1. Introduction
In recent years, with the rapid development of automation technology and artificial intelligence, mining operations have undergone a historic transformation from traditional manual labor to mechanization, automation, digitization, and intelligentization [1,2]. Intelligent mining technology encompasses multiple aspects, including the autonomous operation of intelligent equipment, data collection and big data analysis, IoT monitoring, edge computing, and cloud platform integration. It demonstrates tremendous potential in ensuring safe production, reducing operational costs, improving resource utilization efficiency, and promoting green and environmentally friendly practices [3,4].
The intelligent construction of mines has gradually achieved certain results in equipment implementation, management and maintenance, platform architecture, and industry ecosystem; yet, it still faces significant challenges [5]. First, in terms of the end-side equipment, the intelligence levels are uneven, with incompatible interfaces and inconsistent standards among different devices, making coordinated scheduling difficult. The equipment stability is insufficient in high-dust, high-temperature, and high-vibration environments. The edge computing and on-site decision-making capabilities are also limited, with the excessive reliance on cloud processing being constrained by the communication latency [6]. Second, in terms of operation and maintenance management, there is a lack of a unified data-sharing mechanism among different subsystems. The prediction models exhibit a low accuracy and insufficient early warning capabilities. The operation and maintenance processes lack standardization, increasing system complexity and maintenance costs [7]. Third, in terms of platform development, most platforms suffer from low integration, scattered functionalities, poor scalability, and weak security mechanisms, making it difficult to meet the demands for efficient collaborative control and data security [3]. Finally, at the level of industrial ecology, there is a lack of coordination mechanisms between upstream and downstream players in the industrial chain. Insufficient talent reserves, especially the financial and technical support difficulties faced by small- and medium-sized mines, have constrained the advancement of intelligent construction [8]. These issues reflect that the current development of intelligent mining is still in a critical stage of system integration and capability enhancement. Therefore, this paper will conduct an in-depth analysis of the latest research achievements and practical experiences in intelligent mining technologies, exploring how to further promote the intelligent transformation of mines under unified planning and standard guidance.
The paper first reviews the development history and evolutionary trajectory of intelligent mining technology, analyzing the inherent logic of intelligent mining’s progression from mechanization to automation and then to intelligence. Secondly, we summarize the practical achievements and shortcomings of the existing technologies in three aspects—end side, edge side, and cloud side—with a particular focus on the intelligent mining segment. Cutting-edge technologies are being increasingly implemented and applied in the mining industry, primarily focusing on the mineral resource extraction phase (i.e., the mining segment), specifically the intelligentization of mining equipment, tools, and systems. Finally, based on the existing technologies and application challenges, this paper focuses on cutting-edge issues and proposes urgently needed development directions, providing systematic guidance for relevant researchers and engineering practitioners.
2. Related Work
To ensure a systematic and comprehensive review, this study adopted a structured literature review approach. Relevant publications were retrieved from major databases, including Web of Science, IEEE Xplore, Springer, Elsevier, and CNKI. The search covered the period from 2010 to 2025, using keywords such as “intelligent mining”, “mine automation”, “cloud-edge-end architecture”, “mining industrial internet”, and “edge computing in mines”. Only peer-reviewed journal articles, reviews, and conference papers were included. After preliminary screening, approximately 96 documents were selected for a full-text review. The selected literature was analyzed and categorized to structure the discussion and facilitate a comparative analysis.
With the rapid development of intelligent mining technologies, some comprehensive studies have addressed key issues in the field. Zhang et al. [5] proposed a four-tier framework for intelligent unmanned mining, covering the basic theory, core technologies, operational modes, and overall system architecture. Their study reviewed the applications of 5G, IoT, big data, and related technologies in mining perception, automation, and safety monitoring. They also outlined a three-stage evolution path from remote control to full-process autonomy, providing a clear roadmap for the development of unmanned mining systems.
Li et al. [9] focused on geological support technologies in intelligent coal mines. They introduced a multi-source detection fusion framework based on aerial, drilling, and seismic data, and proposed a dynamic 3D modeling method. The study emphasized the role of visualization platforms in situational awareness and risk prediction. It further suggested integrating knowledge graphs and intelligent algorithms to enhance geological anomaly detection and advocated for open, shared geological data platforms to resolve data fragmentation issues.
From a strategic perspective, Wang et al. [10] outlined the “Smart Coal Mine 2025” vision, proposing eight subsystems supported by IoT, big data, AI, and cloud computing. They highlighted the goal of achieving “unmanned surface operations” and emphasized the importance of unified evaluation standards for intelligent mine construction. Zhang et al. [11] addressed intelligent mining under complex geological conditions, identifying key challenges such as roof stability and roadway deformation. They proposed technical solutions including coordinated roof support and precise hydraulic control, and discussed future directions such as collaborative equipment control, real-time simulation, and intelligent decision systems.
For metal mines, Cai et al. [2] compared the development of deep intelligent mining technologies in China and abroad. They analyzed automation practices in international mines such as Kiruna and Rio Tinto and pointed out gaps in China’s sensing capabilities, efficient mining technologies, and system integration. Zheng et al. [12] reviewed the current status of intelligent monitoring and early warning in coal mines. They identified issues including incomplete precursor data collection, weak system coordination, and limited AI applications, and recommended progress in sensor technology, multi-source data fusion, and algorithm integration. Hu et al. [4] examined the state of intelligent operations in China’s metal mines, covering drilling, blasting, loading, transport, and platform integration. They noted practical issues such as prioritizing display over functionality and data collection over analysis, and called for improvements in system effectiveness and operational maintenance.
In summary, most existing studies focus on specific technologies, application domains, or individual sectors such as coal or metal mining. Although these works provide valuable insights into key technologies and development pathways, several gaps remain. First, few studies examine cross-layer integration from a comprehensive cloud–edge–terminal perspective. Second, foundational elements such as data standards, semantic interoperability, and knowledge representation have received limited attention. Third, there is a lack of systematic analysis on how standardization bottlenecks constrain the scalability and reliability of intelligent mining systems.
To address these gaps, this review is structured around four core themes: intelligent infrastructure, intelligent equipment, system-level collaboration, and technical bottlenecks. Specifically, we look at the following: (1) Mechanized Operations and Autonomous Execution focuses on the closed-loop control and coordination of smart equipment; (2) the section on Systematic Management and Control explores edge-layer data collaboration and protocol adaptation for heterogeneous devices; (3) Edge-side Service Enablement covers dynamic modeling, risk identification, and equipment scheduling; (4) Cloud-side Support Technologies reviews the deployment of industrial internet platforms, intelligent decision systems, and planning tools in real-world mining scenarios; and (5) the section on Challenges and Development Trends synthesizes core issues related to standardization, data intelligence, and intelligent equipment evolution.
Through this framework, the paper aims to provide a comprehensive technical roadmap for cross-device, cross-system, and cross-process coordination in intelligent mining. It also emphasizes the importance of standard systems and knowledge representation as foundational components, thereby supporting future advancements in the digital and intelligent transformation of the mining industry.
3. Overall Development Process of Intelligent Mining Construction
Intelligent mining is a comprehensive system that deeply integrates advanced technologies into the entire mining process, with the goal of achieving safe, efficient, and green production. Over the decades, the construction of intelligent mining has undergone four stages: single-machine automation, comprehensive informatization, digital integration, and full-scale intelligence [13,14], shown in Figure 1. An analysis of the core technological characteristics and existing issues at each stage and period is as follows:
3.1. Stand-Alone Automation Stage
This stage was primarily characterized by mechanization and automation, focusing on the automatic control and monitoring of individual machines [15]. PLC/DCS control systems and basic SCADA monitoring systems were introduced to automate key equipment such as hydraulic supports and shearers in fully mechanized mining faces. Concurrently, basic mine communication networks (field bus, and local area networks) were established to enable localized data transmission. The application of automation control technologies laid the preliminary informatization foundation for mine production processes.
Major mining equipment was retrofitted with automatic control modules, such as the electro-hydraulic control systems and PLC monitoring devices installed on hydraulic supports and shearers in Chinese coal mines during the 1990s [16]. Additionally, some mines began piloting underground safety monitoring systems to enable the online monitoring of critical parameters like th egas concentration and ground pressure. Overseas, mines in Australia and Europe also adopted single-machine automation equipment and safety monitoring devices during this period, but overall system integration remained limited.
At this stage, the level of informatization is uneven, with equipment and systems mostly operating independently. The lack of unified standards and interfaces among various systems prevents effective data sharing. The limited network bandwidth and computing capacity result in poor real-time performance and slow response. Overall, automation is confined to isolated points or localized areas, unable to support mine-wide collaborative control or refined management.
3.2. Integrated Automation and Informatization Stage
At this stage, the mine began large-scale informatization construction, focusing on expanding network coverage and introducing mining Ethernet, fiber-optic communication, and SCADA integrated monitoring systems [16,17]. By establishing a field bus network, centralized monitoring and the coordination of major production equipment and transportation systems were achieved. Enterprise-level IT systems (such as mine GIS and ERP) started to be implemented. Three-dimensional geological modeling, manufacturing execution systems (MESs), and resource management systems were gradually deployed [18].
During this period, large coal enterprises established integrated monitoring centers to interconnect mining, transportation, ventilation, and other systems. Taking the Shendong Coal Mine as an example [19], it initiated the digital mine construction project relatively early and implemented remote monitoring and automated production scheduling for some mining faces. Some large-scale mines abroad have also deployed remote control rooms connected to field equipment via satellite/fiber optics, enabling the remote operation of heavy machinery (such as drills and bulldozers). Overall, this stage marks a transition from “single-machine automation” to “system integration”.
At this stage, there are still numerous isolated monitoring platforms among different subsystems, with a lack of standardized data interfaces, making it difficult to aggregate data onto a unified platform. The network bandwidth and system performance still fall short of meeting the demands of real-time, large-scale data processing. Additionally, the complex underground mining environment imposes higher requirements on the adaptability and reliability of automated equipment, posing challenges for widespread adoption.
3.3. Digital and Intelligent Initial Stage
During this stage, cloud computing, big data, and IoT technologies continued to evolve [10,20]. Mines established end-to-end networks, integrating various underground sensors into the IoT to enable the rapid collection and transmission of equipment and environmental data. Data storage and processing shifted from localized databases to cloud platforms, giving rise to data platforms and data warehouses. Concurrently, intelligent system layers (AI analytics, visual decision-making, and digital twins) began to emerge, with some mines adopting remote-controlled and automated equipment, unmanned inspection robots, and unattended facilities [21]. Architecturally, mines developed a hierarchical system comprising the “equipment layer–control layer–processing and analytics layer–management layer–decision-making layer”, while establishing unified mine databases and standardized workflows.
