Swarm and Evolutionary Computation最新文献

{"title":"Differential evolution with bi-strategy co-deployment framework and diversity improvement","authors":"Liangliang Sun ,&nbsp;Zhenghao Song ,&nbsp;Ge Guo ,&nbsp;Yucheng Zhang ,&nbsp;Natalja Matsveichuk ,&nbsp;Yuri Sotskov","doi":"10.1016/j.swevo.2025.102188","DOIUrl":"10.1016/j.swevo.2025.102188","url":null,"abstract":"<div><div>Differential Evolution (DE) has been adopted as the baseline optimizer for problems with continuous search space because of its stable optimization performance and fast convergence speed. When tackling complex optimization problems, DE faces limitations in its non-adaptive form and fails to utilize the potential information of stagnant individuals to improve the search performance. To address these shortcomings, this paper proposes Differential Evolution with Bi-strategy co-deployment framework and Diversity improvement (BDDE) to enhance the search capacity of DE-based variants. First, a bi-strategy co-deployment framework (BCF) is constructed, which combines a probability-based trial vector generation strategy with a parameter adaptation scheme to leverage their respective advantages. Second, a diversity improvement strategy based on gradient descent is proposed, where diversity level and stagnation detection are both measured. For stagnant individuals at excessively low diversity levels, a gradient descent scheme is introduced to update them, guiding individuals to escape local optima and increasing the population diversity. The performance of BDDE is rigorously evaluated on the standard benchmark test suites developed for the 2013, 2014, 2017, and 2022 Congress on Evolutionary Computation (CEC) real-parameter optimization competitions. In addition, the population diversity of BDDE variants is visualized, and an exploration-exploitation analysis of BDDE is conducted to illustrate the effects of its components. Extensive experimental results indicate that BDDE can outperform other advanced algorithms and achieve highly competitive performance for real-world problems.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"99 ","pages":"Article 102188"},"PeriodicalIF":8.5,"publicationDate":"2025-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145321201","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}

{"title":"A parameter-less decoupling strategic adjustments method for shape reconfiguration of UAV swarm","authors":"Ziquan Wang ,&nbsp;Juan Li ,&nbsp;Chang Liu ,&nbsp;Jie Li","doi":"10.1016/j.swevo.2025.102184","DOIUrl":"10.1016/j.swevo.2025.102184","url":null,"abstract":"<div><div>Unmanned aerial vehicle (UAV) swarms need to efficiently reconfigure different 3D shapes to meet complex task requirements, such as those in agriculture, industry, and battlefield environments. This paper presents a novel parameter-less decoupling strategy adjustment method (PD-SAM) to enable 3D shape reconfiguration of fixed-wing UAV swarms. The PD-SAM includes an improved behavior-rule-based decision-making approach that decouples the decision process into two main components: motion trend adjustment and dynamic position adjustment. Unlike traditional behavior-rule-based methods that rely on weighted parameters for combination, the proposed method reduces the number of parameters and eliminates the need for parameter optimization. Additionally, a mapping mechanism is introduced to translate macroscopic shape parameters into microscopic parameters. By mapping preset swarm shape parameters to microscopic rule actions, this mechanism facilitates flexible and spontaneous swarm reconfiguration. This approach is independent of the UAV swarm size and exhibits high scalability, making it suitable for swarm reconfiguration across a wide range of swarm sizes. Finally, with the aim of validating the effectiveness of the proposed method, this study employs flight control models suitable for real-world flight experiments, alongside high-fidelity aircraft dynamics models. The proposed PD-SAM algorithm is tested through software in the loop simulation and compared with a classical behavior-rule-based approach DWAR (Dynamically Weighting Autonomous Rules), as well as a optimization-based method HPSOGA (Hybrid Particle Swarm Optimization and Genetic Algorithm). Simulation results validate that the proposed PD-SAM method reduces the average error by 74.44 % and the average standard deviation of the swarm motion trend by 51.83 % compared to the DWAR method during the swarm shape reconfiguration process. In terms of computational resource consumption, the PD-SAM method is 3 to 4 orders of magnitude lower than the HPSOGA method.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"99 ","pages":"Article 102184"},"PeriodicalIF":8.5,"publicationDate":"2025-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145321250","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}

