Search

Dynamic task scheduling problem in cloud manufacturing (CMfg) is always challenging because of changing manufacturing requirements and services. To make instant decisions for task requirements, deep reinforcement learning-based (DRL-based) methods have been broadly applied to learn the scheduling policies of service providers. However, the current DRL-based scheduling methods struggle to fine-tune a pre-trained policy effectively. The resulting training from scratch takes more time and may easily overfit the environment. Additionally, most DRL-based methods with uneven action distribution and inefficient output masks largely reduce the training efficiency, thus degrading the solution quality. To this end, this paper proposes an improved DRL-based approach for dynamic task scheduling in CMfg. First, the paper uncovers the causes behind the inadequate fine-tuning ability and low training efficiency observed in existing DRL-based scheduling methods. Subsequently, a novel approach is proposed to address these issues by updating the scheduling policy while considering the distribution distance between the pre-training dataset and the in-training policy. Uncertainty weights are introduced to the loss function, and the output mask is extended to the updating procedures. Numerical experiments on thirty actual scheduling instances validate that the solution quality and generalization of the proposed approach surpass other DRL-based methods at most by 32.8% and 28.6%, respectively. Additionally, our method can effectively fine-tune a pre-trained scheduling policy, resulting in an average reward increase of up to 23.8%. [ABSTRACT FROM AUTHOR]

In the face of increasingly severe global environmental problems, green transformation of the manufacturing industry has become a key issue. This study aims to explore the interaction between green innovation, intelligent decision-making and risk management in manufacturing industry. The purpose of this study is to explore the relationship between intelligent decision-making, risk management and green innovation, as well as the role and impact of risk management in the process of green innovation in manufacturing companies by building a statistical model. In this study, we investigated the intelligent decision-making problem and the risk management problem in the green transformation process through the literature survey method and statistical analysis. The results of the study show that intelligent decision-making positively affects the green innovation capability of manufacturing enterprises, there is a negative correlation between the complexity of risk management and the green innovation capability of manufacturing enterprises, and when manufacturing enterprises rely more on intelligent decision-making, the impact of risk management (positively or negatively) on green innovation is amplified, indicating an interactive effect between intelligent decision-making and risk management. This study provides valuable insights for the manufacturing industry to enhance smart decision-making in the face of environmental challenges. [ABSTRACT FROM AUTHOR]

Aiming at the problems of low solution accuracy and high decision pressure when facing large-scale dynamic task allocation (DTA) and high-dimensional decision space with single agent, this paper combines the deep reinforcement learning (DRL) theory and an improved Multi-Agent Deep Deterministic Policy Gradient (MADDPG-D2) algorithm with a dual experience replay pool and a dual noise based on multi-agent architecture is proposed to improve the efficiency of DTA. The algorithm is based on the traditional Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm, and considers the introduction of a double noise mechanism to increase the action exploration space in the early stage of the algorithm, and the introduction of a double experience pool to improve the data utilization rate; at the same time, in order to accelerate the training speed and efficiency of the agents, and to solve the cold-start problem of the training, the a priori knowledge technology is applied to the training of the algorithm. Finally, the MADDPG-D2 algorithm is compared and analyzed based on the digital battlefield of ground and air confrontation. The experimental results show that the agents trained by the MADDPG-D2 algorithm have higher win rates and average rewards, can utilize the resources more reasonably, and better solve the problem of the traditional single agent algorithms facing the difficulty of solving the problem in the high-dimensional decision space. The MADDPG-D2 algorithm based on multi-agent architecture proposed in this paper has certain superiority and rationality in DTA. [ABSTRACT FROM AUTHOR]

Intelligent medical industry is in a rapid stage of development around the world, followed by are the expanding market size and basic theories of intelligent medical diagnosis and decision-making. Deep learning models have achieved good practical results in medical domain. However, traditional deep learning is almost calculated and developed by crisp values, while imprecise, uncertain, and vague medical data is common in the process of diagnosis and treatment. It is important and significant to review the contributions of fuzzy deep learning for uncertain medical data, because fuzzy deep learning that originated from fuzzy sets, can effectively deal with uncertain and inaccurate information, providing new viewpoints for alleviating the presence of noise, artifact or high dimensional unstructured information in uncertain medical data. Therefore, taking focus on the intersection of both different fuzzy deep learning models and several types of uncertain medical data, the paper first constructs four types of frameworks of fuzzy deep learning models used for uncertain medical data, and investigates the status from three aspects: fuzzy deep learning models, uncertain medical data and application scenarios. Then the performance evaluation metrics of fuzzy deep learning models are analyzed in details. This work has some original points: (1) four types of frameworks of applying fuzzy deep learning models for uncertain medical data are first proposed. (2) Seven fuzzy deep learning models, five types of uncertain medical data, and five application scenarios are reviewed in details, respectively. (3) The advantages, challenges, and future research directions of fuzzy deep learning for uncertain medical data are critically analyzed, providing valuable suggestions for further deep research. [ABSTRACT FROM AUTHOR]

The ability to make informed decisions in complex scenarios is crucial for intelligent automotive systems. Traditional expert rules and other methods often fall short in complex contexts. Recently, reinforcement learning has garnered significant attention due to its superior decision-making capabilities. However, there exists the phenomenon of inaccurate target network estimation, which limits its decision-making ability in complex scenarios. This paper mainly focuses on the study of the underestimation phenomenon, and proposes an end-to-end autonomous driving decision-making method based on an improved TD3 algorithm. This method employs a forward camera to capture data. By introducing a new critic network to form a triple-critic structure and combining it with the target maximization operation, the underestimation problem in the TD3 algorithm is solved. Subsequently, the multi-timestep averaging method is used to address the policy instability caused by the new single critic. In addition, this paper uses Carla platform to construct multi-vehicle unprotected left turn and congested lane-center driving scenarios and verifies the algorithm. The results demonstrate that our method surpasses baseline DDPG and TD3 algorithms in aspects such as convergence speed, estimation accuracy, and policy stability. [ABSTRACT FROM AUTHOR]