The Shendong Jinjie Coal Mine is a landmark digital mine project in China, having developed a 100-million-ton regional coal mine big data collection and analysis platform [22]. This system integrates data from 5 mining areas, 42 equipment manufacturers, and over 3000 devices, covering more than 70,000 collection points to achieve data standardization and interconnectivity. Leveraging this platform, Shendong Coal Mine has implemented big data applications such as energy efficiency analysis, predictive equipment maintenance, and disaster early warning, significantly enhancing safety and operational efficiency. Internationally, the “Remote Operations Center” model has been adopted in countries like Australia [23]. For instance, BHP established a centralized remote control center for mine production in Perth, relocating mine operators from on-site to the center, which substantially improved operational efficiency.
During this stage, the data platform continues to be constructed, and the degree of system integration has significantly improved. However, the data compatibility and fusion between heterogeneous systems remain challenging. In the face of massive sensor data, the real-time requirements have surged, while network latency and computational bottlenecks may lead to suboptimal responsiveness. The promotion of intelligent systems is constrained by the demand for personnel skills and substantial investments, with some enterprises adopting a wait-and-see attitude towards the return on investment. Overall, smart mines at this stage, both domestically and internationally, have entered an exploratory application period but generally remain in the preliminary testing phase.
3.4. Comprehensive Intelligence and “Cloud–Edge–End” Collaboration Stage
The current intelligent mining sector has entered a period of rapid advancement, with a focus on promoting “cloud–edge–end” collaboration and the application of artificial intelligence (AI) and digital twin (DT) technologies [24,25]. The low-latency and high-bandwidth characteristics of 5G communication support underground remote control and high-precision positioning. Edge computing platforms share data processing tasks to ensure real-time response for critical applications. Artificial intelligence and deep-learning technologies are being utilized for fault diagnosis, safety behavior recognition, and resource optimization scheduling. Digital twin technology is beginning to be applied in scenarios such as mine planning, process optimization, and emergency training, providing virtual simulation support for decision-making. The entire system emphasizes “data-driven” operations with full-link coordination across all subsystems.
At present, despite the accelerated iteration of technologies, the comprehensive coordination of intelligent mines still faces challenges. There is an urgent need for national standards and industry norms to bridge the gaps in data interfaces and protocols. Data security and privacy protection have also emerged as new issues, with large-scale remote operations imposing stringent requirements on network reliability. Additionally, big data and AI models rely on high-quality professionals, but the existing talent pool in mining enterprises is generally skewed toward mining engineering, with an insufficient understanding of new technologies. Some small- and medium-sized mines, limited by resources and conditions, struggle to keep up with the construction pace and require differentiated, incremental deployment strategies.
In summary, the construction of intelligent mines has evolved from the stand-alone automation of the late 1990s to the current intelligent system characterized by cloud–edge–end collaboration and AI-driven technologies. At each stage, experience is continuously accumulated by establishing foundations through automation and informatization, achieving data aggregation via digital platforms, and leveraging the role of systematic collaboration. Moving forward, it is essential that we address key challenges such as data integration, standardization coordination, and talent cultivation to fully leverage the benefits of the overall intelligent mine architecture and achieve the goal of safe and efficient modern mining. Future efforts must continue to tackle key issues such as data integration, standard coordination, and talent cultivation to fully leverage the benefits of the intelligent mine’s overall architecture and achieve the goal of safe and efficient modern mining.
4. Current Development Status of Intelligent Mining Technology
The “Guidelines for Intelligent Mine Construction in the Nonferrous Metals Industry” points out the establishment of a “cloud–edge–end” architecture system [26]. The end side achieves comprehensive perception and precise control through the intelligent transformation of production equipment and the application of complete sets of intelligent equipment. The edge side fully utilizes the data from existing and newly built control systems in mines, aggregates regional data resources, and realizes data analysis and real-time decision-making. The cloud-side integrates industrial microservices, big data services, etc., to achieve data aggregation and modeling analysis, and software-based industrial experience and knowledge, as well as the development and operation of various applications. The architecture and scenario illustration are shown in Figure 2.
In the three-tier architecture, the cloud side predominantly leverages public or private cloud platforms to establish a unified data middle platform and control center, responsible for massive data storage and deep-learning model training. The edge side serves as the adhesive connecting the cloud and the device side, acting as an adapter for the normalized operation of intelligent equipment and scenarios. It is responsible for real-time data processing, model inference, instant response, and security protection. The end side encompasses various sensors, smart terminals, and execution devices, acting as the “data source” and “executor” of the entire architecture, responsible for on-site perception and execution control. This section provides a review of the current development status of the technologies at each layer.
4.1. End-Side Technology
The characteristics of end-side technology lie in its comprehensive perception of the working environment and equipment status, with the deployment of multi-source sensors enabling intelligent decision-making in mining operations [27]. End-side technology plays a critical role in data acquisition, preliminary processing, and real-time control. The overall development status of end-side technology is shown in Figure 3.
Data Acquisition and Multi-Source Sensing
Efficient and reliable data acquisition serves as the foundational pillar for intelligent mining systems. In complex and dynamic mining environments, multi-source sensing technologies enable the comprehensive perception of environmental conditions, equipment status, and operational activities. This section systematically reviews key sensing modalities (including environmental sensors, equipment diagnostics, video surveillance, LiDAR scanning, and multimodal data fusion) that collectively form an intelligent perception layer supporting real-time monitoring, safety assurance, and autonomous decision-making in modern mines.
Environmental monitoring sensors: Install gas sensors (methane, gas, oxygen, carbon monoxide, etc.) and environmental sensors (temperature, humidity, dust, etc.) at key locations (working faces, transportation roadways, and stope entrances) to monitor mine ventilation and hazardous gas accumulation in real time. These sensors transmit data to edge nodes via wireless sensor networks, forming an environmental safety monitoring system for the mine. By analyzing the environmental data, abnormal conditions such as hazardous gas exceedances, high temperatures, and fire precursors can be promptly detected [27].
Equipment status sensors: Install sensors on mechanical equipment (such as vibration, pressure, bearing temperature, current, voltage, etc.) to monitor the operational health status of the equipment. These data are aggregated through edge control units, facilitating operational status diagnosis and predictive maintenance, thereby improving safety and equipment utilization. For example, practical applications include technologies such as drill rod fracture warning for rock-drilling jumbos, bearing fault prediction for hoists, and motor overheating protection for autonomous mining trucks [28].
High-definition (hd) video surveillance: Install high-definition cameras at mining faces, ventilation doors, and key operational points to conduct the real-time monitoring of excavation surfaces, equipment operation, and personnel positioning [29]. Hd video data is transmitted via 5G or underground optical fiber to edge computing nodes, where it is processed using video analytics and deep-learning algorithms for pedestrian detection, behavior recognition, and anomaly alerts [30]. Examples include alarms for personnel entering hazardous zones [31], the monitoring of the unloading process by load–haul–dump (LHD) machines, and the detection of conveyor belt damage [32].
LiDAR Scanning: Three-dimensional laser scanning technology is increasingly being applied in mine surveying and safety inspections. Compared with traditional single-point monitoring instruments, LiDAR can rapidly acquire three-dimensional point clouds of tunnels and stopes, enabling the precise modeling of terrain, landforms, and structures [33]. As fixed laser scanning requires substantial manpower and resources, mobile SLAM LiDAR systems have gained popularity, capable of obtaining the complete point cloud data while traversing underground tunnels [34]. By comparing point clouds scanned at different times, potential safety hazards such as tunnel cross-section deformation and rock layer collapse can be accurately identified [35]. Vehicle-mounted and airborne LiDAR systems are widely used in open-pit mines for autonomous driving and scene reconstruction [36].
Multimodal perception fusion: To enhance the perception capability in complex environments, it is necessary to integrate data from multiple sensors. For instance, obstacle detection algorithms combining image and LiDAR point cloud data can improve detection accuracy [37]. Fire safety prediction algorithms incorporating gas concentration and temperature–humidity data enable a more precise risk assessment. On the end side, future development will emphasize the application of multi-sensor information fusion technology to achieve intelligent monitoring throughout the entire mining production process.
2.. Mechanized Operations and Automated Execution
The mechanization of operations refers to the use of electricity or other power sources to drive and operate mechanical equipment, replacing manual labor in production tasks throughout the mining process flow and ore flow. Mechanization serves as the foundation for intelligent mining. It not only aims to reduce physical labor and enhance labor productivity, but also focuses on expanding the extraction scale and promoting scientific production, organization, and management.
Execution automation refers to the ability of the equipment (or facility) on the end-side, which runs through the entire mining process flow and ore flow, to automatically execute its tasks. This automation has two meanings: one is to achieve fixed-mode inputs (data, parameters, etc.) and outputs (data, control, etc.) through programmable means, and the other is to achieve inputs and outputs in specific scenarios through the deep-learning process of neural networks. Ultimately, the equipment (or installation) can achieve automation in detection, information processing, analysis and judgment, operation control, and other aspects in order to accomplish the intended objectives.
Mechanized systems and automated control form the operational backbone of intelligent mining, driving both productivity and safety improvements. This section examines the evolution and integration of key equipment (from precision drilling rigs and autonomous shearers to unmanned haulage vehicles and supportive fixed machinery), highlighting the advances in control algorithms, multi-sensor coordination, and digital twin applications. By detailing these technologies, we illustrate how mechanization and automation synergize to enable high-throughput, resilient, and autonomous mining workflows.
Rock-drilling equipment: In various mining scenarios, the technological evolution of rock-drilling equipment focuses on precision, unmanned operation, and multifunctional integration [21]. Traditional hydraulic drilling rigs have achieved the automatic matching of borehole coordinates with 3D digital maps by incorporating high-precision positioning and navigation systems, with errors controllable within ±100 mm. Intelligent down-the-hole drills represented by the Simba ME7 significantly improve the drilling efficiency in complex rock formations through multi-parameter adaptive control technology (the real-time matching of impact pressure, rotation speed, and feed force). Full-face tunnel boring machines (TBMs) are also being further applied. The Xianshan Iron Mine achieved a rapid excavation of the belt incline shaft through TBM technology, improving the efficiency by three times compared to traditional blasting methods.
