Swarm and Evolutionary Computation最新文献_第3页

Optimizing multi-period multi-mode multi-time window home health care scheduling with an improved tabu search algorithm 基于改进禁忌搜索算法的多周期多模式多时间窗口家庭医疗调度优化

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-26 DOI: 10.1016/j.swevo.2025.102263

Yufeng Zhou, Zimei Pan, Yimeng Zhao, Zhiguo Li

With the rapid growth of online healthcare and increasing demand for personalized medical services, the traditional offline outpatient care model is increasingly unable to meet the diverse needs of various patient groups. To better align with patient preferences and improve care quality, this paper presents a multi-period home health care routing and scheduling problem (HHCRSP), in which patients can select among three service modes: outpatient, door-to-door, and online. The study addresses key challenges in real-world home healthcare delivery, including caregiver-patient matching, time window flexibility, and continuity of care. The objective is to optimize caregiver assignments and scheduling decisions across different service modes while minimizing total costs. We formulate the problem as a mixed-integer nonlinear programming model that captures multiple patient time windows and collaboration between online and offline services. To solve this complex problem efficiently, we propose an improved tabu search (ITS) algorithm. The ITS incorporates a dynamic tabu length mechanism, a novel swap-and-change operator for optimizing patients’ service dates, and a forward start interval algorithm for handling multiple time windows. Numerical experiments demonstrate that ITS outperforms the basic tabu search (TS), competitive simulated annealing (CSA), variable neighborhood search (VNS), and random general variable neighborhood search (RGVNS), achieving average improvements of 21.24 %, 12.28 %, 7.81 %, and 1.76 %, respectively, in solution quality. Sensitivity analyses further reveal that the setting of objective function cost parameters, caregiver-patient skill level deviations, and the number of caregiver workdays significantly impact scheduling performance. The research findings provide valuable decision-making support for healthcare staff scheduling.

随着在线医疗的快速发展和对个性化医疗服务需求的不断增加，传统的线下门诊模式越来越不能满足不同患者群体的多样化需求。为了更好地满足患者的需求，提高护理质量，本文提出了一种多时期家庭健康护理路径与调度问题（HHCRSP），患者可在门诊、上门和在线三种服务模式中进行选择。该研究解决了现实生活中家庭医疗服务的关键挑战，包括护理者与患者的匹配、时间窗口的灵活性和护理的连续性。目标是优化不同服务模式的护理人员分配和调度决策，同时最大限度地降低总成本。我们将该问题表述为一个混合整数非线性规划模型，该模型捕获了多个患者时间窗口以及在线和离线服务之间的协作。为了有效地解决这一复杂问题，我们提出了一种改进的禁忌搜索（ITS）算法。该系统采用了一种动态禁忌长度机制、一种用于优化患者服务日期的新型交换和更改算子，以及一种用于处理多个时间窗口的前向开始间隔算法。数值实验表明，ITS优于基本禁忌搜索（TS）、竞争模拟退火（CSA）、可变邻域搜索（VNS）和随机通用变量邻域搜索（RGVNS），求解质量平均分别提高21.24 %、12.28 %、7.81 %和1.76 %。敏感度分析进一步显示，目标函数成本参数的设定、护理者-患者技能水平偏差和护理人员工作日数显著影响调度绩效。研究结果为医护人员调度提供了有价值的决策支持。

{"title":"Optimizing multi-period multi-mode multi-time window home health care scheduling with an improved tabu search algorithm","authors":"Yufeng Zhou, Zimei Pan, Yimeng Zhao, Zhiguo Li","doi":"10.1016/j.swevo.2025.102263","DOIUrl":"10.1016/j.swevo.2025.102263","url":null,"abstract":"<div><div>With the rapid growth of online healthcare and increasing demand for personalized medical services, the traditional offline outpatient care model is increasingly unable to meet the diverse needs of various patient groups. To better align with patient preferences and improve care quality, this paper presents a multi-period home health care routing and scheduling problem (HHCRSP), in which patients can select among three service modes: outpatient, door-to-door, and online. The study addresses key challenges in real-world home healthcare delivery, including caregiver-patient matching, time window flexibility, and continuity of care. The objective is to optimize caregiver assignments and scheduling decisions across different service modes while minimizing total costs. We formulate the problem as a mixed-integer nonlinear programming model that captures multiple patient time windows and collaboration between online and offline services. To solve this complex problem efficiently, we propose an improved tabu search (ITS) algorithm. The ITS incorporates a dynamic tabu length mechanism, a novel swap-and-change operator for optimizing patients’ service dates, and a forward start interval algorithm for handling multiple time windows. Numerical experiments demonstrate that ITS outperforms the basic tabu search (TS), competitive simulated annealing (CSA), variable neighborhood search (VNS), and random general variable neighborhood search (RGVNS), achieving average improvements of 21.24 %, 12.28 %, 7.81 %, and 1.76 %, respectively, in solution quality. Sensitivity analyses further reveal that the setting of objective function cost parameters, caregiver-patient skill level deviations, and the number of caregiver workdays significantly impact scheduling performance. The research findings provide valuable decision-making support for healthcare staff scheduling.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102263"},"PeriodicalIF":8.5,"publicationDate":"2025-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145839211","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Handling of objectives and constraints with heterogeneous evaluation times for surrogate-assisted evolutionary multi- and many-objective optimization 代理辅助进化多目标和多目标优化中评估时间异构的目标和约束处理