At present, product family design has become an important link in enterprise development and manufacturing. Optimization ideas and technologies are important foundations and core frameworks in product family design. Previous research on product family design has mainly been limited to optimization problems within the product domain. As an important influencing factor in the product family design process, the supply chain not only affects the cost level of the product family in the back-end of the design process, but also affects the modular structure layout of the product family in the front-end of the design process. Therefore, the optimization of the correlation between supply chain and product family design process is a crucial issue that determines the success or failure of product families. However, when researching the personalized needs of users in product family design and configuring product modules, there is very little consideration given to the optimization of supply chain correlation. To address the aforementioned issues, this article develops supply chain oriented product design optimization decision-making method based on improved CUR matrix decomposition. Firstly, based on the customer's functional requirements C matrix and module relationship R matrix, perform customer clustering and corresponding product configuration. Then, utilizing the numerical stability of orthogonal trigonometric decomposition (QR), U matrix is constructed, which represents the inherent relationship between functional requirements and module relationships. Secondly, based on quality/character requirements, functional module levels division and initial supplier configuration are carried out. Finally, determine the supplier configuration for each module with the goal of maximizing total profit. Analyze the customer selection, classification, and product configuration process of a contractor as a case study. The research results indicate that the optimization decision method based on improved CUR matrix decomposition can effectively obtain the optimal solution of the decision problem. [ABSTRACT FROM AUTHOR]

Digital twins (DTs) are an emerging digitalization technology with a huge impact on today’s innovations in both industry and research. DTs can significantly enhance our society and quality of life through the virtualization of a real-world physical system, providing greater insights about their operations and assets, as well as enhancing their resilience through real-time monitoring and proactive maintenance. DTs also pose significant security risks, as intellectual property is encoded and more accessible, as well as their continued synchronization to their physical counterparts. The rapid proliferation and dynamism of cyber threats in today’s digital environments motivate the development of automated and intelligent cyber solutions. Today’s industrial transformation relies heavily on artificial intelligence (AI), including machine learning (ML) and data-driven technologies that allow machines to perform tasks such as self-monitoring, investigation, diagnosis, future prediction, and decision-making intelligently. However, to effectively employ AI-based models in the context of cybersecurity, human-understandable explanations, and their trustworthiness, are significant factors when making decisions in real-world scenarios. This article provides an extensive study of explainable AI (XAI) based cybersecurity modeling through a taxonomy of AI and XAI methods that can assist security analysts and professionals in comprehending system functions, identifying potential threats and anomalies, and ultimately addressing them in DT environments in an intelligent manner. We discuss how these methods can play a key role in solving contemporary cybersecurity issues in various real-world applications. We conclude this paper by identifying crucial challenges and avenues for further research, as well as directions on how professionals and researchers might approach and model future-generation cybersecurity in this emerging field.

With the accelerated evolution of digital technologies such as big data, cloud computing, and artificial intelligence, the trends of intelligence, informatization, and digitization have become imminent across various domains. Deep earth engineering, as a major strategic scientific and technological issue for the nation, is inevitably poised for intelligent upgrades. However, the complex characteristics of “three highs and one disturbance” in deep rock masses pose some severe challenges to the intelligent transformation of deep earth engineering. In order to achieve an efficient integration of deep earth engineering and digital technology, this study focuses on the intelligent practice path based on “perception, transmission, interpretation, analysis, decision”. The representative research progress in intelligent perception, real-time transmission, information interpretation, data analysis, intelligent decision-making and other fields in underground engineering is systematically reviewed, and the targeted development directions for deep earth engineering intelligent development are proposed, including multi perception, fast response, big data, optimal methods, precise models, strong platforms and easy promotion. Research findings indicate that ① The cutting-edge deep earth engineering perception technologies include fiber optic sensors, MEMS sensors, computer vision, automated robots, etc. After data collection is completed, the real-time response of data is achieved through the wireless communication protocols that combine the advantages of simple configuration, strong fault tolerance, and good mobility, in order to achieve an accurate collection and real-time transmission of deep earth engineering monitoring data. ② Deep earth engineering survey technology yields data types such as images, waves, and point clouds. Numerous models for interpreting and analyzing raw data exist. The application of next-generation artificial intelligence technologies, such as artificial neural networks and deep learning, significantly improves the efficiency of interpretation and analysis. ③ The intelligent decision-making system possesses efficient learning capabilities, adapting to the uncertainty in complex environments through cyclic autonomous learning to provide intelligent solutions for decision-making problems. Currently, China’s policies and industrial systems for the intelligent construction in the deep earth engineering have been essentially established, with a plethora of intelligent construction systems already applied in practical settings. Building upon this foundation, the study envisions the development direction of digital deep earth engineering in four aspects: intelligent collection and information interpretation, surrounding rock evaluation and safety assessment, surrounding rock control and dynamic repair, and platform development and application promotion. Furthermore, a comprehensive evaluation and analysis system for surrounding rock stability of deep earth engineering based on multi-source information is constructed.