Excavation equipment: The shearer, as the key cutting equipment in fully mechanized coal mining faces, has evolved from manual operation to automated control. Early mechanized coal mining relied on workers manually controlling the shearer’s movement and cutting. With the development of automation technology, shearers have gradually been equipped with electronic control systems that can automatically adjust the cutting speed and depth. The literature indicates [21] that existing technologies allow shearers to work in coordination with hydraulic supports and armored face conveyors, forming a “three-machine” automation system. Roadheaders are used for the excavation and expansion of mining faces and main inclined roadways. The current research primarily focuses on improving the trajectory control and anti-interference capabilities, employing algorithms such as adaptive robust control to address rock deformation and uncertain disturbances. For example, WANG et al. [38] proposed a sliding mode control strategy based on a disturbance observer to address the precise positioning and trajectory control issues of the roadheader robot cutting head in complex underground environments.
Loading and hauling equipment: Loading and hauling equipment (such as electric shovels, loaders, etc.) is making breakthroughs in unmanned operation and collaborative work, with autonomous navigation, swarm coordination, and remote control as its core technologies [39]. Underground load–haul–dump (LHD) vehicles have achieved unmanned operation in complex tunnels through multi-sensor fusion technology (LiDAR, inertial navigation, and machine vision) [40]. Additionally, automatic weighing and data integration technologies provide precise data support for production scheduling. Current challenges focus on dynamic obstacle avoidance and the optimization of multi-machine collaborative path conflicts, requiring the enhanced adaptability of AI decision-making algorithms.
Transportation equipment: The automated transportation system in mines is a crucial component for enhancing production efficiency [41]. Large-scale mines both domestically and internationally have already adopted unmanned mining vehicles (primarily underground dump trucks, underground unmanned electric locomotives, and open-pit mining trucks). Technological innovations are reflected in three aspects: autonomous driving, intelligent dispatching, and green power. For instance, automatic path tracking is achieved through onboard sensors and control units, enabling obstacle recognition and avoidance in complex environments. The mine scheduling system realizes the real-time data interaction and coordination of loading and transportation, and the loading cycle time is greatly shortened, which further improves the operation efficiency.
Auxiliary mechanical equipment: The technological development characteristics of auxiliary equipment (including shearer auxiliary hydraulic supports, roadway support equipment, hoists, belt conveyors, ventilation doors, electrical doors, etc.) are unmanned operation, dynamic regulation, and digital twins [5,13]. These devices are embedded with sensors and control units for intelligent upgrades. For example, the hydraulic support adopts an intelligent monitoring device, which can automatically detect parameters such as load and displacement, and realize the automatic follow-up and displacement propulsion with the operation of the shearer. The belt conveyor is equipped with belt tension and vibration sensors, which can realize fault warning. The hoist is interconnected with the roadway monitoring system, enabling the automatic adjustment of the lifting speed to adapt to changes in the ore flow. The application of the digital twin platform achieves three-dimensional visual monitoring and fault prediction for ventilation, drainage, and power supply systems.
4.2. Edge-Side Technology
Mine construction generally recognizes the importance of building a “cloud–edge–end” collaborative system. As a key link between the “cloud” and the “end”, the edge layer plays a central role. BL et al. [42] defined mine edge intelligence as a novel technology system specifically designed for mine production environments. Its core goal is to promote the effective integration of multiple types of ‘cloud–edge–end’ systems on the edge layer by constructing a comprehensive perception, analysis, and decision-making system on the edge side of the mining operation. And, by enabling intelligent equipment and enabling upper-level decision-making, the production efficiency of intelligent mines breaks through the bottleneck of improvement. The development of mine edge intelligent technology is shown in Figure 4.
1.. Systematic Management and Control.
Systematic management and control refer to the systematic construction of the entire mining process flow and ore flow on the mine edge side, including infrastructure development, perception fusion, cluster scheduling. The systematicness can be interpreted as systematic engineering, that is, the use of various organizational management techniques to coordinate and cooperate with the overall and local relationships of the entire mining system. Finally, the overall optimal operation is realized.
The seamless integration of communication, data processing, and device interoperability is essential for coherent mine-wide operations. This section explores how edge-based infrastructures (comprising low-latency networks, localized control nodes, and adaptive protocol translation) collectively enable real-time monitoring, decision orchestration, and cross-layer collaboration. By detailing these capabilities, we demonstrate how systematic management at the edge ensures robust, coordinated control across heterogeneous mining assets.
Low-latency network communication: Edge computing must rely on high-speed and reliable networks to enable the rapid transmission of massive data and ensure timely control execution [43]. Modern mines utilize next-generation network technologies such as 5G and private networks to establish a campus-level infrastructure, achieving an ultra-wide bandwidth and low-latency communication. For instance, the large uplink bandwidth of 5G networks supports the high-volume transmission of high-definition video streams with a latency below 100 ms, meeting the real-time requirements of remote control equipment [44]. Communication between edge nodes and the cloud is facilitated via industrial Ethernet or optical fiber. High-speed broadband and network slicing provide isolated guarantees for diverse services (monitoring, control, data synchronization, etc.), thereby significantly enhancing the flexibility and reliability of mine communication networks.
The edge layer coordinates the information interaction between the cloud and end side: Edge computing nodes often serve as local control centers for mining operations, directly participating in monitoring, early warning, and dispatching tasks. They both receive large volumes of raw data from the end side for preprocessing and upload the processed results or summarized data to the cloud for an in-depth analysis [45]. Meanwhile, cloud-based strategies and model updates are also distributed to end devices through the edge layer. As Huawei’s Intelligent Mining Solution indicates that an efficient and reliable communication network is crucial in mining environments to ensure fast and secure data transmission across layers. Therefore, the edge layer acts as a “binding agent” in the architecture, bridging the sensing layer and management layer to enable the collaborative system operation without information silos.
The edge platform serves as the “adapter” for mining scenarios: Mining production equipment comes from various suppliers with different protocols and operates under complex environmental conditions [45]. The edge platform functions as an “adapter” for diverse intelligent devices, handling tasks such as protocol conversion, data format standardization, and communication medium transformation. For instance, for various sensors or edge robots connected to the network, the edge gateway can perform data collection and preliminary fusion while ensuring device compatibility. Moreover, in geographically distributed networks, the edge layer enables the seamless connectivity across different network environments, guaranteeing device interoperability and the smooth transmission of control commands. In summary, the edge layer acts as both the localized “brain” for institutionalized mine control and the critical link supporting “cloud–edge” collaboration, providing intelligent equipment with a seamless communication and cooperative environment.
2.. Edge-Side Service Applications
Building upon the underlying edge infrastructure, service applications translate raw data into actionable insights and coordinated actions. The advancement of edge-side technologies has further consolidated the supporting role of the “cloud–edge–end” integration in intelligent mining operations. It can be summarized into three main application directions: first, the dynamic perception and modeling of the operating environment and work processes; second, the real-time identification and early warning of potential safety hazards; and, third, the efficient scheduling and control of intelligent equipment clusters.
Dynamic perception and modeling of mining environments: The dynamic perception and modeling of mining environments on the edge side primarily rely on multi-source sensor fusion and lightweight machine-learning algorithms. This enables real-time digital twin model updates for mine site topography, equipment distribution, ventilation status, etc. Chen et al. achieved the continuous 3D reconstruction and deformation monitoring of underground tunnel environments by deploying deep learning and sensor data fusion frameworks on edge nodes, improving the timeliness and accuracy of model updates. For mining road and falling object detection, methods such as FusionPlanner [46] utilize multi-sensor fusion on edge nodes to execute multi-task motion planning, balancing lateral and longitudinal control with path optimization to ensure the stable operation of unmanned transport vehicles in high-noise environments. However, multi-source sensor fusion still faces challenges (such as data heterogeneity, spatiotemporal synchronization, and accuracy degradation) in high-noise and heavily occluded mining environments. Additionally, lightweight inference frameworks exhibit insufficient robustness in continuous 3D reconstruction scenarios. Future efforts should focus on developing self-supervised multi-modal fusion algorithms based on sparse–dense hybrid representations. As well as incorporating a multi-level spatiotemporal attention mechanism, the dynamic model achieves adaptive error correction and enables real-time inference with a low power consumption and high throughput.
Safety identification and early warning in mining scenarios: The edge side deploys lightweight computing platforms to run target detection and behavior analysis algorithms, enabling the real-time monitoring and anomaly recognition of scene targets (environmental status, personnel behavior, vehicles, robots, etc.). Imam et al. [47] summarized various computer-vision-based collision avoidance systems. They pointed out that, in extreme low-light and dusty environments, edge-side models need to integrate infrared/visual multimodal data to ensure the detection robustness and response speed. In their PPE compliance study [48], they utilized edge nodes to run a PoseNet variant, achieving the posture estimation and compliance judgment of underground workers wearing safety helmets, providing second-level early warning capabilities for accident prevention. LIU et al. [49] introduced the semi-supervised clustering algorithm (SSCME) with concept drift detection and incremental updates on the edge node side for disaster monitoring in complex underground environments. It can adaptively identify changes in data distribution, eliminate outliers, and update the classification model. This ensures the accurate early warning of sudden ground pressure changes and gas outbursts under low-bandwidth conditions. In summary, there are some problems such as the single-point sensor distortion, centralized model update delay, and lack of interpretability. The existing identification and early warning systems struggle to meet the low-latency response and trustworthiness requirements for emergent risks. The next step is to construct an end-edge federated causal inference framework. For example, lightweight interpretable models are deployed locally to achieve the rapid fusion of multi-source data and mechanism-driven risk decision-making through incremental association reasoning and template adaptation.
Cluster control and scheduling of mining equipment: Data interaction, task allocation, and dynamic scheduling among heterogeneous devices can achieve a leapfrog improvement in the overall efficiency and safety of the mining system. Relevant researchers have integrated edge computing with swarm intelligence algorithms to achieve the coordinated scheduling of multiple excavators, unmanned trucks, and other transportation equipment. Hao et al. [50] proposed an energy-aware edge-scheduling heuristic algorithm. By offloading task allocation and resource migration to edge nodes, it achieved “end–edge” collaborative cluster control, significantly reducing the scheduling latency and energy consumption. By employing deep reinforcement learning for scheduling policy learning, it realized load balancing and real-time task scheduling in heterogeneous device clusters. The cluster scheduling method in the unmanned transportation queue management in the open-pit mine area can reduce the path planning and task switching time of the multi-vehicle formation. In the application of underground unmanned transport vehicles, some studies are automatically planning the driving path according to the mine map and real-time obstacle information [51]. Currently, scheduling strategies driven by deep reinforcement learning have achieved results in simulation environments. However, in heterogeneous edge networks, they still face challenges such as the difficult model migration and the trade-off conflict between “energy efficiency and latency.” Future research should focus on constructing a microservice-based multi-agent collaborative architecture grounded in a service mesh. Additionally, by integrating online transfer learning mechanisms, an intelligent equipment group management and control system with cross-scenario adaptive capabilities should be established.