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-26 DOI: 10.1016/j.swevo.2025.102260

Balija Santoshkumar , Kalyanmoy Deb

Utilizing surrogate-based low-fidelity models can significantly decrease the overall computational time required for optimizing problems where evaluating objectives and constraints is time-consuming. Traditionally, surrogate-assisted evolutionary multi- and many-objective optimization algorithms evaluate each new population member with either high- or low-fidelity models across all objectives and constraints. Recent research, however, indicates that mixed-fidelity evaluation, where certain objectives and constraints are assessed with high-fidelity models and others with low-fidelity models can lead to greater efficiency. This improvement arises because time saved from skipping too computationally cheap or expensive evaluations of constraints for largely feasible and infeasible solutions and objectives for infeasible solutions, can instead be allocated to assessing promising candidates near constraint boundaries and in non-dominated feasible regions. In this study, we introduce a mixed-fidelity selection metric that quantifies the potential advantages of evaluating each objective and constraint for each population member individually. This metric incorporates the likelihood of a solution dominating its neighbors, computational cost, surrogate model error, and the extent of constraint violation. We validate our approach with results on test and engineering design problems ranging from two to 10 variables, with up to eight objectives and 10 constraints. We compare our method against state-of-the-art surrogate-assisted evolutionary algorithms. The findings suggest that this approach offers a promising new direction for surrogate-assisted evolutionary multi- and many-objective optimization research.

利用基于代理的低保真度模型可以显著减少优化问题所需的总体计算时间，其中评估目标和约束是耗时的。传统的代理辅助进化多目标和多目标优化算法在所有目标和约束条件下使用高保真度或低保真度模型评估每个新种群成员。然而，最近的研究表明，混合保真度评估，其中某些目标和约束用高保真度模型评估，而其他目标和约束用低保真度模型评估，可以提高效率。这种改进的出现是因为跳过了对大部分可行和不可行的解决方案以及不可行的解决方案的目标进行计算上过于廉价或昂贵的约束评估所节省的时间，可以将其分配给在约束边界附近和非主导可行区域评估有希望的候选对象。在本研究中，我们引入了一个混合保真度选择度量，该度量量化了单独评估每个群体成员的每个目标和约束的潜在优势。这个度量包含了解决方案支配其邻居的可能性、计算成本、代理模型误差和违反约束的程度。我们用测试和工程设计问题的结果来验证我们的方法，这些问题从2到10个变量不等，最多有8个目标和10个约束。我们将我们的方法与最先进的代理辅助进化算法进行比较。研究结果表明，该方法为代理辅助进化多目标优化研究提供了一个有希望的新方向。

{"title":"Handling of objectives and constraints with heterogeneous evaluation times for surrogate-assisted evolutionary multi- and many-objective optimization","authors":"Balija Santoshkumar , Kalyanmoy Deb","doi":"10.1016/j.swevo.2025.102260","DOIUrl":"10.1016/j.swevo.2025.102260","url":null,"abstract":"<div><div>Utilizing surrogate-based low-fidelity models can significantly decrease the overall computational time required for optimizing problems where evaluating objectives and constraints is time-consuming. Traditionally, surrogate-assisted evolutionary multi- and many-objective optimization algorithms evaluate each new population member with either high- or low-fidelity models across all objectives and constraints. Recent research, however, indicates that mixed-fidelity evaluation, where certain objectives and constraints are assessed with high-fidelity models and others with low-fidelity models can lead to greater efficiency. This improvement arises because time saved from skipping too computationally cheap or expensive evaluations of constraints for largely feasible and infeasible solutions and objectives for infeasible solutions, can instead be allocated to assessing promising candidates near constraint boundaries and in non-dominated feasible regions. In this study, we introduce a mixed-fidelity selection metric that quantifies the potential advantages of evaluating each objective and constraint for each population member individually. This metric incorporates the likelihood of a solution dominating its neighbors, computational cost, surrogate model error, and the extent of constraint violation. We validate our approach with results on test and engineering design problems ranging from two to 10 variables, with up to eight objectives and 10 constraints. We compare our method against state-of-the-art surrogate-assisted evolutionary algorithms. The findings suggest that this approach offers a promising new direction for surrogate-assisted evolutionary multi- and many-objective optimization research.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102260"},"PeriodicalIF":8.5,"publicationDate":"2025-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145839210","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-objective bilevel programming model for optimizing network interdiction deployment 网络拦截部署优化的多目标双层规划模型

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-24 DOI: 10.1016/j.swevo.2025.102265

Wei-Chang Yeh , Chyh-Ming Lai , Tsung-Hua Wu

This study introduces a multi-objective bilevel programming model to address the ground force interdiction deployment problem, which is a hierarchical optimization framework where the defender strategically allocates resources to disrupt the attacker’s operational routes under resource constraints. At the upper level, the defender seeks to minimize interdiction costs while maximizing disruption to the attacker’s most reliable and shortest invasion paths. At the lower level, the attacker responds by minimizing the length of its invasion path and maximizing its reliability. To solve this problem, a novel nested multi-objective evolutionary algorithm, termed iNSSSO, is proposed. The algorithm integrates nondominated sorting simplified swarm optimization to optimize the defender's interdiction strategy at the upper level and Bi-Objective A* to solve the attacker’s bi-objective pathfinding problem at the lower level. To further improve solution quality and diversity, the algorithm incorporates dynamic reliability thresholding and min-cut search mechanisms. Experimental validation on 36 test instances demonstrates that iNSSSO consistently outperforms state-of-the-art algorithms, including MOPSDA, NSGA-II, SPEA2, NSGA-III, MOEA/D, and NSSSO, in terms of solution quality, diversity, and convergence. Furthermore, a practical analysis identifies critical network bottlenecks and frequently interdicted edges, offering valuable insights for resource allocation and defensive strategy planning in network interdiction scenarios.