To study the intelligent decision making methods under limited decision samples, aiming at the problems that operational decision-making experience is difficult to express and the training samples for intelligent decision learning are limited, based on the joint operational simulation and drill environment, a decision-making method based on generative adversarial imitation learning is proposed. This method integrates the operational decision-making experience representation and learning process. On the basis of high-level decision-making and low-level action, rule definitions are used to specify the logic of task execution, and generative adversarial imitation learning algorithms are utilized to improve the generalization ability of intelligent agents in scenarios. This method achieved expected results in the constructed typical adversarial scenarios. The algorithm training converged and the decisions output by the intelligent agent are reasonable. Preliminary experimental results indicate that generative adversarial imitation learning, as an intelligent operational decision-making method, has value for further research.

This paper presents a decision-making approach grounded in rough set theory and evidential reasoning to address the demand for expert decision-making in greenhouse environmental control systems. Furthermore, a decision-making model is developed by integrating the D-S evidence theory with an expert knowledge table for greenhouse environmental control systems. The model's reasoning process encompasses continuous attribute discretization, expert decision table formation, attribute reduction, and evidence combination reasoning. Firstly, the fuzzy C-means clustering algorithm is employed to discretize the original environmental data and cluster it. Subsequently, an attribute reduction algorithm based on information entropy is utilized to optimize the decision table by eliminating unnecessary conditional attributes in expert knowledge. The reduced indicators are then combined using evidential theory. Finally, suitable greenhouse control methods are determined by the confidence decision proposed by the D-S evidence theory. To assess the efficacy of this intelligent decision-making algorithm based on rough set and D-S evidence theory, its performance is compared with traditional SVM algorithms and small-shot learning algorithms. The results indicate that this proposed method significantly enhances the credibility of control decision-making processes, with an average running time of 0.002378s for the fusion decision algorithm and 0.017939s for the support vector machine (SVM) algorithm, respectively. The SVM accuracy rate after testing and training stands at 90.34%. Moreover, retraining based on information entropy attribute reduction leads to a correct decision rate increase of up to 100%. This method notably improves confidence levels in decision-making processes while reducing uncertainty and demonstrates reliability when applied in making decisions regarding greenhouse environments. [ABSTRACT FROM AUTHOR]

To study the intelligent decision making methods under limited decision samples, aiming at the problems that operational decision-making experience is difficult to express and the training samples for intelligent decision learning are limited, based on the joint operational simulation and drill environment, a decision-making method based on generative adversarial imitation learning is proposed. This method integrates the operational decision-making experience representation and learning process. On the basis of high-level decision-making and low-level action, rule definitions are used to specify the logic of task execution, and generative adversarial imitation learning algorithms are utilized to improve the generalization ability of intelligent agents in scenarios. This method achieved expected results in the constructed typical adversarial scenarios. The algorithm training converged and the decisions output by the intelligent agent are reasonable. Preliminary experimental results indicate that generative adversarial imitation learning, as an intelligent operational decision-making method, has value for further research. [ABSTRACT FROM AUTHOR]

This paper reviews the research status of intelligent ship berthing technology at home and abroad in recent years. The hotspots and applications of berthing technology are summarized in the three aspects of berthing mode, mathematical model and control algorithm. The remaining problems in this field are analyzed in terms of autonomy, modeling accuracy, path planning, control algorithm, energy-saving effect and system test. Based on actual navigation needs, it is proposed that the next step is to make breakthroughs in core theories and key technical problems such as information fusion, online modeling, intelligent decision-making, optimization algorithms, green energy conservation and testing technology. The autonomy, robustness, rapidity and "toughness" of berthing technology should be improved so as to realize safe, green and efficient intelligent shipping.

Embodied intelligence stands as a strategic technology in the ongoing scientific and technological revolution, forming a frontier in global competition. The mobile manipulator robot system, with its exceptional mobility, planning, and execution capabilities, has become the preferred hardware carrier for embodied intelligence. Moreover, the mobile manipulator robot system, rooted in embodied intelligence, emerges as a pivotal platform capable of cross-domain functionality. Positioned at the forefront of a new era in information technology and artificial intelligence, this system is integral for future development. Addressing the strategic demand for embodied-intelligence-based mobile manipulator robot systems, this study presents an overview of the current developmental landscape. It delves into the challenges faced by this field, proposing key common technologies such as multimodal perception, world cognition, intelligent autonomous decision-making, and joint planning for movement and manipulation. Furthermore, the study offers recommendations for advancing the field, encompassing national policy support, breakthroughs in common technologies, interdisciplinary collaboration, talent cultivation, and construction of comprehensive verification platforms. These suggestions aim to facilitate the rapid progress of mobile manipulator robots in China amid the wave of embodied intelligence development.

The process steps of intelligent ventilation in mines were proposed, which could make the ventilation system more orderly and controllable. The process of intelligent ventilation in mines was divided into six process sections according to the production process, namely perception monitoring, analysis and diagnosis, intelligent decision-making, scheme approval, remote centralized control and joint control, and execution feedback, which included a total of 24 specific process steps. The input and output elements of each process step and the functional logical relationship between the process steps were clarified. According to the concept of “overall plan of the mine ventilation system+key refinement of the mining air area”, the scheme for implementing the intelligent ventilation process of the entire mine system and two refined intelligent ventilation process application scenarios for the coal mining and excavation working faces were proposed. The intelligent ventilation process steps of the mine were integrated into specific application scenarios. In order to achieve intelligent ventilation application scenarios, the overall architecture of the mine intelligent ventilation system was optimized based on the logical layering concept. The mine intelligent ventilation control platform consisting of hardware driver layer, functional module layer, computing processing layer, data storage layer, and data acquisition layer was planned and designed. The four key process sections of intelligent ventilation involved ten key process sections,which were ventilation perception monitoring, analysis and diagnosis, intelligent decision-making, remote centralized control and joint control. The ten key process steps included monitoring and perception of air volume and wind speed, online monitoring of ventilation resistance, analysis of air volume and pressure in the entire air network, identification and identification of disaster sources, calculation of dynamic air demand in mines, diagnosis of ventilation system faults, decision-making of on-demand air volume control plans, decision-making of emergency air control plans, unmanned remote air control, and unmanned emergency air control. The current technological status of the ten key process steps were summarized and analyzed, and the technical implementation paths for the ten key process steps were proposed. Through iterative upgrades of various intelligent ventilation key technologies, the mine ventilation system was ultimately achieved to be in the stable, reliable, safe, controllable, efficient, energy-saving, and emergency disaster reduction operation state throughout its entire lifecycle.