4.3. Cloud-Side Technology
Industrial Internet [52] is a new type of infrastructure, application model, and industrial ecosystem formed by the deep integration of information and communication technologies with industry. By establishing comprehensive connections among people, machines, objects, and systems, it constructs a new manufacturing and service system that covers the entire industrial chain and value chain. It provides a pathway for achieving digital, networked, and intelligent industrial development. The advancement of cloud-side technologies for intelligent mining is a product of the deep integration between Industrial Internet and the mining industry. It is based on the interconnection among personnel, equipment, mineral resources, information systems, and control systems. Through the management of mine production and operational data (including collection, storage, analysis, modeling, simulation, evaluation, and optimization), it enables intelligent control, dynamic operations, and transformative changes in production organization methods. The cloud-side technology construction architecture is shown in Figure 5.
In this section, we first examine the industrial Internet platform that integrates heterogeneous systems and delivers IaaS, PaaS, and SaaS layers for seamless data aggregation and application deployment. We then explore digital planning and design tools that leverage cloud resources to optimize mine workflows, from geological modeling to production scheduling. Finally, we discuss how cloud-based intelligent decision-making engines (powered by AI, knowledge graphs, and real-time data processing) transform raw information into actionable insights for strategic and operational control.
1.. Industrial Internet Platform
The mining industrial Internet platform serves as the core enabler of intelligent mining. In terms of architectural design, the platform is built upon a unified data layer and algorithm service layer. It establishes a data platform to aggregate operational data from various domains such as excavation, transportation, and safety, providing high-quality data services for upper-layer applications. The IaaS (Infrastructure as a Service) layer virtualizes the hardware layer, shielding the underlying hardware from the upper layers while enabling the elastic expansion of the hardware resources for the upper-layer software. The PaaS (Platform as a Service) layer employs a domain analysis to abstract various business domains, forming managed services that provide business logic processing for the upper layers. Additionally, it offers common technical components and delivers unified cloud-based foundational services. The SaaS layer directly provides application hosting services for user operations. It utilizes the platform’s application configuration capabilities to define business applications.
The platform supports multi-system integration. It interconnects traditional IT/OT systems (such as ERP, SCADA, DCS, MES, etc.) through the industrial Internet to achieve full-process visualization and collaborative control. For example, Huawei proposed a mining industrial Internet architecture centered on Mining HarmonyOS, an industrial bearer network, a cloud infrastructure, a digital platform, and intelligent applications. It constructs a two-level cloud platform consisting of a group cloud and mining edge cloud. Among them, the group central cloud serves as the AI capability and operation management center, supporting group-level business decisions. The mining edge cloud is deployed on-site, enabling real-time data acquisition and control to achieve cloud-edge collaboration. Under this platform architecture, the mining big data platform utilizes artificial intelligence and data mining to realize intelligent decision-making functions such as energy consumption optimization and resource allocation. Digital twin applications in mining planning can simulate the excavation process and optimize process parameters in a virtual environment.
2.. Planning and Design Digitization
Planning and design digitization refers to cloud-based services that support the entire mining process workflow and ore flow across various planning and design operations. It enables the digitization of processes and outcomes through digital tools (such as data collection, processing, storage, and transmission). This digitization can be interpreted as the mining planning, mining design, production scheduling, and production organization processes and their results exhibiting digital characteristics. It transforms information such as the mining environment, resources, engineering, activities, and production records into measurable numbers and data. Its features include accessibility, computability, and cognizability. Professionals can utilize tools like geographic information systems (GISs), 3D mining software, and mining execution systems to accomplish planning and design tasks.
Mine digital planning and design rely on the powerful computing and storage capabilities of the cloud [53]. In terms of mine production planning and scheduling, cloud-based systems can integrate GISs, production dispatching, and real-time information to achieve cross-temporal and cross-process optimized scheduling. For simulation and optimization, by constructing digital twin models encompassing ore bodies, underground facilities, and transportation systems, the cloud enables a simulation of how different production scenarios impact the output, safety, and energy consumption. Leveraging the elastic computing power of cloud technology, multiple operational strategies can be rapidly evaluated to optimize the blast designs, tunneling paths, and transport scheduling. These cloud-based digital planning technologies help build end-to-end mining solutions and provide a solid foundation for the full lifecycle management of mines.
3.. Digital and Intelligent Decision-Making
Digital and Intelligent Decision-Making refers to the cloud-side application of digital and intelligent technological methods to assist people in completing decision-making throughout the entire mining process flow and ore flow. It encompasses deep data mining and intelligent decision execution capabilities. By leveraging advanced information technologies and intelligent algorithms, people can achieve optimized decision-making and enhanced business execution capabilities.
The cloud platform continuously enhances decision-making support capabilities by integrating cloud-edge collaborative computing, real-time data processing, and AI algorithm analysis [54]. Knowledge graphs exhibit unique advantages in associating multi-source data, query reasoning, and providing intelligent recommendations. They have been proven as an effective method for expressing and integrating diverse types of knowledge. For instance, an intelligent coal mine production and operation platform employs large-scale knowledge graphs and retrieval-augmented generation technology (GraphRAG) to construct a structured knowledge base [55]. It supports intelligent Q&A and precise fault diagnosis, improving the system response speed and decision-making accuracy [55]. Research by Zhang et al. [56] demonstrates that knowledge graphs can semantically associate and store multi-source knowledge, providing informational support for complex decision-making. The constructed Coal Mine Safety Experience Knowledge Graph (CMSEKG) can integrate scattered safety management experience and professional knowledge, offering knowledge services for safe production management.
Decision platforms typically integrate rule engines, machine-learning models, and knowledge graph libraries, providing managers with the panoramic monitoring of production efficiency, safety risks, equipment status, and other aspects through visual interfaces. Cloud-based decision platforms can automatically trigger alerts based on real-time data and offer recommendations for production scheduling and maintenance. Zhong et al. [57] constructed an intelligent decision-making platform integrating the dynamic ventilation calculation and disaster linkage control for mine ventilation and safety. It achieved the online simulation and remote control of ventilation networks. Gold mining enterprises represented by China’s Zhaojin Mining have established a “Intelligent Mine Data Decision Platform”. This platform aggregates over 3000 data points from PLCs, sensors, and other sources. It transmits data to the cloud via 5G and industrial networks, and organizes them into nearly 200 multidimensional data scenarios. A monitoring system is built around indicators such as output, efficiency, equipment, and energy, providing quantitative support for identifying production bottlenecks and evaluating efficiency. Based on this platform, the crushing system was optimized through process modeling and historical data analysis, improving the crushing efficiency by approximately 15% without altering the process.
5. Technical Challenges and Future Development Trends
Intelligent mining is a form of complex systems engineering that integrates information perception, data processing, intelligent decision-making, and automated execution. Facing complex scenarios and multidisciplinary challenges, a unified standard system, comprehensive data governance, reliable equipment intelligence, and high-quality talent support are the keys to achieving systemic breakthroughs.
To better understand the practical implementation and outcomes of intelligent mining, representative cases from domestic large-scale metal and non-metal mines are summarized. These cases reflect the diverse implementation paths of intelligent mining systems, including automated equipment, digital platforms, safety management, and production scheduling, and further demonstrate the integrated application of cloud–edge–end architecture in practical mining scenarios. Table 1 presents an overview of intelligent construction practices across different mining sites, reflecting their technical features and operational benefits.
As shown in the table, intelligent mining construction has yielded significant gains in operational efficiency, safety, and data-driven management. However, the deployment process also reveals persistent challenges, including system interoperability, data fusion barriers, equipment adaptability, and talent shortages. In the following sections, we analyze the key technical obstacles facing intelligent mining and outline the prospective research directions aimed at fostering sustainable and scalable development across the industry.
5.1. Standard Specifications and Evaluation Model Development
As a typical multi-level collaborative architecture, intelligent mine systems require cross-vendor multi-agent collaborative operations. The current technical standard system exhibits a “fragmented” development characteristic. For instance, existing technical specifications such as the “Mine Safety Regulations” lack sufficient integration depth with safety management systems, making it difficult to support needs such as safety situation awareness, risk assessment, and closed-loop control. Notably, industry-consensus evaluation metrics for the Capability Maturity Model (CMM) have yet to be established [79]. And there is a lack of dynamic assessment mechanisms covering the entire system lifecycle, severely constraining the iterative upgrades and overall efficiency improvements of intelligent systems.
One practical issue in system integration stems from communication protocol incompatibility. Many mining enterprises continue to deploy legacy protocols like Modbus, which lack semantic context and scalability, while newer automation systems favor OPC-UA, which supports complex object modeling and security features. The absence of bridging standards complicates interoperability, often requiring custom-built middleware, increasing the system complexity and maintenance burdens [80,81,82]. Similar gaps are evident in data modeling: standards such as ISO 15,926 and ISO 19,156 provide general frameworks for process industries and observation data, respectively, but are not tailored to mining-specific entities such as orebody structures, equipment–event relationships, or dynamically changing geotechnical environments [83,84].
To address these limitations, three strategic directions for standardization and evaluation model development are proposed: Unified semantic modeling system for the mining domain: Ontology-based semantic frameworks—common in domains like manufacturing and construction—are still underdeveloped in mining. Efforts such as the IFC-Mining extension project (BuildingSMART International, 2022) are currently working to extend the Industry Foundation Classes (IFC) to cover the mining infrastructure and operational data, aiming to support semantic interoperability in mine planning and design [85,86]. Similarly, initiatives within the W3C SSN/SOSA framework suggest potential pathways to formalize mining sensor data using RDF/OWL vocabularies [87,88]. However, a domain-specific, multi-layer mining ontology that integrates geology, process engineering, equipment status, and safety knowledge remains a critical research and development gap. Intelligent mining capability maturity evaluation system: While maturity models such as CMMI or CMMM are widely used in IT and smart manufacturing, the mining industry lacks a unified capability maturity model specific to the deployment of intelligent systems. Some coal mines in China have adopted internal classification schemes for automation levels, but these assessments often rely on subjective or isolated metrics [89]. A credible system should follow the principles of the Smart Manufacturing Capability Maturity Model (SMCMM) developed by NIST, incorporating dimensions such as human–machine interaction, data lifecycle management, and autonomous decision capability. This model could guide intelligent transformation through a structured evaluation involving baseline identification, target planning, and iterative improvement. Standard-driven modular technology ecosystem: To support scalable and sustainable development, intelligent mining systems should embrace modular, microservice-based architectures. Although some mining software vendors (e.g., RPMGlobal, Dassault Systèmes, and ABB) have begun modularizing their platforms, true interoperability requires open APIs, standardized data formats, and well-defined service contracts. Organizations such as the China National Institute of Standardization and IEEE Global Mining Guidelines Group (GMG) have advocated for open architecture mining platforms to reduce vendor lock-in and enable the plug-and-play integration of third-party modules, such as perception algorithms or energy optimization components.