针对地面部队拦截部署问题，提出了一种多目标双层规划模型，该模型是防御方在资源约束下战略性地分配资源以破坏攻击方作战路线的分层优化框架。在上层，防御者寻求最大限度地减少拦截成本，同时最大限度地破坏攻击者最可靠和最短的入侵路径。在较低的级别上，攻击者通过最小化其入侵路径的长度和最大化其可靠性来响应。为了解决这一问题，提出了一种新的嵌套多目标进化算法iNSSSO。该算法集成了非支配排序简化群算法，在上层优化防御者的拦截策略，在下层解决攻击者的双目标寻路问题。为了进一步提高解的质量和多样性，算法引入了动态可靠性阈值和最小切搜索机制。36个测试实例的实验验证表明，iNSSSO在解决质量、多样性和收敛性方面始终优于最先进的算法，包括MOPSDA、NSGA-II、SPEA2、NSGA-III、MOEA/D和NSSSO。此外，实际分析确定了关键的网络瓶颈和经常被拦截的边缘，为网络拦截场景中的资源分配和防御策略规划提供了有价值的见解。

{"title":"Multi-objective bilevel programming model for optimizing network interdiction deployment","authors":"Wei-Chang Yeh , Chyh-Ming Lai , Tsung-Hua Wu","doi":"10.1016/j.swevo.2025.102265","DOIUrl":"10.1016/j.swevo.2025.102265","url":null,"abstract":"<div><div>This study introduces a multi-objective bilevel programming model to address the ground force interdiction deployment problem, which is a hierarchical optimization framework where the defender strategically allocates resources to disrupt the attacker’s operational routes under resource constraints. At the upper level, the defender seeks to minimize interdiction costs while maximizing disruption to the attacker’s most reliable and shortest invasion paths. At the lower level, the attacker responds by minimizing the length of its invasion path and maximizing its reliability. To solve this problem, a novel nested multi-objective evolutionary algorithm, termed iNSSSO, is proposed. The algorithm integrates nondominated sorting simplified swarm optimization to optimize the defender's interdiction strategy at the upper level and Bi-Objective A* to solve the attacker’s bi-objective pathfinding problem at the lower level. To further improve solution quality and diversity, the algorithm incorporates dynamic reliability thresholding and min-cut search mechanisms. Experimental validation on 36 test instances demonstrates that iNSSSO consistently outperforms state-of-the-art algorithms, including MOPSDA, NSGA-II, SPEA2, NSGA-III, MOEA/D, and NSSSO, in terms of solution quality, diversity, and convergence. Furthermore, a practical analysis identifies critical network bottlenecks and frequently interdicted edges, offering valuable insights for resource allocation and defensive strategy planning in network interdiction scenarios.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102265"},"PeriodicalIF":8.5,"publicationDate":"2025-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145839308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Pareto fronts guided co-evolutionary algorithm for constrained multi-objective optimization 约束多目标优化的Pareto前沿引导协同进化算法

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-24 DOI: 10.1016/j.swevo.2025.102266

Hui Wang , Jinyu Xu , Jiayao Qian , Xinyu Zhou , Wei Li , Zhihua Cui , Jia Zhao

Solving constrained multi-objective optimization problems (CMOPs) is a challenging task, because they need to optimize multiple conflicting objectives and satisfy several constraints. In the past several years, though many constrained multi-objective evolutionary algorithms (CMOEAs) have been proposed to deal with CMOPs, they still have some limitations. For instance, different CMOEAs may be required for CMOPs with different characteristics. To tackle this issue, this paper proposes a Pareto fronts (PFs) guided co-evolutionary algorithm (called PFCEA) for CMOPs. Inspired by competitive multitasking (CMT), PFCEA employs two populations with specialized roles. The first population aims to use the unconstrained PF (UPF) and the most useful single-constrained PF (SPF) to find the constrained PF (CPF). To choose the most useful SPF, a novel single constraint priority (SCP) method is designed. The second population considers total constraint violation and objective values by using an improved fuzzy constraint handling technique. During the search, a reward accumulation method is used for the co-evolution of two populations. To verify the performance of PFCEA, four popular benchmark sets and six real-world CMOPs are tested. The performance of PFCEA is compared with eleven other state-of-the-art CMOEAs. Experimental results demonstrate that the proposed PFCEA is competitive in solving CMOPs with different properties.