This study evaluates quality management practices in Industry 4.0 in a specific case of steel manufacturing. We formulate a novel proposal based on Semantic Data Mining techniques a step towards knowledge‐driven decision support based on the industrial Six Sigma approach. In our research we combine machine learning classifiers, and explanation generation algorithms with the Six Sigma practice to automate the evaluation of quality of steel products and determine origins of their defects. We describe our original method, and provide evaluation of the results with real–life data from our industrial partner. [ABSTRACT FROM AUTHOR]

In the current era, the prevalence of common ailments is leading to an increasing number of fatalities. Various infections, viruses, and other pathogens can cause these illnesses. Some illnesses can give rise to tumors that seriously threaten human health. Distinct forms of tumors exist, including benign, premalignant, and malignant, with cancer being present only in malignant forms. Deep learning (DL) algorithms have emerged as one of the most promising methods for detecting cancers within the human body. However, existing models face criticism for their limitations, such as lack of support for large datasets, and reliance on a limited number of attributes from input images. To address these limitations and enable efficient cancer detection throughout the human body, an intelligent decision-making approach model (IDMA) is proposed. The IDMA is combined with the pre-trained VGG19 for improved training. The IDMA analyses convolutional neural network (CNN) layer images for signs of malignancy and rules out false positives. Various performance indicators, like sensitivity, precision, recall, and F1-score, are used to assess the system's performance. The suggested system has been evaluated and proven to outperform similar current systems, achieving an impressive 98.67% accuracy in detecting cancer cells. [ABSTRACT FROM AUTHOR]

Copyright of Coal Science & Technology (0253-2336) is the property of Coal Science & Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The decision-making method of tunnel boring machine (TBM) operating parameters has a significant guiding significance for TBM safe and efficient construction, and it has been one of the TBM tunneling research hotspots. For this purpose, this paper introduces an intelligent decision-making method of TBM operating parameters based on multiple constraints and objective optimization. First, linear cutting tests and numerical simulations are used to investigate the physical rules between different cutting parameters (penetration, cutter spacing, etc.) and rock compressive strength. Second, a dual-driven mapping of rock parameters and TBM operating parameters based on data mining and physical rules of rock breaking is established with high accuracy by combining rock-breaking rules and deep neural networks (DNNs). The decision-making method is established by dual-driven mapping, using the effective rock-breaking capacity and the rated value of mechanical parameters as constraints and the total excavation cost as the optimization objective. The best operational parameters can be obtained by searching for the revolutions per minute and penetration that correspond to the extremum of the constrained objective function. The practicability and effectiveness of the developed decision-making model is verified in the Second Water Source Channel of Hangzhou, China, resulting in the average penetration rate increasing by 11.3% and the total cost decreasing by 10%.

Multiple unmanned aerial vehicle (multi-UAV) confrontation is becoming an increasingly important combat mode in intelligent air combat. The confrontation highly relies on the intelligent collaboration and real-time decision-making of the UAVs. Thus, a decomposed and prioritized experience replay (PER)-based multi-agent deep deterministic policy gradient (DP-MADDPG) algorithm has been proposed in this paper for the moving and attacking decisions of UAVs. Specifically, the confrontation is formulated as a partially observable Markov game. To solve the problem, the DP-MADDPG algorithm is proposed by integrating the decomposed and PER mechanisms into the traditional MADDPG. To overcome the technical challenges of the convergence to a local optimum and a single dominant policy, the decomposed mechanism is applied to modify the MADDPG framework with local and global dual critic networks. Furthermore, to improve the convergence rate of the MADDPG training process, the PER mechanism is utilized to optimize the sampling efficiency from the experience replay buffer. Simulations have been conducted based on the Multi-agent Combat Arena (MaCA) platform, wherein the traditional MADDPG and independent learning DDPG (ILDDPG) algorithms are benchmarks. Simulation results indicate that the proposed DP-MADDPG improves the convergence rate and the convergent reward value. During confrontations against the vanilla distance-prioritized rule-empowered and intelligent ILDDPG-empowered blue parties, the DP-MADDPG-empowered red party can improve the win rate to 96% and 80.5%, respectively.

The ability to make informed decisions in complex scenarios is crucial for intelligent automotive systems. Traditional expert rules and other methods often fall short in complex contexts. Recently, reinforcement learning has garnered significant attention due to its superior decision-making capabilities. However, there exists the phenomenon of inaccurate target network estimation, which limits its decision-making ability in complex scenarios. This paper mainly focuses on the study of the underestimation phenomenon, and proposes an end-to-end autonomous driving decision-making method based on an improved TD3 algorithm. This method employs a forward camera to capture data. By introducing a new critic network to form a triple-critic structure and combining it with the target maximization operation, the underestimation problem in the TD3 algorithm is solved. Subsequently, the multi-timestep averaging method is used to address the policy instability caused by the new single critic. In addition, this paper uses Carla platform to construct multi-vehicle unprotected left turn and congested lane-center driving scenarios and verifies the algorithm. The results demonstrate that our method surpasses baseline DDPG and TD3 algorithms in aspects such as convergence speed, estimation accuracy, and policy stability.