5.2. Data Collection and Intelligent Analysis Construction
The mine working environment has complex physical characteristics such as high dust, high noise, and strong vibration. Traditional IoT nodes face data loss, delay jitter, and energy consumption bottlenecks. The big data platform is limited by the bandwidth and computing power distribution, and cannot perform the real-time deep mining of massive sensor data. The intelligent analysis often stays in the post-diagnosis stage.
In the future, the construction of data acquisition and intelligent analysis has the following three research focuses: Adaptive perception–analysis mechanisms: Future systems must incorporate on-device learning and self-optimizing algorithms to adjust modeling parameters and network architectures dynamically. By embedding AutoML pipelines at the edge, perception models can automatically select optimal neural network configurations and fine-tune hyperparameters in response to shifting environmental conditions (such as the varying illumination in tunnels or changes in acoustic background), thereby maintaining a high detection accuracy without manual retraining [90]. This adaptive approach ensures that anomaly detection and predictive maintenance models remain robust across different geological sites and operational states. Multi-modal information fusion: To overcome the limitations of any single sensor modality, intelligent mining platforms should fuse heterogeneous data streams (video, vibration, acoustic emissions, temperature, and gas concentration) into a unified representation [91,92]. Real-time fusion at edge gateways can reconcile temporal misalignments and sensor noise, producing a coherent feature set for downstream analytics. For instance, correlating vibration signatures with high-definition video frames can pinpoint conveyor belt defects more reliably than either source alone. This unified feature space enhances fault diagnosis, reduces false alarms, and accelerates hazard identification, even under harsh field conditions. Knowledge-driven decision engine: Purely data-driven models often lack contextual understanding, hindering their applicability in safety-critical mining operations. Integrating domain knowledge graphs with explainable AI techniques provides a framework for traceable decision logic [93,94]. A mining knowledge graph can encode relationships among geological formations, equipment types, operational procedures, and historical incidents. When coupled with interpretable model outputs (such as feature importance scores or rule-based explanations), operators gain transparent insights into risk predictions and optimization recommendations. This hybrid approach not only improves trust in automated systems but also supports continuous learning, as newly captured data and expert feedback can be incorporated back into the knowledge base to refine future decision making.
5.3. Intelligent Equipment and Automation Construction
The current intelligence of mining equipment is primarily limited to localized perception and remote control, making it difficult to achieve cross-equipment and cross-process collaborative autonomous operations. When production and safety incidents dynamically change, the real-time decision-making and self-stabilization capabilities of the equipment still require significant improvement. There are three future development trends: Autonomous operations with lightweight intelligence: The end-side equipment deploys the algorithm model issued by the cloud, and constantly optimizes the operation strategy and driving path by receiving the instructions sent dynamically by the edge side. Meanwhile, the calculation of the equipment should be lightweight, which focuses on achieving a stable closed loop of “data acquisition and upload, instruction reception and execution”. Flexible and reconfigurable production scheduling: The equipment structure and control logic adopt the concept of Reconfigurable Manufacturing Systems (RMS), enabling the rapid adjustment of operational modes based on orebody parameters and production demands [95,96]. Embedded scheduling agents, aware of each machine’s current configuration and performance envelope, can negotiate task assignments and temporal overlaps, optimizing resource utilization without manual intervention. Self-Protection and Safety Autonomy: In high-risk scenarios, the equipment can utilize multi-sensors for omnidirectional detection, triggering functions such as self-shutdown, protection, or trajectory avoidance. This self-protective logic completes a robust closed loop that not only safeguards equipment integrity but also preserves human lives in rapidly evolving mining contexts.
5.4. Intelligent Mine Talent Cultivation and Team Building
The intelligent mine is an interdisciplinary complex, which puts forward high requirements for the multi-ability of personnel. However, the current workforce is predominantly composed of mining engineering backgrounds, with insufficient representation of information technology and automation professionals. This has led to a pronounced disconnect between humans and machines after the deployment of intelligent systems.
Currently, some universities and mining enterprises have jointly established “collaborative training bases” and “industry–academia–research integrated centers”. These initiatives leverage real mining data to conduct project-based teaching and collaborative research. Some mining companies have developed virtual simulation training platforms, utilizing digital twin environments to provide intelligent mining technology training and operation certification. This approach reduces the implementation time for new technologies and lowers training costs. Moving forward, it is necessary that we further refine the training system for interdisciplinary talent, introduce cross-disciplinary professionals, and enhance intelligent skills development.
6. Conclusions and Prospects
Intelligent mining technology is driving the transformation of traditional mining into a safer, more efficient, and greener intelligent model by integrating information technology, automation technology, and artificial intelligence. This paper provides a comprehensive review of intelligent mining construction from the perspectives of technological development history, current technological status, technological challenges, and future trends.
The technological development process can be summarized into four stages: Stand-alone Automation Stage (1990s–2000s): (1) The core technologies included PLC/DCS systems and localized automated equipment control. The limitations encompassed data silos, poor real-time performance, and weak collaborative capabilities. (2) Integrated Informatization Stage (2000s–2010s): The core technologies involved SCADA centralized monitoring and enterprise-level IT systems (GIS/ERP). The limitations included insufficient system integration and non-uniform protocols. (3) Digitalization and Intelligentization Initiation (2010s–2020s): The core technologies comprised IoT, cloud computing, and autonomous driving. The limitations involved inadequate data compatibility, high capital investment, and insufficient standardization guidance. (4) Comprehensive Intelligentization Stage (2020s–Present): The core technologies feature cloud–edge–device collaboration, AI-driven solutions, and deep integration of 5G/digital twins. Persistent challenges include data heterogeneity, network security vulnerabilities, and talent shortages.
The development status of relevant technologies and applications is summarized from three aspects: the end side, edge side, and cloud side. (1) On the end side, multi-source sensing technology enables the real-time monitoring and precise control of mine environments and equipment status. Intelligent equipment significantly enhances operational mechanization and execution automation. (2) The edge side leverages the network infrastructure to establish localized decision-making hubs. It accomplishes real-time data processing, 3D dynamic modeling, and safety warning tasks. Through protocol adaptation and cluster scheduling optimization, this side improves systematic management and control capabilities. (3) The cloud side constructs an industrial Internet platform. It integrates data middle platforms and digital twin technologies to support the fusion analysis, simulation optimization, and intelligent decision-making of massive heterogeneous data. The three-layer synergy drives the evolution of mining operations toward a data-driven, autonomous, and collaborative intelligent paradigm.
Current intelligent mining technologies face multiple challenges: (1) the insufficient interoperability among heterogeneous systems (lack of standardization and protocol gaps); (2) the limited capability for multi-source heterogeneous data fusion and real-time analysis (bandwidth bottlenecks and computing power allocation conflicts); (3) the weak adaptive decision-making ability of intelligent equipment under complex working conditions (inadequate robustness in extreme environments and insufficient cross-process coordination); and (4) the structural shortage of interdisciplinary and versatile talents.
While this review provides a comprehensive overview of intelligent mining technologies, it has several limitations. First, due to the vast amount of interdisciplinary literature, the review may not fully cover some specialized domains such as AI model robustness under extreme mining conditions or low-level hardware integration challenges. Second, the categorization based on the “cloud–edge–end” framework, though systematic, may oversimplify the dynamic interactions among subsystems. Third, regional policy differences and socio-economic factors influencing intelligent mining deployment were not deeply explored.
The future development trend focuses on coordinated breakthroughs in “technology–ecology–talent”. First, through the construction of a full-stack standard system (such as mine ontology modeling and capability maturity assessment) and a modular open-source ecosystem, it promotes plug-and-play equipment and industrial chain collaboration. Second, by integrating edge federated learning with multimodal knowledge graphs, it achieves cross-domain collaborative analysis and explainable decision-making under data privacy protection. Third, the development of self-learning reconfigurable equipment and cluster intelligent scheduling algorithms enhances autonomous operational resilience in extreme environments. Fourth, relying on virtual simulation platforms and deep industry–academia–research coupling mechanisms, it cultivates interdisciplinary composite talents. Technological advancements will accelerate the transformation of mines into intelligent entities with full-factor interconnectivity, full-process autonomy, and all-scenario resilience, providing a core driving force for green and efficient mining development.
Conceptualization, Z.W. (Zhuo Wang) and L.B.; methodology, Z.W. (Zhuo Wang); validation, Z.W. (Zhuo Wang), J.L., Z.W. (Zhaohao Wu) and Z.Z.; formal analysis, Z.W. (Zhuo Wang) and L.B.; investigation, Z.W. (Zhuo Wang), J.L., Z.W. (Zhaohao Wu) and Z.Z.; resources, L.B.; data curation, Z.W. (Zhuo Wang); writing—original draft preparation, Z.W. (Zhuo Wang); writing—review and editing, Z.W. (Zhuo Wang) and L.B.; visualization, Z.W. (Zhuo Wang); supervision, L.B.; project administration, L.B.; funding acquisition, L.B. All authors have read and agreed to the published version of the manuscript.
No new data were created or analyzed in this study.
We also thank the reviewers for their comments and suggestions to improve the quality of the paper.
The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of the data; in the writing of the manuscript; or in the decision to publish the results.
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Figure 1 Overall development process of intelligent mining construction.
Figure 2 “Cloud–edge–end” architecture and intelligent mining scenarios.
Figure 3 Current development status of end-side technologies.
Figure 4 Current development status of edge-side technologies.
Figure 5 Cloud-side technology construction architecture.