{"title":"Green flexible job-shop scheduling considering transportation time and machine multi-rotation speeds","authors":"Minghai Yuan ,&nbsp;Zhen Zhang ,&nbsp;Zichen Li ,&nbsp;Yang Ye ,&nbsp;Fengque Pei ,&nbsp;Wenbin Gu","doi":"10.1016/j.swevo.2025.102181","DOIUrl":"10.1016/j.swevo.2025.102181","url":null,"abstract":"<div><div>This paper tackles, for the first time, a Green Flexible Job-Shop Scheduling Problem simultaneously considering Automated Guided Vehicle (AGV) transportation time and multi-rotation-speed machines under multi-objective optimization. Such an integrated setting, rarely addressed in existing studies, poses new challenges due to the tight coupling between transportation, processing speed decisions, and energy–time trade-offs. To solve this complex problem, we propose a novel Dueling Double Deep Q-Network (D3QN)-based scheduling framework with a hierarchical action space. By embedding expert-designed dispatching rules into four sub-decision layers (job, machine, speed, AGV), the framework drastically reduces action dimensionality and improves convergence and generalization. A layered reward mechanism is further designed to balance makespan and energy consumption, while a vectorized state representation enables adaptive decision-making in dynamic environments. Extensive simulations and real-world case studies show that the proposed approach achieves up to 18.6% energy savings and notable scheduling efficiency improvements over benchmark algorithms.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"99 ","pages":"Article 102181"},"PeriodicalIF":8.5,"publicationDate":"2025-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145321196","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}

{"title":"Surrogate-Assisted Niching Differential Evolution for hyperparameter optimization in Convolutional Neural Networks","authors":"Wenhao Du ,&nbsp;Zhigang Ren ,&nbsp;Fan Li ,&nbsp;Yidi Lin","doi":"10.1016/j.swevo.2025.102176","DOIUrl":"10.1016/j.swevo.2025.102176","url":null,"abstract":"<div><div>Hyperparameter optimization in deep Convolutional Neural Networks (CNNs) plays a crucial role in enhancing model performance. However, such problems often exhibit inherent challenges, including high-dimensionality, multimodality, expensive evaluations, and mixed-variable characteristic, which impose high demands on optimization algorithms. To address these challenges, this study proposes a Surrogate-Assisted Niching Differential Evolution (SANDE), which efficiently optimizes CNN architecture hyperparameters through landscape delineation, surrogate-assisted optimization, and computational resource allocation. Specifically, SANDE employs a niching technique that integrates information from both the decision and objective spaces to divide the hyperparameter landscape into multiple simpler and promising sub-regions. These sub-regions are then efficiently searched using a surrogate-assisted integrated differential evolution, where a hybrid differential evolution and a surrogate model serve as the optimizer and objective function, respectively. An information integration strategy is also incorporated to enhance the optimization robustness and convergence. Furthermore, a dynamic resource allocation strategy is introduced to distribute computational resources across sub-regions based on their accumulated historical optimization outcomes, thereby enhancing resource utilization efficiency. The resulting SANDE-CNN is compared with six manually designed CNNs and 23 CNNs obtained by 14 advanced CNN optimization methods. Experimental results demonstrate that SANDE can achieve competitive performance at the cost of limited computational resources.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"99 ","pages":"Article 102176"},"PeriodicalIF":8.5,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266867","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}

{"title":"A matheuristic approach for the mobile parcel locker delivery system with delivery robots and drone resupply","authors":"Cheng Chen ,&nbsp;Emrah Demir ,&nbsp;Wenke Li ,&nbsp;Xisheng Hu ,&nbsp;Hainan Huang ,&nbsp;Jian Li","doi":"10.1016/j.swevo.2025.102182","DOIUrl":"10.1016/j.swevo.2025.102182","url":null,"abstract":"<div><div>Motivated by the rapid advancement of autonomous technologies in urban logistics, this research introduces a novel variant of vehicle routing problem with autonomous resources, including mobile parcel lockers (MPLs), delivery robots and drones. In this problem, customers choose between home delivery and self-pickup from lockers at designated parking areas. Robots are deployed from MPLs which are resupplied by drones as needed. We define this problem as the Mobile Parcel Locker Problem with Delivery Robot and Drone Resupply (MPLPDR-DR). To solve it, we formulate a mixed-integer linear programming (MILP) model and develop a matheuristic approach. This approach integrates a hybrid metaheuristic algorithm for optimizing the routing of MPLs and delivery robots, while a MILP model determines the optimal drone resupply decisions. The hybrid metaheuristic is built on the artificial bee colony framework and integrates a large neighborhood search procedure, a variable neighborhood descent procedure, and a mutation mechanism. The proposed approach also addresses synchronization challenges related to timing in parallel and sequential deliveries. Extensive experiments highlight the algorithm’s effectiveness on large set MPLPDR-DR instances, and the results offer valuable managerial insights.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"99 ","pages":"Article 102182"},"PeriodicalIF":8.5,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266866","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}

