, Z. Guan, E. Verschuur, and Y. Chen, ‘‘Automatic high-resolution
IOP Publishing, 2020, Art. no. 012021. microseismic event detection via supervised machine learning,’’ Geophys.
[189] T. Olsson and U. Franke, ‘‘Risks and assets: A qualitative study of J. Int., vol. 222, no. 3, pp. 1881–1895, 2020.
a software ecosystem in the mining industry,’’ in Proc. 27th ACM [209] P. C. Sen, M. Hajra, and M. Ghosh, Supervised Classification Algorithms
Joint Meeting Eur. Softw. Eng. Conf. Symp. Found. Softw. Eng. New in Machine Learning: A Survey and Review. Singapore: Springer, 2020,
York, NY, USA: Association for Computing Machinery, Aug. 2019, pp. 99–111.
[210] X. Wu, L. Xiao, Y. Sun, J. Zhang, T. Ma, and L. He, ‘‘A survey of human-
pp. 895–904.
in-the-loop for machine learning,’’ Future Gener. Comput. Syst., vol. 135,
[190] S. Blaser, S. Nebiker, and D. Wisler, ‘‘Portable image-based high perfor-
pp. 364–381, May 2022.
mance mobile mapping system in underground environments—System
[211] S. Feng, Z. Chen, H. Luo, S. Wang, Y. Zhao, L. Liu, D. Ling, and L.
configuration and performance evaluation,’’ ISPRS Ann. Photogramm.,
Jing, ‘‘Tunnel boring machines (tbm) performance prediction: A case
Remote Sens. Spatial Inf. Sci., vol. 4, pp. 255–262, May 2019.
study using big data and deep learning,’’ Tunnelling Underground Space
[191] H. Rehman, A. M. Naji, J.-J. Kim, and H. Yoo, ‘‘Extension of tun-
Technol., vol. 110, p. 103636, 2021.
neling quality index and rock mass rating systems for tunnel support
[212] K. Larsen, ‘‘Data exploration with weight of evidence and
design through back calculations in highly stressed jointed rock mass: An
information value in R,’’ MultiThread, San Francisco, CA, USA,
empirical approach based on tunneling data from Himalaya,’’ Tunnelling
Tech. Rep. 1, 2015. [Online]. Available: https://multithreaded.stitchfix.
Underground Space Technol., vol. 85, pp. 29–42, Mar. 2019.
com/blog/2015/08/13/weight-of-evidence/
[192] E. Brzychczy and A. Trzcionkowska, ‘‘Process-oriented approach for
[213] H. Boström, S. F. Andler, M. Brohede, R. Johansson, A. Karlsson,
analysis of sensor data from longwall monitoring system,’’ in Intelligent
J. Van Laere, L. Niklasson, M. Nilsson, A. Persson, and T. Ziemke, ‘‘On
Systems in Production Engineering and Maintenance (Advances in Intel-
the definition of information fusion as a field of research,’’ Digitala Veten-
ligent Systems and Computing), A. Burduk, E. Chlebus, T. Nowakowski,
skapliga Arkivet, Institutionen för Kommunikation och Inf., Skövde,
and A. Tubis, Eds. Cham, Switzerland: Springer, 2019, pp. 611–621.
Sweden, Tech. Rep., 2007.
[193] N. Bar, S. Nicoll, M. Reynolds, and D. Bran, ‘‘Geotechnical data manage- [214] H. Wang, X. Fang, Y. Li, Z. Zheng, and J. Shen, ‘‘Research and appli-
ment and visualization systems: Meeting the data challenge of the 21st cation of the underground fire detection technology based on multi-
century and maximizing value for open pit mines,’’ in Proc. ISRM Eur. dimensional data fusion,’’ Tunnelling Underground Space Technol.,
Rock Mech. Symp. (EUROCK). Richardson, TX, USA: OnePetro, 2018, vol. 109, May 2021, Art. no. 103753.
pp. 1–6. [215] L. R. E. Sousa, T. Miranda, R. L. E. Sousa, and J. Tinoco, ‘‘The use of
[194] X. Feng, H. Zhou, S. Li, Q. Sheng, and Q. Jiang, ‘‘System of intelligent data mining techniques in rockburst risk assessment,’’ Engineering, vol. 3,
evaluation and prediction in space-time for safety of rock engineering no. 4, pp. 552–558, 2017.
under hazardous environment,’’ Chin. J. Rock Mech. Eng., vol. 27, no. [216] D. Lahat, T. Adali, and C. Jutten, ‘‘Multimodal data fusion: An overview
9, pp. 1741–1756, 2008. of methods, challenges, and prospects,’’ Proc. IEEE, vol. 103, no. 9,
[195] Y. Vasuki, L. Yu, E.-J. Holden, P. Kovesi, D. Wedge, and A. H. Grigg, pp. 1449–1477, Sep. 2015.
‘‘The spatial–temporal patterns of land cover changes due to mining [217] L. Wang, S. Xu, J. Qiu, K. Wang, E. Ma, C. Li, and C. Guo, ‘‘Automatic
activities in the darling range, Western Australia: A visual analytics monitoring system in underground engineering construction: Review and
approach,’’ Ore Geol. Rev., vol. 108, pp. 23–32, May 2019. prospect,’’ Adv. Civil Eng., vol. 2020, pp. 1–16, Jun. 2020.
[196] J.-D. Wansink, ‘‘The potential of advanced data analytics using machine [218] T. Meng, X. Jing, Z. Yan, and W. Pedrycz, ‘‘A survey on machine learning
learning to increase overall equipment effectiveness in an underground for data fusion,’’ Inf. Fusion, vol. 57, pp. 115–129, May 2020.
mining operation,’’ thesis, TU Delft Civil Eng. Geosci. TU Delft Geosci. [219] A. Chauhan, O. P. Malviya, M. Verma, and T. S. Mor, ‘‘Blockchain and
Eng., 2017. scalability,’’ in Proc. IEEE Int. Conf. Softw. Qual., Rel. Secur. Companion
[197] T. V. Randhavane, A. Bera, E. Kubin, K. Gray, and D. Manocha, ‘‘Mod- (QRS-C), Aug. 2018, pp. 122–128.
eling data-driven dominance traits for virtual characters using gait analy- [220] M. Nofer, P. Gomber, O. Hinz, and D. Schiereck, ‘‘Blockchain,’’ Bus. Inf.
sis,’’ IEEE Trans. Vis. Comput. Graphics, vol. 27, no. 6, pp. 2967–2979, Syst. Eng., vol. 59, no. 3, pp. 183–187, Mar. 2017.
Jun. 2021. [221] P. Wang, C. Hu, and M. Xu, ‘‘A blockchain based secure data transmission
[198] C. Lee, S.-M. Kim, and Y. Choi, ‘‘Case analysis for introduction of mechanism for electric company,’’ in Proc. Data Sci., 6th Int. Conf., in
machine learning technology to the mining industry,’’ Tunnel Under- Communications in Computer and Information Science, J. He, P. S. Yu,
ground Space, vol. 29, no. 1, pp. 1–11, 2019. Y. Shi, X. Li, Z. Xie, G. Huang, J. Cao, F. Xiao, Eds. Singapore: Springer,
[199] X.-T. Feng and Q. Jiang, ‘‘Application of intelligent rock mechanics 2020, pp. 515–520.
methodology to rock engineering,’’ Water Eenrgy Int., vol. 67, no. 7, [222] K. Salah, M. Rehman, N. Nizamuddin, and A. Al-Fuqaha, ‘‘Blockchain
pp. 1–10, 2010. for AI: Review and open research challenges,’’ IEEE Access, vol. 7,
[200] G. Su, X. Feng, Q. Jiang, G. Chen, and Q. Ge, ‘‘Intelligent method of pp. 10127–10149, 2019.
combinatorial optimization of excavation sequence and support param- [223] D. Mazzei, G. Baldi, G. Fantoni, G. Montelisciani, A. Pitasi, L. Ricci, and
eters for large underground caverns under condition of high geostress,’’ L. Rizzello, ‘‘A blockchain Tokenizer for industrial IoT trustless applica-
Chin. J. Rock Mech. Eng., vol. 26, no. S1, pp. 2800–2808, 2007. tions,’’ Future Gener. Comput. Syst., vol. 105, pp. 432–445, Apr. 2020.
VOLUME 11, 2023 40627