求解约束多目标优化问题是一项具有挑战性的任务，因为它需要优化多个相互冲突的目标并满足多个约束条件。近年来，虽然提出了许多约束多目标进化算法（cmoea）来处理约束多目标问题，但它们仍然存在一定的局限性。例如，具有不同特性的cmoea可能需要不同的cmoea。为了解决这一问题，本文提出了一种基于Pareto front （PFs）的协同进化算法（PFCEA）。受竞争多任务（CMT）的启发，PFCEA雇佣了两个具有特定角色的群体。第一个种群旨在使用无约束PF （UPF）和最有用的单约束PF （SPF）来找到有约束PF （CPF）。为了选择最有用的SPF，设计了一种新的单约束优先级（SCP）方法。第二种群采用改进的模糊约束处理技术，考虑总约束违反和目标值。在搜索过程中，采用奖励累积法对两个种群进行协同进化。为了验证PFCEA的性能，测试了四个流行的基准集和六个真实的cops。将PFCEA的性能与其他11种最先进的cmoea进行了比较。实验结果表明，所提出的PFCEA在求解不同性质的CMOPs方面具有竞争力。

{"title":"Pareto fronts guided co-evolutionary algorithm for constrained multi-objective optimization","authors":"Hui Wang , Jinyu Xu , Jiayao Qian , Xinyu Zhou , Wei Li , Zhihua Cui , Jia Zhao","doi":"10.1016/j.swevo.2025.102266","DOIUrl":"10.1016/j.swevo.2025.102266","url":null,"abstract":"<div><div>Solving constrained multi-objective optimization problems (CMOPs) is a challenging task, because they need to optimize multiple conflicting objectives and satisfy several constraints. In the past several years, though many constrained multi-objective evolutionary algorithms (CMOEAs) have been proposed to deal with CMOPs, they still have some limitations. For instance, different CMOEAs may be required for CMOPs with different characteristics. To tackle this issue, this paper proposes a Pareto fronts (PFs) guided co-evolutionary algorithm (called PFCEA) for CMOPs. Inspired by competitive multitasking (CMT), PFCEA employs two populations with specialized roles. The first population aims to use the unconstrained PF (UPF) and the most useful single-constrained PF (SPF) to find the constrained PF (CPF). To choose the most useful SPF, a novel single constraint priority (SCP) method is designed. The second population considers total constraint violation and objective values by using an improved fuzzy constraint handling technique. During the search, a reward accumulation method is used for the co-evolution of two populations. To verify the performance of PFCEA, four popular benchmark sets and six real-world CMOPs are tested. The performance of PFCEA is compared with eleven other state-of-the-art CMOEAs. Experimental results demonstrate that the proposed PFCEA is competitive in solving CMOPs with different properties.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102266"},"PeriodicalIF":8.5,"publicationDate":"2025-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145839307","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Real-time optimization of energy consumption and load balancing in server clusters based on model decomposition and a new differential evolution variant 基于模型分解和一种新的差分进化变体的服务器集群能耗与负载平衡实时优化

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-23 DOI: 10.1016/j.swevo.2025.102230

Zhi Xiong , Ziyu Wang , Jianlong Xu , Fei Wang

Adjusting server deployment in a server cluster in real time with a changing workload to balance energy saving and load balancing is an urgent problem. Addressing the limitations of existing research in decision variable reduction and constraint handling, this work proposes an online real-time optimization strategy for energy consumption and load balancing of server clusters based on model decomposition and a new differential evolution (DE) variant. The optimization content includes the on/off state, CPU frequency, and workload allocation of each server. The decision variables are reasonably defined to derive the cluster energy consumption and load balancing models, and the cluster optimization is described as a bi-objective optimization model. Then, based on the model characteristics, the model is decomposed into two layers of optimization to reduce the difficulty of solving the problem, where the outer layer is a bi-objective nonlinear optimization problem and the inner layer is a single-objective mixed-integer linear programming problem. Finally, the inner-layer optimization is solved using the Gurobi optimizer, and the outer-layer optimization is solved using the non-dominated sorting genetic algorithm II and DE algorithm. To address the constraints existing in the outer-layer optimization, a new DE variant, DE/rand/1/while-if-either-or, is proposed. This variant can increase the feasible probability of mutant individuals and reduce the interference with the evolution mechanism, improving the population quality. Tests in various scenarios verify the real-time optimization capability of the proposed strategy and demonstrate the feasibility and effectiveness of the model decomposition and the new DE variant.

在服务器集群中，随着工作负载的变化，实时调整服务器部署，实现节能和负载均衡是一个迫切需要解决的问题。针对现有研究在决策变量约简和约束处理方面的局限性，本文提出了一种基于模型分解和一种新的差分进化（DE）变体的服务器集群能耗和负载平衡在线实时优化策略。优化内容包括每台服务器的开/关状态、CPU频率和工作负载分配。合理定义决策变量，推导出集群能耗和负载均衡模型，并将集群优化描述为双目标优化模型。然后，根据模型特点，将模型分解为两层优化，降低问题的求解难度，其中外层为双目标非线性优化问题，内层为单目标混合整数线性规划问题。最后，采用Gurobi优化器求解内层优化，采用非支配排序遗传算法II和DE算法求解外层优化。为了解决存在于外层优化中的约束，提出了一个新的DE变体DE/rand/1/while-if-非此即彼。这种变异可以增加突变个体的生存概率，减少对进化机制的干扰，提高种群质量。各种场景的测试验证了所提出策略的实时优化能力，并证明了模型分解和新的DE变体的可行性和有效性。