Unmanned aerial vehicle (UAV) swarm confrontation jamming offers a cost-effective and long-range countermeasure against hostile swarms. Intelligent decision-making is a key factor in ensuring its effectiveness. In response to the low-timeliness problem caused by linear programming in current algorithms, this paper proposes an intelligent decision-making algorithm for UAV swarm confrontation jamming based on the multi-agent actor–critic (M2AC) model. First, based on Markov games, an intelligent mathematical decision-making model is constructed to transform the confrontation jamming scenario into a symbolized mathematical problem. Second, the indicator function under this learning paradigm is designed by combining the actor–critic algorithm with Markov games. Finally, by employing a reinforcement learning algorithm with multithreaded parallel training–contrastive execution for solving the model, a Markov perfect equilibrium solution is obtained. The experimental results indicate that the algorithm based on M2AC can achieve faster training and decision-making speeds, while effectively obtaining a Markov perfect equilibrium solution. The training time is reduced to less than 50% compared to the baseline algorithm, with decision times maintained below 0.05 s across all simulation conditions. This helps alleviate the low-timeliness problem of UAV swarm confrontation jamming intelligent decision-making algorithms under highly dynamic real-time conditions, leading to more effective and efficient UAV swarm operations in various jamming and electronic warfare scenarios.

In addressing the morphing problem in vehicle flight, some scholars have primarily employed reinforcement learning methods to make morphing decisions based on task. However, they have not considered the constraints associated with the task process. The innovation of this article is that it proposes an intelligent morphing decision method based on deep neural networks (DNNs) for the autonomous morphing decision problem of hypersonic boost-glide morphing vehicles under process constraints. Firstly, we established a dynamic model of a hypersonic boost-glide morphing vehicle with a continuously variable sweep angle. Then, in order to address the decision optimality problem considering errors and the heat flux density constraint problem during the gliding process, interference was introduced to the datum trajectory in segments. Subsequently, re-optimization was performed to generate a trajectory sample library, which was used to train an intelligent decision-maker using a DNN. The simulation results demonstrated that, compared with the conventional programmatic morphing approach, the intelligent morphing decision maker could dynamically determine the sweep angle based on the current flight state, leading to improved range while still adhering to the heat flux density constraint. This validates the effectiveness and robustness of the proposed intelligent decision-maker. [ABSTRACT FROM AUTHOR]

Aiming at the problems of traditional dynamic weapon-target assignment algorithms in command decisionmaking, such as large computational amount, slow solution speed, and low calculation accuracy, combined with deep reinforcement learning theory, an improved Deep Deterministic Policy Gradient algorithm with dual noise and prioritized experience replay is proposed, which uses a double noise mechanism to expand the search range of the action, and introduces a priority experience playback mechanism to effectively achieve data utilization. Finally, the algorithmis simulated and validated on the ground-to-air countermeasures digital battlefield. The results of the experiment show that, under the framework of the deep neural network for intelligent weapon-target assignment proposed in this paper, compared to the traditional RELU algorithm, the agent trained with reinforcement learning algorithms, such asDeepDeterministic Policy Gradient algorithm, Asynchronous Advantage Actor-Critic algorithm, Deep Q Network algorithm performs better. It shows that the use of deep reinforcement learning algorithms to solve the weapon-target assignment problem in the field of air defense operations is scientific. In contrast to other reinforcement learning algorithms, the agent trained by the improved Deep Deterministic Policy Gradient algorithm has a higher win rate and reward in confrontation, and the use of weapon resources is more efficient. It shows that the model and algorithm have certain superiority and rationality. The results of this paper provide new ideas for solving the problem of weapon-target assignment in air defense combat command decisions. [ABSTRACT FROM AUTHOR]

This study addresses the challenge of intelligent decision‐making for command‐and‐control systems in air defense combat operations. Current autonomous decision‐making systems suffer from limited rationality and insufficient intelligence during operation processes. Recent studies have proposed methods based on deep reinforcement learning (DRL) to address these issues. However, DRL methods typically face challenges related to weak interpretability, lack of convergence guarantees, and high‐computing power requirements. To address these issues, a novel technique for large‐scale air defense decision‐making by combining a DRL technique with game theory is discussed. The proposed method transforms the target assignment problem into a potential game that provides theoretical guarantees for Nash equilibrium (NE) from a distributed perspective. The air‐defense decision problem is decomposed into separate target selection and target assignment problems. A DRL method is used to solve the target selection problem, while the target assignment problem is translated into a target assignment optimization game. This game is proven to be an exact potential game with theoretical convergence guarantees for an NE. Having simulated the proposed decision‐making method using a digital battlefield environment, the effectiveness of the proposed method is demonstrated. [ABSTRACT FROM AUTHOR]