Intelligent construction status of some mines.
| Name | Location | Scale | Intelligent Construction | Role and Impact |
|---|---|---|---|---|
| Dexing Copper Mine [ | Jiangxi, China | Largest open-pit copper mine in Asia | Deployment of 3D digital mining platform; remote drilling and autonomous haulage; intelligent truck dispatch system | Realized integration of geological–mining data, and enhanced resource control precision; drilling/haulage efficiency reached 85% of manual level; enabled large-scale equipment coordination over 15 km2 |
| Pulang Copper Mine [ | Yunnan, China | Super-large underground copper mine | Mechanized mining and unmanned haulage; AI-based video and sound surveillance; full-process 3D visualization | Ensured uninterrupted operation under extreme weather; improved risk detection and early warning capability; strengthened safety transparency and control |
| Fankou Lead-Zinc Mine [ | Guangdong, China | Largest lead-zinc mine in Asia | Construction of high-speed data center; geological and resource digital management; intelligent safety training and risk control system | Enhanced data transmission and emergency response by >30%; improved resource utilization by 3%; established intelligent safety supervision framework |
| Beiya Gold Mine [ | Yunnan, China | Large open-pit gold mine | Mine data center and execution system; full-process digital mining software; integrated safety monitoring system | Enabled collaborative management of mining and processing; realized intelligent geological modeling and production design; achieved real-time hazard monitoring and environmental integration |
| Shangfanggou Molybdenum Mine [ | Henan, China | Large porphyry molybdenum deposit | Visual digital mining platform; truck dispatch and ore allocation system | Constructed a 3D visualized mining model; achieved integrated safety, production, and scheduling management |
| Xingshan Iron Mine [ | Hebei, China | Large underground iron mine | Autonomous underground transport; full automation of fixed systems; visual AI for security monitoring | Achieved unmanned control of transport and crushing; enabled unattended operation of fixed infrastructure; enhanced intrusion detection and safety automation |
| Yimin Coal Mine [ | Inner Mongolia, China | Large open-pit coal mine | Fully unmanned mining face; integrated business control platform | Reached 98.6% operation success rate and >80% manual efficiency; solved data fragmentation and enabled 24 h autonomous operation |
| Tashan Coal Mine [ | Shanxi, China | Large underground coal mine | Multi-system intelligent decision platform; mine-wide data integration center; comprehensive dispatch command center | Integrated ten major systems for real-time decision-making; achieved centralized data governance; enabled remote, unattended, full-process digital scheduling |
1. Chen, L.; Xie, Y.; Wang, Y.; Ge, S.; Wang, F. Sustainable Mining in the Era of Artificial Intelligence. IEEE/CAA J. Autom. Sin.; 2024; 11, pp. 1-4. [DOI: https://dx.doi.org/10.1109/JAS.2023.124182]
2. Cai, M.; Tan, W.; Wu, X.; Zhang, L. Current situation and development strategy of deep intelligent mining in metal mines. Chin. J. Nonferrous Metals; 2021; 31, pp. 3409-3421. [DOI: https://dx.doi.org/10.11817/j.ysxb.1004.0609.2021-42115]
3. Wang, J.; Huang, Z. The Recent Technological Development of Intelligent Mining in China. Engineering; 2017; 3, pp. 439-444. [DOI: https://dx.doi.org/10.1016/J.ENG.2017.04.003]
4. Hu, N.; Li, G. Discussion on the Status and Issues of Intelligence for Metal Mines in China. Met. Mine; 2024; 1, pp. 7-19. [DOI: https://dx.doi.org/10.19614/j.cnki.jsks.202401002]
5. Zhang, K.; Kang, L.; Chen, X.; He, M.; Zhu, C.; Li, D. A Review of Intelligent Unmanned Mining Current Situation and Development Trend. Energies; 2022; 15, 513. [DOI: https://dx.doi.org/10.3390/en15020513]
6. Dong, L.; Wang, J.; Wang, J.; Wang, H. Safe and intelligent mining: Some explorations and challenges in the era of big data. J. Cent. South Univ.; 2023; 30, pp. 1900-1914. [DOI: https://dx.doi.org/10.1007/s11771-023-5350-4]
7. Bing, Z.; Wang, X.; Dong, Z.; Dong, L.; He, T. A novel edge computing architecture for intelligent coal mining system. Wirel. Netw.; 2023; 29, pp. 1545-1554. [DOI: https://dx.doi.org/10.1007/s11276-021-02858-x]
8. Bi, L.; Wang, Z.; Wu, Z.; Zhang, Y. A New Reform of Mining Production and Management Modes under Industry 4.0: Cloud Mining Mode. Appl. Sci.; 2022; 12, 2781. [DOI: https://dx.doi.org/10.3390/app12062781]
9. Li, D.; Peng, S.; Guo, Y.; Lin, P. Development status and prospect of geological guarantee technology for intelligent coal mining in China. Green Smart Min. Eng.; 2024; 1, pp. 433-446. [DOI: https://dx.doi.org/10.1016/j.gsme.2024.12.002]
10. Wang, G.; Wang, H.; Ren, H.; Zhao, G.; Pang, Y.; Du, Y.; Zhang, J.; Hou, G. 2025 scenarios and development path of intelligent coal mine. J. China Coal Soc.; 2018; 43, pp. 295-305. [DOI: https://dx.doi.org/10.13225/j.cnki.jccs.2018.0152]
11. Zhang, S.; Ren, H.; Han, A.; Gong, S. Key technology and development trend of intelligent mining in complex condition working face. J. Mine Autom.; 2022; 48, pp. 16-25. [DOI: https://dx.doi.org/10.13272/j.issn.1671-251x.2021090041]
12. Zheng, X.; Tong, X.; Guo, J.; Zhang, D. Research status and development trend of intelligent monitoring and early warning technology in coal mine. J. Mine Autom.; 2020; 46, pp. 35-40. [DOI: https://dx.doi.org/10.13272/j.issn.1671-251x.17530]
13. Wang, G.; Ren, H.; Zhao, G.; Zhang, D.; Wen, Z.; Meng, L.; Gong, S. Research and practice of intelligent coal mine technology systems in China. Int. J. Coal Sci. Technol.; 2022; 9, 24. [DOI: https://dx.doi.org/10.1007/s40789-022-00491-3]
14. Wang, X.; Yang, J.; Liu, Y.; Wang, Y.; Wang, F.; Kang, M.; Tian, Y.; Rudas, I.; Li, L.; Fanti, M.
15. Wang, H. The 40 years developmental review of the fully mechanized mine roadway heading technology in China. J. China Coal Soc.; 2010; 35, pp. 1815-1820. [DOI: https://dx.doi.org/10.13225/j.cnki.jccs.2010.11.009]
16. Zhang, S. Development of coal mine automation. Ind. Mine Autom.; 2013; 39, pp. 27-33. [DOI: https://dx.doi.org/10.7526/j.issn.1671-251X.2013.02.008]
17. Zhou, B. Development History and Trend Analysis of Coal Mine Safety Monitoring and Control Systems. Min. Saf. Environ. Prot.; 2007; pp. 76-78.
18. Ding, E.; Yu, X.; Xia, B.; Zhao, X.; Zhang, D.; Liu, T.; Wang, W. Development of mine informatization and key technologies of intelligent mines. J. China Coal Soc.; 2022; 47, pp. 564-578. [DOI: https://dx.doi.org/10.13225/j.cnki.jccs.YG21.1930]
19. Cu, Y.; Ren, Y.; Bai, M. Practice of intelligent construction of Shendong Coal Mine. Coal Sci. Technol.; 2022; 50, pp. 218-226.
20. Yuan, L.; Yu, X.; Ding, E.; Zhao, X.; Feng, S.; Zhang, D.; Liu, T.; Wang, W.; Huang, Y. Research on key technologies of human-machine-environment states perception in mine Internet of things. J. Commun.; 2020; 41, pp. 1-12. [DOI: https://dx.doi.org/10.11959/j.issn.1000-436x.2020036]
21. Li, J.; Zhan, K. Intelligent Mining Technology for an Underground Metal Mine Based on Unmanned Equipment. Engineering; 2018; 4, pp. 381-391. [DOI: https://dx.doi.org/10.1016/j.eng.2018.05.013]
22. Li, Y. Building Intelligent New Mines and Setting Industry Benchmarks—The Intelligent Mine Construction Experience of Shendong Jinjie Coal Mine. J. Intell. Mine; 2021; 2, pp. 83-88.
23. Bellamy, D.; Pravica, L. Assessing the impact of driverless haul trucks in Australian surface mining. Resour. Policy; 2011; 36, pp. 149-158. [DOI: https://dx.doi.org/10.1016/j.resourpol.2010.09.002]
24. Liu, K.; Mei, B.; Li, Q.; Sun, S.; Zhang, Q. Collaborative Production Planning Based on an Intelligent Unmanned Mining System for Open-Pit Mines in the Industry 4.0 Era. Machines; 2024; 12, 419. [DOI: https://dx.doi.org/10.3390/machines12060419]
25. Chen, L.; Xie, J.; Zhang, X.; Deng, J.; Ge, S.; Wang, F. Mining 5.0: Concept and Framework for Intelligent Mining Systems in CPSS. IEEE Trans. Intell Veh.; 2023; 8, pp. 3533-3536. [DOI: https://dx.doi.org/10.1109/TIV.2023.3285417]
26. Guidelines for Intelligent Mine Construction in the Nonferrous Metals Industry. China Nonferrous Met. News; 2020; 2, pp. 1-11. [DOI: https://dx.doi.org/10.38265/n.cnki.nysjs.2020.000191]
27. Muduli, L.; Mishra, D.P.; Jana, P.K. Application of wireless sensor network for environmental monitoring in underground coal mines: A systematic review. J. Netw. Comput. Appl.; 2018; 106, pp. 48-67. [DOI: https://dx.doi.org/10.1016/j.jnca.2017.12.022]
28. Brodny, J.; Tutak, M. Applying Sensor-Based Information Systems to Identify Unplanned Downtime in Mining Machinery Operation. Sensors; 2022; 22, 2127. [DOI: https://dx.doi.org/10.3390/s22062127]
29. Xie, B.J.; Li, H.; Luan, Z.; Li, X.; Lei, Z. A lightweight coal mine pedestrian detector for video surveillance systems with multi-level feature fusion and channel pruning. Sci. Rep.; 2025; 15, 5757. [DOI: https://dx.doi.org/10.1038/s41598-025-87157-7]
30. Surucu, O.; Gadsden, S.A.; Yawney, J. Condition Monitoring using Machine Learning: A Review of Theory, Applications, and Recent Advances. Expert Syst. Appl.; 2023; 221, 119738. [DOI: https://dx.doi.org/10.1016/j.eswa.2023.119738]
31. Xu, Z.; Li, J.; Zhang, M. A Surveillance Video Real-Time Analysis System Based on Edge-Cloud and FL-YOLO Cooperation in Coal Mine. IEEE Access; 2021; 9, pp. 68482-68497. [DOI: https://dx.doi.org/10.1109/ACCESS.2021.3077499]
32. Liu, M.; Zhu, Q.; Yin, Y.; Fan, Y.; Su, Z.; Zhang, S. Damage Detection Method of Mining Conveyor Belt Based on Deep Learning. IEEE Sens. J.; 2022; 22, pp. 10870-10879. [DOI: https://dx.doi.org/10.1109/JSEN.2022.3170971]
33. Kang, J.; Li, M.; Mao, S.; Fan, Y.; Wu, Z.; Li, B. A Coal Mine Tunnel Deformation Detection Method Using Point Cloud Data. Sensors; 2024; 24, 2299. [DOI: https://dx.doi.org/10.3390/s24072299]
34. Wang, J.; Wang, L.; Peng, P.; Jiang, Y.; Wu, J.; Liu, Y. Efficient and accurate mapping method of underground metal mines using mobile mining equipment and solid-state lidar. Measurement; 2023; 221, 113581. [DOI: https://dx.doi.org/10.1016/j.measurement.2023.113581]
35. Chen, B.; Zhao, Z.; Bi, L.; Wang, Z. RM2D: An automated and robust laser-based framework for mobile tunnel deformation detection. Undergr. Space; 2025; 20, pp. 241-258. [DOI: https://dx.doi.org/10.1016/j.undsp.2024.07.002]
36. Chen, L.; Li, Y.; Li, L.; Qi, S.; Zhou, J.; Tang, Y.; Yang, J.; Xin, J. High-Precision Positioning, Perception and Safe Navigation for Automated Heavy-Duty Mining Trucks. IEEE Trans. Intell. Veh.; 2024; 9, pp. 4644-4656. [DOI: https://dx.doi.org/10.1109/TIV.2024.3375273]
37. Yang, T.; Wang, S.; Tong, J.; Wang, W. Accurate real-time obstacle detection of coal mine driverless electric locomotive based on ODEL-YOLOv5s. Sci. Rep.; 2023; 13, 17441. [DOI: https://dx.doi.org/10.1038/s41598-023-44746-8]
38. Wang, S.; Wang, D.; Ma, A.; Yan, X.; Zhao, S. Intelligent Robust Control of Roadheader Based on Disturbance Observer. Actuators; 2025; 14, 36. [DOI: https://dx.doi.org/10.3390/act14010036]
39. Fu, T.; Zhang, T.; Cui, Y.; Song, X. Deep Learning-enhanced Intelligent Electrie Shovel Digging Trajectory Tracking Control. J. Mech. Eng.; 2024; 60, pp. 357-366.