{"title":"Real-time optimization of energy consumption and load balancing in server clusters based on model decomposition and a new differential evolution variant","authors":"Zhi Xiong , Ziyu Wang , Jianlong Xu , Fei Wang","doi":"10.1016/j.swevo.2025.102230","DOIUrl":"10.1016/j.swevo.2025.102230","url":null,"abstract":"<div><div>Adjusting server deployment in a server cluster in real time with a changing workload to balance energy saving and load balancing is an urgent problem. Addressing the limitations of existing research in decision variable reduction and constraint handling, this work proposes an online real-time optimization strategy for energy consumption and load balancing of server clusters based on model decomposition and a new differential evolution (DE) variant. The optimization content includes the on/off state, CPU frequency, and workload allocation of each server. The decision variables are reasonably defined to derive the cluster energy consumption and load balancing models, and the cluster optimization is described as a bi-objective optimization model. Then, based on the model characteristics, the model is decomposed into two layers of optimization to reduce the difficulty of solving the problem, where the outer layer is a bi-objective nonlinear optimization problem and the inner layer is a single-objective mixed-integer linear programming problem. Finally, the inner-layer optimization is solved using the Gurobi optimizer, and the outer-layer optimization is solved using the non-dominated sorting genetic algorithm II and DE algorithm. To address the constraints existing in the outer-layer optimization, a new DE variant, DE/rand/1/while-if-either-or, is proposed. This variant can increase the feasible probability of mutant individuals and reduce the interference with the evolution mechanism, improving the population quality. Tests in various scenarios verify the real-time optimization capability of the proposed strategy and demonstrate the feasibility and effectiveness of the model decomposition and the new DE variant.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102230"},"PeriodicalIF":8.5,"publicationDate":"2025-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145839209","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Enhanced cooperative learning and local search integrated particle swarm optimization for global optimization and coverage enhancement in wireless sensor network 基于增强合作学习和局部搜索的粒子群算法实现无线传感器网络的全局优化和覆盖增强

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-23 DOI: 10.1016/j.swevo.2025.102262

Shubham Gupta, Kuldeep Gautam

Particle Swarm Optimization (PSO) is an efficient and widely used swarm intelligence-based metaheuristic algorithm for global optimization. Its global exploration ability has attracted researchers to solve their complex multimodal problems. However, its shortcomings of slow convergence, stagnation at sub-optimal solutions, and premature convergence cannot be ignored while solving complex and multimodal optimization problems. This paper introduces an enhanced cooperative learning-based PSO, named ECL-PSO, by combining different learning schemes, including Cooperative Comprehensive Learning (CCL), Cooperative Elite Learning (CEL) and Aggregate Learning (AGL) strategies to enhance the velocity rule of the PSO. These schemes are introduced to improve the quality of exploration through cooperation among the personal best information of particles. To avoid premature convergence, the Aggregate Learning combines the new information explored by the CCL and CEL strategies to generate new guiding particles. Further, to achieve exact solutions in the neighborhood of explored regions, we have employed the Sequential Quadratic Programming-based local search scheme. This local search is applied to the personal best positions of the swarm at a later period of the search procedure by assuming that the swarm has sufficiently explored the solution space in the previous stages of optimization. The validation of the ECL-PSO is conducted on the IEEE CEC2017 benchmark set using various performance measures and applied to the coverage optimization problem in a wireless sensor network. The analysis of results confirms the efficacy of all the applied strategies in improving the search quality of the ECL-PSO. The source code of the ECL-PSO can be found in https://github.com/shubh-mnnit93/ECL-PSO-algorithm.git.

粒子群算法（PSO）是一种高效且应用广泛的基于群智能的全局优化元启发式算法。它的全球勘探能力吸引了研究人员来解决复杂的多模态问题。然而，在求解复杂多模态优化问题时，其收敛速度慢、次优解处停滞不前、过早收敛等缺点不可忽视。通过结合合作综合学习（CCL）、合作精英学习（CEL）和聚合学习（AGL）等不同的学习策略，提高了粒子群算法的速度规律，提出了一种增强的基于合作学习的粒子群算法ecl -粒子群算法。引入这些方案是为了通过粒子个人最佳信息之间的合作来提高探测质量。为了避免过早收敛，聚合学习结合了CCL和CEL策略探索的新信息来生成新的引导粒子。此外，为了在探索区域的邻域中获得精确解，我们采用了基于顺序二次规划的局部搜索方案。这种局部搜索在搜索过程的后期应用于群体的个人最优位置，假设群体在之前的优化阶段已经充分探索了解空间。ECL-PSO在IEEE CEC2017基准集上使用各种性能指标进行验证，并应用于无线传感器网络中的覆盖优化问题。结果分析证实了所采用的策略在提高ECL-PSO搜索质量方面的有效性。ECL-PSO的源代码可以在https://github.com/shubh-mnnit93/ECL-PSO-algorithm.git找到。