Intelligent decision-making support for multi-behavior collaborative control in coal mine excavation is one of the core functions of the coal mine excavation working face. The optimal time series planning of multi-behavior collaborative control in excavation is the key to intelligent decision-making. In order to solve the problems of single control mode, solidification and poor collaborative operation capability of multi-behavior in coal mine excavation, an intelligent decision-making model of multi-behavior collaborative control in coal mine excavation is designed. It realizes the collaborative operation of multi-behavior in the optimal time series. Firstly, an intelligent decision-making method for excavation multi-behavior collaborative control is proposed. The feasible time series planning set and multi-objective optimal time series planning strategy for excavation multi-behavior are determined. Secondly, based on the regulations and process requirements of the excavation site, a set of excavation action events is determined. By analyzing the time relationship between two action events in the event set, a constraint matrix for the time relationship of excavation multi-behaviors is obtained. Thirdly, based on the transformation method of the time relationship constraint matrix, the multi-behavior time relationship constraint matrix of excavation is transformed into a time relationship constraint matrix. The feasible time series planning set of excavation multi-behavior is obtained. Finally, the solving functions for different excavation objectives are defined and the optimal time series for different excavation objectives is obtained. The experimental results show that the excavation robot can work collaboratively without interference according to the optimal time series planning results of the excavation action determined by the model under different excavation objects. The execution time of one working cycle of the excavation operation is basically consistent with the time calculated by the decision-making model.

The classical TODIM considers the crisp numbers to handle the information. However, in a real-world applicative context, this information is bounded by noise and vagueness and hence uncertain. There are wide range of works in the literature which utilizes fuzzy sets to handle the uncertainty in the various dimensions. However, there is a constraint of hesitancy in such decision-making problems due to the involvement of various decision-makers. Also, in the TODIM method, decision-maker's bounded rationality and psychological behavior are also taken into consideration which adds up the hesitation and considers the problem with higher dimension of uncertainty. There are various applications of fuzzy TODIM using type-2 fuzzy numbers where uncertainty is being handled in more than one dimensions and also the introduction of intuitionistic fuzzy numbers where hesitancy factor of a decision-maker is taken into account. This paper targets to handle the uncertainty in more than one dimension keeping the hesitancy part into consideration for intelligent decision-making. Therefore, a novel trapezoidal type-2 intuitionistic fuzzy set (TrT2 IFS) is proposed, which is an aggregation of several triangular IFSs having upper and lower membership along with a non-membership value. For this TrT2 IFS, we have defined the generation procedure, operations, comparison and distance between such TrT2 IFSs. In addition, the decision-maker weights and criterion weight computation are also presented with respect to the TODIM approach. Furthermore, we have applied this extended TrT2 IFS-based TODIM method into a renewable energy resource selection problem of multi-criteria decision-making. [ABSTRACT FROM AUTHOR]

Due to the rising dependency on digital technology, cybersecurity has emerged as a more prominent field of research and application that typically focuses on securing devices, networks, systems, data and other resources from various cyber‐attacks, threats, risks, damages, or unauthorized access. Artificial intelligence (AI), also referred to as a crucial technology of the current Fourth Industrial Revolution (Industry 4.0 or 4IR), could be the key to intelligently dealing with these cyber issues. Various forms of AI methodologies, such as analytical, functional, interactive, textual as well as visual AI can be employed to get the desired cyber solutions according to their computational capabilities. However, the dynamic nature and complexity of real‐world situations and data gathered from various cyber sources make it challenging nowadays to build an effective AI‐based security model. Moreover, defending robustly against adversarial attacks is still an open question in the area. In this article, we provide a comprehensive view on "Cybersecurity Intelligence and Robustness," emphasizing multi‐aspects AI‐based modeling and adversarial learning that could lead to addressing diverse issues in various cyber applications areas such as detecting malware or intrusions, zero‐day attacks, phishing, data breach, cyberbullying and other cybercrimes. Thus the eventual security modeling process could be automated, intelligent, and robust compared to traditional security systems. We also emphasize and draw attention to the future aspects of cybersecurity intelligence and robustness along with the research direction within the context of our study. Overall, our goal is not only to explore AI‐based modeling and pertinent methodologies but also to focus on the resulting model's applicability for securing our digital systems and society. [ABSTRACT FROM AUTHOR]

From the perspective of historical review, the methodology of economics develops from qualitative to quantitative, from a small sampling of data to a vast amount of data. Because of the superiority in learning inherent law and representative level, deep learning models assist in realizing intelligent decision-making in economics. After presenting some statistical results of relevant researches, this paper systematically investigates deep learning in economics, including a survey of frequently-used deep learning models in economics, several applications of deep learning models used in economics. Then, some critical reviews of deep learning in economics are provided, including models and applications, why and how to implement deep learning in economics, research gap and future challenges, respectively. It is obvious that several deep learning models and their variants have been widely applied in different subfields of economics, e.g., financial economics, macroeconomics and monetary economics, agricultural and natural resource economics, industrial organization, urban, rural, regional, real estate and transportation economics, health, education and welfare, business administration and microeconomics, etc. We are very confident that decision-making in economics will be more intelligent with the development of deep learning, because the research of deep learning in economics has become a hot and important topic recently. [ABSTRACT FROM AUTHOR]

Aiming at the low efficiency and intelligence degree in the parameter setting of computer-aided manufacturing (CAM) numerical control (NC) programming caused by manual work, an intelligent decision-making method of CAM NC programming parameters based on model-based definition (MBD) is proposed. The hierarchical process model uses MBD technology to express the process flow and manage the process information. The tightness between process and NC machining is enhanced by sorting out NC machining steps and extracting their process information automatically. Then a back propagation (BP) neural network is used to mine the experience of NC programming in historical cases. Sparrow search algorithm (SSA) and other methods are used to optimize BP neural network's initial weight and threshold, learning rate, and feedback process. A CAM parameter decision model based on BP neural network is constructed. Taking the connecting rod of a certain type of diesel engine as the experimental object, the method increases programming efficiency by 47%, with an accuracy rate of 80%. It is proved that the intelligent decision-making method proposed in this paper has practicability and feasibility. [ABSTRACT FROM AUTHOR]