{"title":"Scheduling in a three-stage remanufacturing system with machine blockage, deterioration and maintenance using metaheuristic algorithm","authors":"Zihao Luo,&nbsp;Wenyu Zhang,&nbsp;Mengfei Liu","doi":"10.1016/j.swevo.2025.102185","DOIUrl":"10.1016/j.swevo.2025.102185","url":null,"abstract":"<div><div>Remanufacturing plays a significant role in sustainable development. A complete remanufacturing process integrates three stages: disassembly, reprocessing, and reassembly. To bring the problem closer to real-world scenarios, this study proposes a scheduling problem for three-stage remanufacturing system considering machine blockage, deterioration and maintenance. Rate-modifying activity (RMA), as a type of maintenance activity, is executed to address the time-dependent deterioration. To solve this problem, first, a new deterioration model with RMAs is proposed to estimate the actual reprocessing time and determine the strategy for RMA execution. Second, a new blocking scheduling model is established to minimize the makespan. To find satisfactory solutions in a reasonable time, a new metaheuristic called modified monarch butterfly optimization (MMBO) algorithm is proposed. In MMBO algorithm, a problem-specific constructive heuristic and a new machine load balancing strategy are proposed to generate high-quality initial solutions. Then, two improved operators, adaptive to the solution representation scheme, are designed for exploring the solution space. Finally, experiments and comparison with state-of-the-art algorithms are made to demonstrate the effectiveness and superiority of the MMBO algorithm for this problem.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"99 ","pages":"Article 102185"},"PeriodicalIF":8.5,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145267418","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}

{"title":"Improved crayfish optimization algorithm based on multimodal collaborative search and dynamic distribution perturbation for estimating State-of-Health of lithium-ion batteries","authors":"Yilin Yang ,&nbsp;Shuxia Jiang ,&nbsp;Yongjun Zhou ,&nbsp;Hao Xue ,&nbsp;Shuai Yan ,&nbsp;Pengcheng Guo","doi":"10.1016/j.swevo.2025.102178","DOIUrl":"10.1016/j.swevo.2025.102178","url":null,"abstract":"<div><div>The crayfish optimization algorithm (COA) is a novel metaheuristic algorithm. In response to issues such as poor search capability, as well as the tendency to fall into premature convergence when COA solves complex optimization problems, an improved crayfish optimization algorithm based on multimodal collaborative search and dynamic distribution perturbation (MDCOA) is proposed. In MDCOA, a multimodal collaborative search strategy is proposed, which consists of two sub-strategies: dimension learning-based hunting (DLH) search and equilibrium hybrid search (EHS). Firstly, the DLH strategy is utilized to expand the neighborhood of crayfish population, enhancing the crayfish's utilization of neighborhood information. Secondly, the EHS is proposed to balance the intensity of global and local searches, and the global optimal solution is updated by comparing the fitness of DLH and EHS. To avoid premature convergence, dynamic distribution perturbation is proposed to nonlinearly disturb the algorithm. To verify the performance of the MDCOA, the parameter sensitivity of the algorithm and the impact of the two improvement mechanisms are analyzed using the CEC 2020 benchmark suite. Subsequently, MDCOA is compared with 18 other algorithms across multiple dimensions using the CEC 2022 and CEC 2017 benchmark suites. To verify the ability of MDCOA to deal with practical problems, it is used to optimize the hyperparameters of the Transformer-LSTM model for establishing a lithium-ion battery State-of-Health (SOH) estimation model. Simulation results based on actual data demonstrate that the Transformer-LSTM model optimized by MDCOA exhibits high estimation accuracy, with <em>R²</em> values above 97%, <em>RMSE</em> below 0.035, and <em>MAE</em> below 0.02 across four different lithium-ion battery datasets under various operating conditions. Therefore, MDCOA can be used to optimize the hyperparameters of Transformer-LSTM and apply it to lithium-ion batteries SOH estimation. The source code of MDCOA is publicly available on <span><span>https://github.com/yylcsuft/MDCOA</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"99 ","pages":"Article 102178"},"PeriodicalIF":8.5,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145267419","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}