{"title":"Enhanced cooperative learning and local search integrated particle swarm optimization for global optimization and coverage enhancement in wireless sensor network","authors":"Shubham Gupta, Kuldeep Gautam","doi":"10.1016/j.swevo.2025.102262","DOIUrl":"10.1016/j.swevo.2025.102262","url":null,"abstract":"<div><div>Particle Swarm Optimization (PSO) is an efficient and widely used swarm intelligence-based metaheuristic algorithm for global optimization. Its global exploration ability has attracted researchers to solve their complex multimodal problems. However, its shortcomings of slow convergence, stagnation at sub-optimal solutions, and premature convergence cannot be ignored while solving complex and multimodal optimization problems. This paper introduces an enhanced cooperative learning-based PSO, named ECL-PSO, by combining different learning schemes, including Cooperative Comprehensive Learning (CCL), Cooperative Elite Learning (CEL) and Aggregate Learning (AGL) strategies to enhance the velocity rule of the PSO. These schemes are introduced to improve the quality of exploration through cooperation among the personal best information of particles. To avoid premature convergence, the Aggregate Learning combines the new information explored by the CCL and CEL strategies to generate new guiding particles. Further, to achieve exact solutions in the neighborhood of explored regions, we have employed the Sequential Quadratic Programming-based local search scheme. This local search is applied to the personal best positions of the swarm at a later period of the search procedure by assuming that the swarm has sufficiently explored the solution space in the previous stages of optimization. The validation of the ECL-PSO is conducted on the IEEE CEC2017 benchmark set using various performance measures and applied to the coverage optimization problem in a wireless sensor network. The analysis of results confirms the efficacy of all the applied strategies in improving the search quality of the ECL-PSO. The source code of the ECL-PSO can be found in <span><span>https://github.com/shubh-mnnit93/ECL-PSO-algorithm.git</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102262"},"PeriodicalIF":8.5,"publicationDate":"2025-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145839213","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-agent trajectory prediction with trend-aware attention and mixed graph convolutional networks 基于趋势感知注意力和混合图卷积网络的多智能体轨迹预测

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-20 DOI: 10.1016/j.swevo.2025.102259

Haiyan Tu, Chunli Wang, Jiawei Pan, Xiujuan Zheng

Accurate trajectory prediction is critical in intelligent transportation, autonomous driving, and robot navigation. We present TAA-MGCN, a novel multi-agent trajectory prediction model that combines temporal trend awareness with a hybrid graph attention mechanism. TAA-MGCN employs a temporal trend-aware attention mechanism to capture short-term motion trends and a hybrid graph convolution (integrating feature and topological graphs) to model complex spatial interactions. During graph convolution, a distribution modulation module dynamically adjusts the attention weight distribution. Within an encoder–decoder framework, TAA-MGCN generates multimodal trajectory distributions. Extensive experiments on three benchmark datasets (ETH-UCY, SDD and INTERACTION) demonstrate that TAA-MGCN achieves superior prediction accuracy, surpassing recent advanced models by over 10% on average in both ADE and FDE, highlighting its superior effectiveness and robustness, particularly in complex interaction scenarios.

准确的轨迹预测在智能交通、自动驾驶和机器人导航中至关重要。我们提出了一种新的多智能体轨迹预测模型TAA-MGCN，该模型结合了时间趋势感知和混合图注意机制。TAA-MGCN采用时间趋势感知注意机制来捕捉短期运动趋势，并采用混合图卷积（整合特征图和拓扑图）来模拟复杂的空间相互作用。在图卷积过程中，分布调制模块动态调整注意力权重分布。在编码器-解码器框架内，TAA-MGCN生成多模态轨迹分布。在三个基准数据集（ETH-UCY， SDD和INTERACTION）上进行的大量实验表明，TAA-MGCN具有优越的预测精度，在ADE和FDE方面平均比最近的先进模型高出10%以上，突出了其优越的有效性和鲁棒性，特别是在复杂的交互场景中。

引用次数: 0

Autonomous scheduling method for multi-satellite collaboration with multi-dimensional optimization 多维优化的多卫星协同自主调度方法

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-19 DOI: 10.1016/j.swevo.2025.102255

Xiaoyu Chen , Jiaqi He , Guangming Dai , Maocai Wang , Lei Peng , Tian Tian

Satellite network systems play an important role in the fields of communication, navigation and remote sensing. Aiming at the complex constraints and dynamic dependencies in multi-satellite collaborative mission scheduling, this paper designs a multi-satellite collaborative multi-dimensional optimization mission autonomous scheduling method. Based on the dynamic weighted graph, the weights of nodes and edges are used to reflect the constraints and dependencies between missions and resources, and between missions in the spatiotemporal dimension. The graph neural network and multi-layer attention mechanism are combined to capture the interaction characteristics and to express the complex associations in a refined manner. On this basis, through the real-time update mechanism of the weighted graph and the hierarchical strategy network, the algorithm can dynamically respond to environmental changes at each decision time step, thereby improving the adaptability to dynamic changes and the solution efficiency. Finally, a large number of simulation experiments are conducted to verify the flexibility and scalability of the algorithm in multi-satellite collaborative mission scheduling. The results show that the algorithm designed in this paper can adaptively schedule missions in complex and changing environments. In terms of four key indicators, the total benefit of mission scheduling is increased by an average of 19.2%, the number of successfully scheduled missions is increased by 13.8%, and the smooth control of running time and load balancing of resource utilization are effectively achieved.

卫星网络系统在通信、导航和遥感等领域发挥着重要作用。针对多卫星协同任务调度中存在的复杂约束和动态依赖关系，设计了一种多卫星协同多维优化任务自主调度方法。在动态加权图的基础上，利用节点和边的权重来反映任务与资源之间、任务与任务之间在时空维度上的约束和依赖关系。将图神经网络和多层注意机制相结合，捕捉交互特征，以精细的方式表达复杂的关联。在此基础上，通过加权图和分层策略网络的实时更新机制，算法可以在每个决策时间步动态响应环境变化，从而提高对动态变化的适应能力和求解效率。最后，通过大量仿真实验验证了该算法在多卫星协同任务调度中的灵活性和可扩展性。结果表明，本文所设计的算法能够在复杂多变的环境下自适应调度任务。在4个关键指标方面，任务调度的总效益平均提高19.2%，调度任务成功数平均提高13.8%，有效实现了运行时间的平稳控制和资源利用的负载均衡。