This study addresses the challenge of intelligent decision‐making for command‐and‐control systems in air defense combat operations. Current autonomous decision‐making systems suffer from limited rationality and insufficient intelligence during operation processes. Recent studies have proposed methods based on deep reinforcement learning (DRL) to address these issues. However, DRL methods typically face challenges related to weak interpretability, lack of convergence guarantees, and high‐computing power requirements. To address these issues, a novel technique for large‐scale air defense decision‐making by combining a DRL technique with game theory is discussed. The proposed method transforms the target assignment problem into a potential game that provides theoretical guarantees for Nash equilibrium (NE) from a distributed perspective. The air‐defense decision problem is decomposed into separate target selection and target assignment problems. A DRL method is used to solve the target selection problem, while the target assignment problem is translated into a target assignment optimization game. This game is proven to be an exact potential game with theoretical convergence guarantees for an NE. Having simulated the proposed decision‐making method using a digital battlefield environment, the effectiveness of the proposed method is demonstrated.

Copyright of Telecommunication Engineering is the property of Telecommunication Engineering and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

With the development and progress of various sensors, the Internet of Things technology has continued to penetrate into people's lives. Successful application cases of the Internet of Things technology have been seen in various industries. Some scholars have begun to try to apply the Internet of Things technology to the management of various industries. Deep learning technology belongs to the category of artificial intelligence technology. Its main goal is to combine a large number of training samples to train the trained object into an object that can learn sample patterns independently. At present, deep learning has a large number of applications in various intelligent operations and key intelligent recognition fields. Due to the characteristics of its learnable training set, deep learning techniques are often used in the direction of decision analysis. The research adopts a combination of questionnaire survey and modeling analysis to explore the construction of in-depth learning methods for enterprise management intelligent decision-making in the context of the Internet of Things. The research was conducted by telephone interview and online questionnaire, and was supported by nearly 100 technical backbone personnel from 25 large enterprises. The research finds that applying the Internet of Things and deep learning technology to enterprise intelligent management decisions can improve the decision-making accuracy of enterprises, and also provide enterprises with more significant decision-making analysis advantages in the industry competition. This paper first describes the concepts of the Internet of Things, deep learning and intelligent decision-making of enterprise management. Then, in order to sort out the problems existing in the current intelligent decision-making of enterprise management, the author conducted interviews and questionnaires. Through the analysis of the questionnaire data, he found the problems existing in the current intelligent decision-making of enterprise management. Finally, he designed a new strategy of intelligent decision-making of enterprise management based on the Internet of Things and deep learning technology, through the simulation and subsequent investigation and analysis of the newly designed strategy. It is found that the enterprise management intelligent decision-making method designed by combining the Internet of Things and deep learning technology is superior to the traditional enterprise decision-making method in all aspects. The design of the system has been reformed mainly from the aspects of human resource management, digital file management and real-time monitoring of production and operation, so that the popularity of the enterprise's comprehensive information management has reached 84.2%, the efficiency of intelligent decision-making increased by 28.7% on average. The achievements of the research not only prove the role of IoT technology in promoting enterprises' in-depth learning, but also put forward new ideas for enterprises' scientific decision-making. [ABSTRACT FROM AUTHOR]

In view of the problems such as incomplete data collection and implementation layer, non embodiment of edge computing function, and unclear construction standards of integrated prevention and control platform during the construction of intelligent mine fire prevention and control subsystem, an architecture model based on “cloud-edge-end” collaboration was proposed in comparison with relevant provisions of national and local coal mine intelligent construction standards for fire prevention and control. Fire data sensing devices such as optical fiber temperature measurement, MEMS temperature sensors and inspection robot, as well as fire execution control devices such as grouting, nitrogen injection, automatic sprinkler and roadway rapid closure device were added to form the data sensing and execution control layer, which is responsible for sensing fire data information and executing control commands issued by cloud computing center or edge side; Edge computing devices were added to form the edge computing layer to realize the linkage control of on-site fire prevention and extinguishing devices; with the “Intelligent Mine Basic Information Platform” as the middle platform base, intelligent mine fire prevention and control business application was established to form a platform prevention and control layer, expert knowledge base was integrated, and professional analysis on key indicators of fire prevention and control were conducted. A comprehensive prevention and control system for intelligent analysis at the cloud platform layer, nearby analysis and control at the edge layer, and data collection and execution at the perception and execution layer were formed. The opposition and unity relationship among the three layers of “cloud-edge-end” fire prevention and control system were dialectically analyzed. Key devices and technologies such as wireless low-power fire sensor, patrol robot, monitoring point deployment optimization in fire hazard area, edge side supporting physical devices, edge side fire early warning algorithm, decision edge calculation algorithm, edge node deployment strategy, comprehensive prevention and control platform, fire data governance, fire early warning and decision model were discussed.

As a densely populated place, the occurrence of accidents on campus causes serious concerns, while the construction of an intelligent and safe campus is also one of the hotly debated topics. Starting from the perspective of the development process of security events, this paper comprehensively perfects each stage and applies reasonable methods, hoping to reduce the cost of human and material resources in campus security. First, the abnormal behaviour data set and the identification model based on time and space fusion are constructed to identify the abnormal behaviours on campus. Further on, through the agent reward function construction and state transfer function solution and algorithm research, the multi-intelligent-agent tracking and cooperation based on the game idea are realised to provide efficient real-time security alarm and decision suggestion. Finally, the safety resource path planning platform is designed, and the simulation analysis is conducted combined with the traffic network of New Normal University. The results turn out to identify the abnormal behaviour, work out the optimal decision as well as figure out the optimal path of security resources planning and verify the effectiveness of the mathematical model in a simulation manner.