40. Wang, J.; Wang, L.; Jiang, Y.; Peng, P.; Wu, J.; Liu, Y. A novel global re-localization method for underground mining vehicles in haulage roadways: A case study of solid-state LiDAR-equipped load-haul-dump vehicles. Tunn. Undergr. Space Technol.; 2025; 156, 106270. [DOI: https://dx.doi.org/10.1016/j.tust.2024.106270]
41. Jiang, Y.; Li, Z.; Yang, G.; Zhang, Y.; Zhang, X. Recent progress on smart mining in China: Unmanned electric locomotive. Adv. Mech. Eng.; 2017; 9, 755450555. [DOI: https://dx.doi.org/10.1177/1687814017695045]
42. Zhao, Z.; Bi, L. Mine Edge Intelligence: A New Trend in the Construction of the Edge Layer for Smart Mines. Gold Sci. Technol.; 2023; 31, pp. 964-977.
43. Singh, A.; Kumar, D.; Hötzel, J. IoT Based information and communication system for enhancing underground mines safety and productivity: Genesis, taxonomy and open issues. Ad Hoc Netw.; 2018; 78, pp. 115-129. [DOI: https://dx.doi.org/10.1016/j.adhoc.2018.06.008]
44. Hasan, S.; Sjöberg, K.; Wallin, P. Wireless Communication in Underground Mining Teleoperation: A Systematic Review. IEEE Access; 2025; 13, pp. 72577-72602. [DOI: https://dx.doi.org/10.1109/ACCESS.2025.3563138]
45. Kubiak, K.; Dec, G.; Stadnicka, D. Possible Applications of Edge Computing in the Manufacturing Industry-Systematic Literature Review. Sensors; 2022; 22, 2445. [DOI: https://dx.doi.org/10.3390/s22072445]
46. Teng, S.; Li, L.; Li, Y.; Hu, X.; Li, L.; Ai, Y.; Chen, L. FusionPlanner: A multi-task motion planner for mining trucks via multi-sensor fusion. Mech. Syst. Signal Practice; 2024; 208, 111051. [DOI: https://dx.doi.org/10.1016/j.ymssp.2023.111051]
47. Imam, M.; Baina, K.; Tabii, Y.; Ressami, E.; Adlaoui, Y.; Benzakour, I.; Abdelwahed, E. The Future of Mine Safety: A Comprehensive Review of Anti-Collision Systems Based on Computer Vision in Underground Mines. Sensors; 2023; 23, 4294. [DOI: https://dx.doi.org/10.3390/s23094294]
48. Imam, M.; Baïna, K.; Tabii, Y.; Ressami, E.; Adlaoui, Y.; Benzakour, I.; Bourzeix, F.; Abdelwahed, E. Ensuring Miners’ Safety in Underground Mines Through Edge Computing: Real-Time PPE Compliance Analysis Based on Pose Estimation. IEEE Access; 2024; 12, pp. 145721-145739. [DOI: https://dx.doi.org/10.1109/ACCESS.2024.3470558]
49. Liu, C.; Su, X.; Li, C. Edge Computing for Data Anomaly Detection of Multi-Sensors in Underground Mining. Electronics; 2021; 10, 302. [DOI: https://dx.doi.org/10.3390/electronics10030302]
50. Hao, Y.; Cao, J.; Wang, Q.; Du, J. Energy-aware scheduling in edge computing with a clustering method. Future Gener. Comput. Syst.; 2021; 117, pp. 259-272. [DOI: https://dx.doi.org/10.1016/j.future.2020.11.029]
51. Zhang, S.; Lu, C.; Jiang, S.; Shan, L.; Xiong, N. An Unmanned Intelligent Transportation Scheduling System for Open-Pit Mine Vehicles Based on 5G and Big Data. IEEE Access; 2020; 8, pp. 135524-135539. [DOI: https://dx.doi.org/10.1109/ACCESS.2020.3011109]
52. Li, J.; Yu, F.R.; Deng, G.; Luo, C.; Ming, Z.; Yan, Q. Industrial Internet: A Survey on the Enabling Technologies, Applications, and Challenges. IEEE Commun. Surv. Tutor.; 2017; 19, pp. 1504-1526. [DOI: https://dx.doi.org/10.1109/COMST.2017.2691349]
53. Xie, J.; Li, S.; Wang, X. A digital smart product service system and a case study of the mining industry: MSPSS. Adv. Eng. Inf.; 2022; 53, 101694. [DOI: https://dx.doi.org/10.1016/j.aei.2022.101694]
54. Chen, L.; Xie, Y.; He, Y.; Ai, Y.; Tian, B.; Li, L.; Ge, S.; Wang, F. Autonomous mining through cooperative driving and operations enabled by parallel intelligence. Commun. Eng.; 2024; 3, 75. [DOI: https://dx.doi.org/10.1038/s44172-024-00220-5]
55. Wang, Y.; Feng, Y.; Xi, C.; Wang, B.; Tang, B.; Geng, Y. Development of an Intelligent Coal Production and Operation Platform Based on a Real-Time Data Warehouse and AI Model. Energies; 2024; 17, 5205. [DOI: https://dx.doi.org/10.3390/en17205205]
56. Zhang, J.; Li, Y.; Wu, J.; Ren, X.; Wang, Y.; Jia, H.; Xue, M. Constructing a Coal Mine Safety Knowledge Graph to Promote the Association and Reuse of Risk Management Empirical Knowledge. Sustainability; 2024; 16, 8848. [DOI: https://dx.doi.org/10.3390/su16208848]
57. Wen, L.; Zhong, D.; Bi, L.; Wang, L.; Liu, Y. Optimization Method of Mine Ventilation Network Regulation Based on Mixed-Integer Nonlinear Programming. Mathematic; 2024; 12, 2632. [DOI: https://dx.doi.org/10.3390/math12172632]
58. Wang, X.; Wang, J.; Ye, R.; Zhou, J.; Bai, D. Exploration and practical application of the digital intelligent mining system at Dexing Copper Mine. Nonferrous Met. Min. Sect.; 2025; 77, pp. 111-118. [DOI: https://dx.doi.org/10.20238/j.issn.1671-4172.2025.03.015]
59. Yu, H.; Zahidi, I.; Liang, D. Spatiotemporal variation of vegetation cover in mining areas of Dexing City, China. Envrion. Res.; 2023; 225, 115634. [DOI: https://dx.doi.org/10.1016/j.envres.2023.115634]
60. Feng, X.; Feng, M.; Zhu, T.; Wu, A.; Yang, C.; Wang, S.; Sui, S. Construction and Application of Debris Flow Comprehensive Monitoring and Early Warning Platform Based on Sky-Ground-Underground Integration Technology. Min. Res. Dev.; 2024; 44, pp. 190-196.
61. Zeng, Q.; Zhu, S.; Li, Z.; Wu, A.; Wang, M.; Su, Y.; Wang, S.; Qu, X.; Feng, M. Research on Real-Time Monitoring and Warning Technology for Multi-Parameter Underground Debris Flow. Sustainability; 2023; 15, 15006. [DOI: https://dx.doi.org/10.3390/su152015006]
62. Ling, W.; Feng, X.; Wang, L.; Zhu, Z.; Wang, S.; Fu, H.; Zhang, S.; Zhao, Y. Prediction method of surface subsidence induced by block caving method based on UAV oblique photogrammetry. Sci. Rep.; 2024; 14, 24630. [DOI: https://dx.doi.org/10.1038/s41598-024-74864-w] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39428408]
63. Liu, X.; Ma, S.; Lou, H.; Yuan, G.; Tan, L.; You, Z.; Wang, R.; Liu, L. Dynamic Management System of Mineral Resources Based on 3D Design and Collaborative Platform. Copp. Eng.; 2025; 3, pp. 83-92.
64. Zhu, S.; Jiang, W.; Yu, X.; Li, J. Study on the Positioning Principle and Application of Microseismic Monitoring System in Fankou Lead-zine Mine. Min. Res. Dev.; 2017; 37, pp. 31-33. [DOI: https://dx.doi.org/10.13827/j.cnki.kyyk.2017.06.008]
65. Ouyang, S.; Shi, Y.; Cao, W. Development of Automatic Monitoring and Pre-warning Systemfor Groundwater in Fankou Lead-zinc Mine. Met. Mine; 2015; 4, pp. 267-272.