{"title":"Autonomous scheduling method for multi-satellite collaboration with multi-dimensional optimization","authors":"Xiaoyu Chen , Jiaqi He , Guangming Dai , Maocai Wang , Lei Peng , Tian Tian","doi":"10.1016/j.swevo.2025.102255","DOIUrl":"10.1016/j.swevo.2025.102255","url":null,"abstract":"<div><div>Satellite network systems play an important role in the fields of communication, navigation and remote sensing. Aiming at the complex constraints and dynamic dependencies in multi-satellite collaborative mission scheduling, this paper designs a multi-satellite collaborative multi-dimensional optimization mission autonomous scheduling method. Based on the dynamic weighted graph, the weights of nodes and edges are used to reflect the constraints and dependencies between missions and resources, and between missions in the spatiotemporal dimension. The graph neural network and multi-layer attention mechanism are combined to capture the interaction characteristics and to express the complex associations in a refined manner. On this basis, through the real-time update mechanism of the weighted graph and the hierarchical strategy network, the algorithm can dynamically respond to environmental changes at each decision time step, thereby improving the adaptability to dynamic changes and the solution efficiency. Finally, a large number of simulation experiments are conducted to verify the flexibility and scalability of the algorithm in multi-satellite collaborative mission scheduling. The results show that the algorithm designed in this paper can adaptively schedule missions in complex and changing environments. In terms of four key indicators, the total benefit of mission scheduling is increased by an average of 19.2%, the number of successfully scheduled missions is increased by 13.8%, and the smooth control of running time and load balancing of resource utilization are effectively achieved.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102255"},"PeriodicalIF":8.5,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145789845","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Integration of deep reinforcement learning with simulation optimization applied to semiconductor backend assembly scheduling problem 深度强化学习与仿真优化的集成应用于半导体后端装配调度问题

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-19 DOI: 10.1016/j.swevo.2025.102252

Chun-Chih Chiu , Chyh-Ming Lai , Yu-Shun Liao , Wei-Chang Yeh

This study investigates the semiconductor backend assembly scheduling problem, a critical challenge in make-to-order manufacturing where minimizing flow time is essential. The problem features identical and unrelated parallel machines, setup times, and entity transformations (e.g., wafers diced into dies, dies bonded to substrates, substrates packed into magazines). Additional complexities—including batch formation, job splitting, and machine eligibility—further increased the difficulty, rendering it an NP-hard extension of the hybrid flow shop problem. Traditional heuristic algorithms and simulation based optimization often face scalability and efficiency limitations in large-scale applications. To address these challenges, we propose an integrated framework that combines deep reinforcement learning with simulation-based optimization, supported by simplified swarm optimization and an optimal replication allocation strategy. Simplified swarm optimization explores the solution space, while optimal replication allocation directs simulation resources towards promising solutions, enhancing efficiency. Deep reinforcement learning strengthens the framework by dynamically adjusting dispatching rules, accelerating decision-making, and avoiding local optima. Experiments on 18 datasets from a Taiwanese semiconductor factory demonstrated that the proposed approach substantially reduces average flow time while preserving computational efficiency, offering an effective solution to this complex scheduling problem.

本研究探讨了半导体后端组装调度问题，这是一个关键的挑战，在按订单制造中，最小化流程时间是必不可少的。该问题具有相同和不相关的并行机器，设置时间和实体转换（例如，将晶圆切成模具，模具粘合到基板上，基板装入弹匣）。额外的复杂性——包括批生成、作业分割和机器资格——进一步增加了难度，使其成为混合流车间问题的NP-hard扩展。传统的启发式算法和基于仿真的优化在大规模应用中往往面临可扩展性和效率的限制。为了应对这些挑战，我们提出了一个集成框架，将深度强化学习与基于模拟的优化相结合，并以简化的群优化和最佳复制分配策略为支持。简化的群优化探索解空间，而最优复制分配将模拟资源导向有前途的解，提高效率。深度强化学习通过动态调整调度规则、加速决策和避免局部最优来增强框架。在台湾半导体工厂的18个数据集上进行的实验表明，该方法在保持计算效率的同时大大减少了平均流程时间，为这一复杂的调度问题提供了有效的解决方案。

{"title":"Integration of deep reinforcement learning with simulation optimization applied to semiconductor backend assembly scheduling problem","authors":"Chun-Chih Chiu , Chyh-Ming Lai , Yu-Shun Liao , Wei-Chang Yeh","doi":"10.1016/j.swevo.2025.102252","DOIUrl":"10.1016/j.swevo.2025.102252","url":null,"abstract":"<div><div>This study investigates the semiconductor backend assembly scheduling problem, a critical challenge in make-to-order manufacturing where minimizing flow time is essential. The problem features identical and unrelated parallel machines, setup times, and entity transformations (e.g., wafers diced into dies, dies bonded to substrates, substrates packed into magazines). Additional complexities—including batch formation, job splitting, and machine eligibility—further increased the difficulty, rendering it an NP-hard extension of the hybrid flow shop problem. Traditional heuristic algorithms and simulation based optimization often face scalability and efficiency limitations in large-scale applications. To address these challenges, we propose an integrated framework that combines deep reinforcement learning with simulation-based optimization, supported by simplified swarm optimization and an optimal replication allocation strategy. Simplified swarm optimization explores the solution space, while optimal replication allocation directs simulation resources towards promising solutions, enhancing efficiency. Deep reinforcement learning strengthens the framework by dynamically adjusting dispatching rules, accelerating decision-making, and avoiding local optima. Experiments on 18 datasets from a Taiwanese semiconductor factory demonstrated that the proposed approach substantially reduces average flow time while preserving computational efficiency, offering an effective solution to this complex scheduling problem.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102252"},"PeriodicalIF":8.5,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145789842","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multiform optimization with coupled subspace alignment and two-stage adaptive differential evolution for interval constrained scheduling of coal mine integrated energy systems 煤矿综合能源系统区间约束调度的耦合子空间定位和两阶段自适应差分演化多形式优化