{"title":"Modelling the acoustic behaviour of tricot fabrics using metaheuristic optimization algorithms","authors":"Siddhi Vardhan Singh Rao ,&nbsp;Apurba Das ,&nbsp;Bipin Kumar ,&nbsp;Nandan Kumar","doi":"10.1016/j.swevo.2025.102186","DOIUrl":"10.1016/j.swevo.2025.102186","url":null,"abstract":"<div><div>Tricot fabrics, though widely used in household and automotive sectors, have not been investigated for their acoustic behaviour despite structural advantages. This study examines their sound absorption performance and models it using metaheuristic optimisation algorithms. Experimental results showed underlap pitch as the most influential structural variable, linked to reductions in straight pore fraction and increases in areal density. Acoustic modelling was performed with Maa’s micro-perforated panel, Garai-Pompoli’s equivalent fluid, and Johnson-Champoux-Allard (JCA) microstructural models. Microstructural parameters (tortuosity, shape factor, scale factor, and porosity correction factor) were estimated using particle swarm optimisation (PSO), dynamic multi-swarm PSO (MPSO), artificial bee colony optimisation (ABCO), and fish school search optimisation (FSSO). Among these, JCA-PSO achieved the best agreement with experimental sound absorption coefficient (SAC) data, reducing the root mean square error (RMSE) from 0.053 to 0.045 when effective porosity was introduced. While all algorithms gave comparable SAC predictions, MPSO proved the fastest and most stable, and FSSO the least efficient. Overall, the integration of metaheuristic algorithms with microstructural modelling offers a robust and computationally efficient approach for parameter estimation in fibrous media. The findings highlight tricot fabrics as lightweight, high-performance acoustic materials with potential for automotive and architectural noise-control applications.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"99 ","pages":"Article 102186"},"PeriodicalIF":8.5,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266868","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}

{"title":"A multifactorial evolutionary algorithm to detect stably influential seeds from competitive networks under multiple damage scenarios","authors":"Shuai Wang,&nbsp;Junru Tang,&nbsp;Xiaojun Tan,&nbsp;Mengtang Li","doi":"10.1016/j.swevo.2025.102187","DOIUrl":"10.1016/j.swevo.2025.102187","url":null,"abstract":"<div><div>The complex network has garnered significant attention over the past few decades, with optimization and information extraction problems show significance in practical applications. The competitive influence maximization, along with its robustness, has emerged as a recent focal point, which is aimed at identifying seeds with robust and influential capabilities across multiple propagative groups within a specific network. Existing studies indicate the damage percentage of link-based failures can be pre-defined, and solving the problem in a single-objective manner. However, it has been demonstrated that multiple damage scenarios are prevalent, and the corresponding search processes may yield the synergy. Therefore, the correlation between the optimization directed at different damage scenarios of link-based attacks is analyzed first, which has shown non-conflict relation. Consequently, the multitasking optimization paradigm is thus introduced to modeling the related robust influence maximization problem. A numerical metric is also designed to reflect the significance of links on competitive networks. Equipped with this metric, a multifactorial evolutionary algorithm has been developed to tackle the seed determination problem under multiple damage scenarios, termed MFEA-RCIM_MD. The involved operators consider diverse information from both genetic and fitness domains, and a multi-phase transfer operation is included to leverage knowledge across different tasks. Experiments on synthetic and real-world networks demonstrate the remarkable performance of the algorithm over existing single-objective and multitasking approaches. With enhanced efficiency, multiple candidates are provided for decision-makers to address diffusive challenges in practical systems.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"99 ","pages":"Article 102187"},"PeriodicalIF":8.5,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266869","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}

{"title":"Incorporating exhibitor competitiveness and price sensitivity: A Staged Crossover Hybrid Genetic Algorithm (SCHGA) for optimizing booth pricing","authors":"Chen Yang ,&nbsp;Weijia Liu ,&nbsp;Qiong Chen ,&nbsp;Hailong Wang ,&nbsp;Ben Niu","doi":"10.1016/j.swevo.2025.102180","DOIUrl":"10.1016/j.swevo.2025.102180","url":null,"abstract":"<div><div>Exhibitor participation is crucial to the success of an exhibition. However, a key challenge for organizers is how to meet the diverse demands of enterprises regarding booth pricing and related services. By analyzing real exhibition data, this study explores a method to improve the overall revenue of both exhibitors and organizers. However, previous studies have overlooked three key factors that affect exhibitors’ overall revenue: booth pricing, enterprise competitiveness, and internal enterprise strength. To address this gap, we propose a booth pricing model that incorporates exhibitor competitiveness and price sensitivity, along with multiple constraints, to better accommodate their flexible demands. To effectively solve the proposed mathematical model, this study presents the Staged Crossover Hybrid Genetic Algorithm (SCHGA). The algorithm adopts a fine-grained coordinate-point crossover mechanism, in which crossover is performed at the coordinate-point level. After identifying the better-performing coordinate points in one direction, the algorithm accelerates the matching of better points in the other direction, thereby speeding up convergence and improving solution quality. Experimental results on real exhibition data show that SCHGA outperforms two advanced algorithms and five basic algorithms in terms of convergence quality and stability. Therefore, SCHGA can effectively assist exhibition organizers in booth pricing and allocation decisions.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"99 ","pages":"Article 102180"},"PeriodicalIF":8.5,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145267417","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}