66. Li, Y.; Deng, H.; Wen, L.; Qin, Y.; Xu, X. Method for Identifying and Forecasting Mining-Induced Earthquakes Based on Spatiotemporal Characteristics of Microseismic Activities in Fankou Lead/Zinc Mine. Minerals; 2022; 12, 318. [DOI: https://dx.doi.org/10.3390/min12030318]
67. Li, R.; Liu, X.; Ma, S.; Chen, S.; Zhen, Y.; Zhou, Y. Innovative Practice of LlMS Based on the CNAS System Standard. Met. Mine; 2024; 1, pp. 72-79. [DOI: https://dx.doi.org/10.19614/j.cnki.jsks.202401007]
68. Dong, Z.; Zhai, H.; Ji, X.; Yang, Y.; Hu, W.; Sun, C. Thoughts on digital mine construction in Beiya open pit gold mine. Min. Metall.; 2020; 29, pp. 10-15. [DOI: https://dx.doi.org/10.3969/j.issn.1005-7854.2020.06.003]
69. Li, J.; Zhao, Y. Remote monitoring application of semi-continuous waste rock transportation system for large open pit mine based on cloud platform. China Min. Mag.; 2020; 29, pp. 75-80.
70. Xiao, C.; Zhang, D.; Lu, C.; Wang, L. Application of intelligent safety assessment and monitoring technology inShangfanggou Molybdenum Mine. Nonferrous Met. Min. Sect.; 2021; 73, pp. 46-53. [DOI: https://dx.doi.org/10.3969/j.issn.1671-4172.2021.06.008]
71. Feng, J. Development and application of underground mineproduction monitoring system. Nonferrous Met. Min. Sect.; 2015; 67, pp. 107-111. [DOI: https://dx.doi.org/10.3969/j.issn.1671-4172.2015.06.023]
72. Jiang, Z.; Li, N.; Li, Y.; Wu, G.; Liu, X.; Men, D. Research on Information Model Construction Technology of Multi-source Heterogeneous Industrial Data Collaborative Management Based on OPC UA. Mach. Hydraul.; 2024; 52, pp. 8-15.
73. Zhang, B.; Yu, J.; Zhao, Y.; Liu, Q.; Ma, G.; Liu, Y.; Tian, W.; Xian, J.; Cao, Y. Construction in Yimin Open pit Coal Mine Study on key technology and development prospect of intelligent mine. China Coal; 2021; 47, pp. 183-187. [DOI: https://dx.doi.org/10.3969/j.issn.1006-530X.2021.z1.026]
74. Liu, C.; Zhao, J.; Wang, T.; Zhao, Y.; Liu, Y.; Gu, Y.; Wu, T. Research and application of coal mine data center based on cloud platform. Therm. Power Gener.; 2024; 53, pp. 153-161. [DOI: https://dx.doi.org/10.19666/j.rlfd.202308129]
75. Zhang, B.; Liu, Q.; Xu, G.; Cai, R.; Chen, F. Application of GNSS+INS integrated navigation inmine auto-driving trucks. Bull. Surv. Mapp.; 2022; 7, pp. 143-147. [DOI: https://dx.doi.org/10.13474/j.cnki.11-2246.2022.0219]
76. Yan, H.; Zhang, J.; Feng, R.; Wang, W.; Lan, Y.; Xu, Z. Surrounding rock failure analysis of retreating roadways and the control technique for extra-thick coal seams under fully-mechanized top caving and intensive mining conditions: A case study. Tunn. Undergr. Space Technol.; 2020; 97, 103241. [DOI: https://dx.doi.org/10.1016/j.tust.2019.103241]
77. Pang, H.; Wang, S.; Dou, X.; Liu, H.; Chen, X.; Yang, S.; Wang, T.; Wang, S. A Feature Extraction Method Using Auditory Nerve Response for Collapsing Coal-Gangue Recognition. Appl. Sci.; 2020; 10, 7471. [DOI: https://dx.doi.org/10.3390/app10217471]
78. Yu, B.; Zhou, Y.; Xu, G.; Li, Y.; Li, D.; Wang, S.; Kuang, T.; Meng, E. Theory and key technologies for intelligent fully-mechanized top-coal caving faces ofannual production of millions of tons. Coal Sci. Technol.; 2024; 52, pp. 48-67. [DOI: https://dx.doi.org/10.12438/cst.2024-0880]
79. Henriquez, V.; Calvo-Manzano, J.A.; Moreno, A.M.; Feliu, T.S. Agile-CMMI V2.0 alignment: Bringing to light the agile artifacts pointed out by CMMI. Comput. Stand. Interfaces; 2022; 82, 103610. [DOI: https://dx.doi.org/10.1016/j.csi.2021.103610]
80. Vance, D.; Jin, M.; Price, C.; Nimbalkar, S.U.; Wenning, T. Smart manufacturing maturity models and their applicability: A review. J. Manuf. Technol. Manag.; 2023; 34, pp. 735-770. [DOI: https://dx.doi.org/10.1108/JMTM-03-2022-0103]
81. Gökalp, M.O.; Gökalp, E.; Kayabay, K.; Koçyigit, A.; Eren, P.E. Data-driven manufacturing: An assessment model for data science maturity. J. Manuf. Syst.; 2021; 60, pp. 527-546. [DOI: https://dx.doi.org/10.1016/j.jmsy.2021.07.011]
82. Gökalp, E.; Martinez, V. Digital transformation capability maturity model enabling the assessment of industrial manufacturers. Comput. Ind.; 2021; 132, 103522. [DOI: https://dx.doi.org/10.1016/j.compind.2021.103522]
83. Kim, B.C.; Teijgeler, H.; Mun, D.; Han, S. Integration of distributed plant lifecycle data using ISO 15926 and Web services. Ann. Nucl. Energy; 2011; 38, pp. 2309-2318. [DOI: https://dx.doi.org/10.1016/j.anucene.2011.07.033]
84. Jiménez, M.; González, M.; Amaro, A.; Fernández-Renau, A. Field Spectroscopy Metadata System Based on ISO and OGC Standards. ISPRS Int. J. Geo-Inf.; 2014; 3, pp. 1003-1022. [DOI: https://dx.doi.org/10.3390/ijgi3031003]
85. Li, W.; Li, S.; Lin, Z.; Li, Q. Information modeling of mine working based on BIM technology. Tunn. Undergr. Space Technol.; 2021; 115, 103978. [DOI: https://dx.doi.org/10.1016/j.tust.2021.103978]
86. Mittal, S.; Khan, M.A.; Purohit, J.K.; Menon, K.; Romero, D.; Wuest, T. A smart manufacturing adoption framework for SMEs. Int. J. Prod. Res.; 2020; 58, pp. 1555-1573. [DOI: https://dx.doi.org/10.1080/00207543.2019.1661540]
87. Janowicz, K.; Haller, A.; Cox, S.J.D.; Phuoc, D.L.; Lefrançois, M. SOSA: A lightweight ontology for sensors, observations, samples, and actuators. J. Web Semant.; 2019; 56, pp. 1-10. [DOI: https://dx.doi.org/10.1016/j.websem.2018.06.003]
88. Haller, A.; Janowicz, K.; Cox, S.J.D.; Lefrançois, M.; Taylor, K.; Phuoc, D.L.; Lieberman, J.; García-Castro, R.; Atkinson, R.; Stadler, C. The modular SSN ontology: A joint W3C and OGC standard specifying the semantics of sensors, observations, sampling, and actuation. Semant. Web; 2018; 10, pp. 9-32. [DOI: https://dx.doi.org/10.3233/SW-180320]
89. Wang, G.; Xu, Y.; Meng, X.; Fan, J.; Wu, Y.; Ren, H.; Pang, Y.; Du, Y.; Zhao, G.; Li, M. Specification, classification and grading evaluation index forsmart longwall mining face. J. China Coal Soc.; 2020; 45, pp. 3033-3044. [DOI: https://dx.doi.org/10.13225/j.cnki.jccs.2020.1059]
90. Zender, A.; Humm, B.G. Ontology-based Meta AutoML. Integr. Comput.-Aided Eng.; 2022; 29, pp. 351-366. [DOI: https://dx.doi.org/10.3233/ICA-220684]
91. Jiang, W.; Wang, X.; Zhang, S. Integrating multi-modal data into AFSA-LSTM model for real-time oil production prediction. Energy; 2023; 279, 127935. [DOI: https://dx.doi.org/10.1016/j.energy.2023.127935]
92. Feng, D.; Haase-Schutz, C.; Rosenbaum, L.; Hertlein, H.; Glaser, C.; Timm, F.; Wiesbeck, W.; Dietmayer, K. Deep Multi-Modal Object Detection and Semantic Segmentation for Autonomous Driving: Datasets, Methods, and Challenges. IEEE Trans. Intell. Transp.; 2021; 22, pp. 1341-1360. [DOI: https://dx.doi.org/10.1109/TITS.2020.2972974]
93. Xu, X.; Liu, Z.; Liu, W.; Pei, C.; Wu, X.; Nie, Z. A sustainable development benchmarking framework for energy companies based on topic mining and knowledge graph: The case of oil and gas industry. Renew. Sustain. Energy Rev.; 2024; 196, 114350. [DOI: https://dx.doi.org/10.1016/j.rser.2024.114350]
94. Bai, Y.; Wu, J.; Ren, Q.; Jiang, Y.; Cai, J. A BN-based risk assessment model of natural gas pipelines integrating knowledge graph and DEMATEL. Process Saf. Envrion.; 2023; 171, pp. 640-654. [DOI: https://dx.doi.org/10.1016/j.psep.2023.01.060]
95. Seddighi, S. CO2-based reconfigurable enhanced oil recovery using the oxy-fuel-fired boilers: Thermal and flow properties. J. Clean. Prod.; 2017; 162, pp. 1346-1355. [DOI: https://dx.doi.org/10.1016/j.jclepro.2017.06.170]
96. Koren, Y.; Shpitalni, M. Design of reconfigurable manufacturing systems. J. Manuf. Syst.; 2010; 29, pp. 130-141. [DOI: https://dx.doi.org/10.1016/j.jmsy.2011.01.001]
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Bi, Lin 1
; Li, Jinbo 2 ; Wu Zhaohao 1 ; Zhao Ziyu 1
1 School of Resources and Safety Engineering, Central South University, Changsha 410083, China; [email protected] (Z.W.); [email protected] (J.L.); [email protected] (Z.W.); [email protected] (Z.Z.)
2 School of Resources and Safety Engineering, Central South University, Changsha 410083, China; [email protected] (Z.W.); [email protected] (J.L.); [email protected] (Z.W.); [email protected] (Z.Z.), School of Mining Engineering and Geology, Xinjiang Institute of Engineering, Urumqi 830023, China