RUIYU LIANG received the B.E. and M.E. SERKAN SAYDAM received the B.Sc., M.Sc., and
degrees in mining engineering from the School Ph.D. degrees in mining engineering from Dokuz
of Resources and Civil Engineering, Northeast- Eylul University, Izmir, Turkey. He has completed
ern University, Shenyang, China, in 2016 and the Postdoctoral Fellowship with the University of
2020, respectively. He is currently pursuing the Witwatersrand, Johannesburg, South Africa. He is
Ph.D. degree with the School of Minerals and currently the Chair of Mining Engineering with the
Energy Resources Engineering, UNSW, Sydney, School of Minerals and Energy Resources Engi-
in 2020. He focuses on unity-based 3-D interactive neering, UNSW. A key focus of his research is
visualization development, visual analytics, and to address the current needs and future challenges
fusion of visualization and analytics in applica- faced by the minerals industry. He has established
tions. In the big data era, he is looking forward to fusing data visualiza- research collaboration with NASA, ESA, and Luxembourg Space Agency,
tion and data analytics and providing more insights into geological hazard as well as more than 40 research organizations and universities globally. His
prevention. His research interests include data visualization and analytics in research interests include space resources engineering, ground control, mine
underground mining. systems design, the mine Internet of Things, and technology integration and
management. He is currently a Fellow Member of the AusIMM; the President
of the ISRM Commission on Planetary Rock Mechanics; the Deputy Director
CHAORAN HUANG (Member, IEEE) is currently of the Australian Centre for Space Engineering Research (ACSER), UNSW;
a Postdoctoral Fellow (a Research Associate) with and the Deputy Secretary-General and a Council Member of the Society of
the School of Minerals and Energy Resources Mining Professors (SOMP).
Engineering, The University of New South Wales,
working with Dr. Chengguo Zhang and Prof. Ismet
Canbulat. He was supervised by Prof. Lina Yao
with the School of Computer Science and Engi-
neering. He is working closely with her and her
group D2 . He has concentrated on service-oriented
computing, while much of his current research is
focused on the data mining, machine learning, and the IoT data analytics.
CHENGGUO ZHANG received the B.E. degree in