IF 8.5 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Swarm and Evolutionary Computation

Pub Date : 2025-12-17 DOI: 10.1016/j.swevo.2025.102256

Canyun Dai , Xiaoyan Sun , Yong Zhang , Dunwei Gong

Coal mine integrated energy system (CMIES) scheduling faces challenges from high dimensionality, strongly coupled constraints, multi-objective demands, and source-load uncertainties. However, existing optimization methods often struggle to balance computational efficiency with comprehensive problem solving in such scenarios. To address this, this paper proposes a multiform optimization algorithm with coupled subspace alignment and two-stage adaptive differential evolution (MFO-CSA-TADE). First, the strong coupling relationship of energy system is decoupled. This decomposes the original problem into two weakly constrained surrogate tasks: an interval-valued electro-thermal uncertainty task and a deterministic electro-cooling task, forming a multiform optimization framework. Second, a knowledge transfer strategy via coupled subspace alignment is developed. Through alignment mapping and cooperative evolution of cross-task coupled electrical variables, efficient knowledge exchange between surrogate tasks is achieved. Third, a two-stage adaptive differential evolution (TADE) strategy is presented for task optimization. The first stage uses diversity-driven TADE to concurrently evolve two surrogate tasks, while the second stage utilizes convergence-driven TADE to focus on the original task’s feasible space. Case studies in a Shanxi mine demonstrate that compared with six state-of-the-art algorithms, the proposed algorithm significantly outperforms others in key metrics such as IGD, HV, IM and T. It not only maintains superior Pareto front quality and uncertainty handling capability but also reduces computation time to 1/6–1/3 of that of comparative algorithms.

煤矿综合能源系统（CMIES）调度面临着高维、强耦合约束、多目标需求和源负荷不确定性的挑战。然而，在这种情况下，现有的优化方法往往难以平衡计算效率和全面的问题解决。为了解决这一问题，本文提出了一种耦合子空间对齐和两阶段自适应差分进化的多形式优化算法（MFO-CSA-TADE）。首先，对能量系统的强耦合关系进行解耦。将原问题分解为两个弱约束代理任务：区间值电热不确定性任务和确定性电冷却任务，形成多形式优化框架。其次，提出了一种基于耦合子空间对齐的知识转移策略。通过跨任务耦合电变量的对齐映射和协同演化，实现了代理任务之间高效的知识交换。第三，提出了一种两阶段自适应差分进化（TADE）策略用于任务优化。第一阶段使用多样性驱动的TADE并行演化两个代理任务，第二阶段使用收敛驱动的TADE关注原始任务的可行空间。通过对山西某矿山的实例研究表明，与现有的6种算法相比，该算法在IGD、HV、IM和t等关键指标上明显优于其他算法，不仅保持了优越的帕累托前质量和不确定性处理能力，而且将计算时间缩短至同类算法的1/6 ~ 1/3。

{"title":"Multiform optimization with coupled subspace alignment and two-stage adaptive differential evolution for interval constrained scheduling of coal mine integrated energy systems","authors":"Canyun Dai , Xiaoyan Sun , Yong Zhang , Dunwei Gong","doi":"10.1016/j.swevo.2025.102256","DOIUrl":"10.1016/j.swevo.2025.102256","url":null,"abstract":"<div><div>Coal mine integrated energy system (CMIES) scheduling faces challenges from high dimensionality, strongly coupled constraints, multi-objective demands, and source-load uncertainties. However, existing optimization methods often struggle to balance computational efficiency with comprehensive problem solving in such scenarios. To address this, this paper proposes a multiform optimization algorithm with coupled subspace alignment and two-stage adaptive differential evolution (MFO-CSA-TADE). First, the strong coupling relationship of energy system is decoupled. This decomposes the original problem into two weakly constrained surrogate tasks: an interval-valued electro-thermal uncertainty task and a deterministic electro-cooling task, forming a multiform optimization framework. Second, a knowledge transfer strategy via coupled subspace alignment is developed. Through alignment mapping and cooperative evolution of cross-task coupled electrical variables, efficient knowledge exchange between surrogate tasks is achieved. Third, a two-stage adaptive differential evolution (TADE) strategy is presented for task optimization. The first stage uses diversity-driven TADE to concurrently evolve two surrogate tasks, while the second stage utilizes convergence-driven TADE to focus on the original task’s feasible space. Case studies in a Shanxi mine demonstrate that compared with six state-of-the-art algorithms, the proposed algorithm significantly outperforms others in key metrics such as IGD, HV, IM and T. It not only maintains superior Pareto front quality and uncertainty handling capability but also reduces computation time to 1/6–1/3 of that of comparative algorithms.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"100 ","pages":"Article 102256"},"PeriodicalIF":8.5,"publicationDate":"2025-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145789843","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0