As one of the important way for constructing smart agriculture, unmanned farms are the most attractive in nowadays, and have been explored in many countries. Generally, data, knowledge and intelligent equipment are the core elements of unmanned farms. It deeply integrates modern information technologies such as the Internet of Things, big data, cloud computing, edge computing, and artificial intelligence with agriculture to realize agricultural production information perception, quantitative decision-making, intelligent control, precise input and personalized services. In the paper, the overall technical architecture of unmanned farms is introduced, and five kinds of key technologies of unmanned farms are proposed, which include information perception and intelligent decision-making technology, precision control technology and key equipment for agriculture, automatic driving technology in agriculture, unmanned operation agricultural equipment, management and remote controlling system for unmanned farms. Furthermore, the latest research progress of the above technologies both worldwide are analyzed. Based on which, critical scientific and technological issues to be solved for developing unmanned farms in China are proposed, include unstructured environment perception of farmland, automatic drive for agriculture machinery in complex and changeable farmland environment, autonomous task assignment and path planning of unmanned agricultural machinery, autonomous cooperative operation control of unmanned agricultural machinery group. Those technologies are challenging and absolutely, and would be the most competitive commanding height in the future. The maize unmanned farm constructed in the city of Gongzhuling, Jilin province, China, was also introduced in detail. The unmanned farms is mainly composed of information perception system, unmanned agricultural equipment, management and controlling system. The perception system obtains and provides the farmland information, maize growth, pest and disease information of the farm. The unmanned agricultural machineries could complete the whole process of the maize mechanization under unattended conditions. The management and controlling system includes the basic GIS, remote controlling subsystem, precision operation management subsystem and working display system for unmanned agricultural machineries. The application of the maize unmanned farm has improved maize production efficiency (the harvesting efficiency has been increased by 3－4 times) and reduced labors. Finally, the paper summarizes the important role of the unmanned farm technology were summarized in solving the problems such as reduction of labors, analyzes the opportunities and challenges of developing unmanned farms in China, and put forward the strategic goals and ideas of developing unmanned farm in China.

Copyright of Tunnel Construction / Suidao Jianshe (Zhong-Yingwen Ban) is the property of Tunnel Construction Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Smart Rail Transit is the property of Smart Rail Transit Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

With the development of science and technology, resource consumption has increasingly become a social problem. The circular economy is an economic form of a comprehensive utilization of resources. At present, China's circular economy industry has begun to develop towards intensification, scale and specialization. With the deepening of research on the value of a comprehensive utilization of resources, the traditional economic field has undergone tremendous changes. According to the current situation of circular economy industry, an intelligent decision-making system is designed using artificial intelligence technology. Empirically, using the circular economy model and combining the sequence parameters to determine the sequence parameter indicators of the scientific and technological innovation subsystem and the energy saving and emission reduction subsystem, the correlation analysis of each indicator is carried out, and the order degree of the two subsystems is obtained respectively, so as to obtain for the enterprise the degree of synergy of the evolution system of the circular economy synergy of industries and industrial clusters. After repeated exploratory tests, it is found that the calculation has achieved over 98% of the precision of an astute direction, and has specific use and value because of the enormous number of preparing modes and trial dependability. Making a specific commitment to the circular economy is a trusted and logical solution. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) is one of the most widely used technologies today, and it has a significant effect on our lives in a variety of ways, including social, commercial, and economic aspects. In terms of automation, productivity, and comfort for consumers across a wide range of application areas, from education to smart cities, the present and future IoT technologies hold great promise for improving the overall quality of human life. However, cyber-attacks and threats greatly affect smart applications in the environment of IoT. The traditional IoT security techniques are insufficient with the recent security challenges considering the advanced booming of different kinds of attacks and threats. Utilizing artificial intelligence (AI) expertise, especially machine and deep learning solutions, is the key to delivering a dynamically enhanced and up-to-date security system for the next-generation IoT system. Throughout this article, we present a comprehensive picture on IoT security intelligence, which is built on machine and deep learning technologies that extract insights from raw data to intelligently protect IoT devices against a variety of cyber-attacks. Finally, based on our study, we highlight the associated research issues and future directions within the scope of our study. Overall, this article aspires to serve as a reference point and guide, particularly from a technical standpoint, for cybersecurity experts and researchers working in the context of IoT. [ABSTRACT FROM AUTHOR]

Due to the increasing number of transportation vessels, marine traffic has become more congested. According to the statistics, 89% to 95% of maritime accidents are related to human factors. In order to reduce marine incidents, ship automatic collision avoidance has become one of the most important research issues in the field of ocean engineering. A generalized behavior decision-making (GBDM) model, trained via a reinforcement learning (RL) algorithm, is proposed in this paper, and it can be used for ship autonomous driving in multi-ship encounter situations. Firstly, the obstacle zone by target (OZT) is used to calculate the area of future collisions based on the dynamic information of ships. Meanwhile, a virtual sensor called a grid sensor is taken as the input of the observation state. Then, International Regulations for Preventing Collision at Sea (COLREGs) is introduced into the reward function to make the decision-making fully comply with COLREGs. Different from the previous RL-based collision avoidance model, the interaction between the ship and the environment only works in the collision avoidance decision-making stage. Finally, 60 complex multi-ship encounter scenarios clustered by the COLREGs are taken as the ship's GBDM model training environments. The simulation results show that the proposed GBDM model and training method has flexible scalability in solving the multi-ship collision avoidance problem complying with COLREGs in different scenarios. [ABSTRACT FROM AUTHOR]