mining engineering from the China University of
Mining and Technology, in 2010, and the Ph.D.
degree in mining engineering from UNSW, Syd-
ney, in 2015. He works in the discipline of mining
engineering, particularly in mine geomechanics
research throughout his career, aiming to improve
safety and sustainability in mines by conducting
fundamental and applied research. His research ISMET CANBULAT received the B.Sc. degree
interests include the critical areas in the mining from Istanbul Technical University, the M.Sc.
industry, such as dynamic rock mass failures, the interaction of mining and degree from the University of the Witwatersrand,
groundwater, and mine data analysis and visualization. His current research and the Ph.D. degree from the University of Preto-
focuses on integrating data analysis and visualization technologies into the ria. Since 2014, he has been a Professor, the Head
assessment of rock mass behavior and management of geotechnical risks in of the School, and the Kenneth Finlay Chair of
deep mining environments, with more emphasis on the hard rock mines. His rock mechanics with the School of Minerals and
research leads to novel and unique solutions that previously have limited Energy Resources Engineering, UNSW Sydney.
understanding and directly applies and translates to industry practices. He is the author or coauthor of numerous technical
papers published in refereed journals and confer-
ence proceedings. His peer-reviewed research reports were submitted to
BINGHAO LI (Senior Member, IEEE) received funding organizations in Australia and South Africa and the consultancy
the B.E. degree in electrical and mechanical engi- reports were submitted to over 60 metalliferous and coal mines. His principle
neering from Beijing Jiaotong University, China, research interests and expertise include rock mechanics theory and applica-
in 1994, the M.E. degree in civil engineering tion, ground control and rock mass classification, design of pillar and ground
from Tsinghua University, China, in 2001, and support systems, numerical modeling in addition to other areas in mining
the Ph.D. degree in spatial information systems engineering, including mine design, layout selection, geotechnical impact
from The University of New South Wales, Sydney, assessment, advanced monitoring, risk management and advanced risk-based
Australia, in 2006. He is an Associate Professor design and evaluation, and mine closure and subsidence assessment. He is a
and a Leader of the MIoT and IPIN Laboratory, fellow of the Institute of Engineers Australia and the Australasian Institute
School of Minerals and Energy Resources Engi- of Mining and Metallurgy; and a member of the Australian Geomechanics
neering, The University of New South Wales. His research interests include Society, the International Society for Rock Mechanics, and the Society of
indoor positioning, pedestrian navigation, satellite positioning and naviga- Mining, Metallurgy and Exploration.
tion, and the mine Internet of Things.
40628 VOLUME 11, 2023

Exploring The Fusion Potentials of Data Visualization and Data Analytics in The Process of Mining Digitalization

Uploaded by

Copyright:

Available Formats

Exploring The Fusion Potentials of Data Visualization and Data Analytics in The Process of Mining Digitalization

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

Exploring The Fusion Potentials of Data Visualization and Data Analytics in The Process of Mining Digitalization

Uploaded by

Copyright:

Available Formats

Received 9 March 2023, accepted 11 April 2023, date of publication 17 April 2023, date of current version 28 April 2023.

Digital Object Identifier 10.1109/ACCESS.2023.3267813

Exploring the Fusion Potentials of Data

I. INTRODUCTION challenges in managing, visualising, and analysing raw sen-

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

VOLUME 11, 2023 40609

40610 VOLUME 11, 2023

FIGURE 1. Visualisation revolution routine [86].

dimensions which poses challenge on data analytics and visu-

VOLUME 11, 2023 40611

40612 VOLUME 11, 2023

to visualise geological data such as seismic data, geolog-

VOLUME 11, 2023 40613

applications in handling complex non-standarded structures

40614 VOLUME 11, 2023

VOLUME 11, 2023 40615

FIGURE 6. The workflow and an earth model data structure [16].

FIGURE 5. Data sources in the mining industry.

mining equipment fleet are vital in mining production,

40616 VOLUME 11, 2023

TABLE 3. Some examples of mining information systems.

text, speech, and images are increasingly being used to gain

VOLUME 11, 2023 40617

40618 VOLUME 11, 2023

VOLUME 11, 2023 40619

40620 VOLUME 11, 2023

C. DATA FUSION analytics to these diverse data sources, it is possible to

VOLUME 11, 2023 40621

40622 VOLUME 11, 2023

VOLUME 11, 2023 40623

40624 VOLUME 11, 2023

VOLUME 11, 2023 40625

40626 VOLUME 11, 2023

VOLUME 11, 2023 40627

CHENGGUO ZHANG received the B.E. degree in

40628 VOLUME 11, 2023

You might also like