Swarm and Evolutionary Computation最新文献

{"title":"Modeling and optimization of a fatigue-aware distributed flexible job shop scheduling problem with dual resources","authors":"Kunpeng Teng ,&nbsp;Chao Lu ,&nbsp;Caozheng Yan","doi":"10.1016/j.swevo.2026.102411","DOIUrl":"10.1016/j.swevo.2026.102411","url":null,"abstract":"<div><div>The rapid development of intelligent manufacturing and globalized production has accelerated the transition from traditional single-workshop scheduling to distributed multi-factory production systems. In such environments, production efficiency depends not only on machine allocation but also on effective coordination of human operators. However, most existing distributed flexible job shop scheduling studies implicitly assume fatigue-free workers, which limits their applicability in real shop-floor environments. This paper investigates a fatigue-aware distributed flexible job shop scheduling problem, in which machines and workers are jointly modeled as dual constrained resources across multiple factories. Worker fatigue evolves dynamically during processing and directly influences scheduling quality. The objective is to simultaneously minimize the maximum worker fatigue level and the makespan. A mixed-integer linear programming formulation is first established to characterize machine assignment, worker assignment, fatigue propagation, and inter-factory job distribution. To efficiently solve large-scale instances, a Fatigue-Aware Hybrid Multi-Objective Optimization algorithm (FAHMO) is proposed. FAHMO incorporates a fatigue-weighted initialization strategy, a decoding-based critical path analysis, and multiple complementary neighborhood search mechanisms, which collaboratively enhance convergence, diversity, and fatigue mitigation. Extensive computational experiments on benchmark instances, together with comparisons against six state-of-the-art and recently published multi-objective evolutionary algorithms, demonstrate the effectiveness and robustness of the proposed approach.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"105 ","pages":"Article 102411"},"PeriodicalIF":8.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147952561","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":"Forecasting continuous optimization algorithm performance improvement within each run","authors":"Peter Korošec","doi":"10.1016/j.swevo.2026.102399","DOIUrl":"10.1016/j.swevo.2026.102399","url":null,"abstract":"<div><div>This research introduces predictive models to forecast continuous optimization algorithm performance improvement during runtime, departing from conventional approaches focusing solely on final performance outcomes. These models leverage sequential, iteration-scaled data, which includes both the solution locations and their corresponding objective values observed throughout the algorithm’s execution. Using this information, the models aim to forecast performance improvements in future iterations, evaluated over a predetermined horizon points. These real-time forecasting capabilities could be utilized in algorithm configuration scenarios to provide additional insights, further enhancing algorithm performance in black-box continuous optimization problems. We evaluate several long short-term memory networks, selected for their effectiveness with sequential data, in two scenarios: development of personalized models using runs from a particular algorithm instance on a single problem instance and development models using runs from a particular algorithm instance on diverse set of instances of optimization problems. Employing the CEC2014 benchmark suite with 10-dimensional unconstrained problem instances and various configurations of Differential Evolution algorithm, our approach demonstrates promising results by exceeding a baseline forecasting model.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"105 ","pages":"Article 102399"},"PeriodicalIF":8.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147805242","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 Hybrid Attraction–Repulsion Optimization Algorithm for Multi-Robot Task Scheduling in Intelligent Greenhouses","authors":"Jiawei Zhao ,&nbsp;Hongbing Li ,&nbsp;Binbin Zhou ,&nbsp;Tingrui Zhang ,&nbsp;Yanting Huang ,&nbsp;Xiaoxing Zhang ,&nbsp;Yongtao Li ,&nbsp;Xiaoyu Ma ,&nbsp;YuQiao Liang","doi":"10.1016/j.swevo.2026.102401","DOIUrl":"10.1016/j.swevo.2026.102401","url":null,"abstract":"<div><div>In intelligent greenhouse environments, multi-robot cooperation over complex road networks and multi-stage operations involves strong coupling among task allocation, operation sequencing, and path selection. Solving these decisions separately may lead to an elongated critical path and inefficient resource utilization. This study addresses the Greenhouse Multi-Robot Task Scheduling (GMRTS) problem by developing a mathematical model aimed at minimizing the system makespan <span><math><msub><mi>C</mi><mrow><mi>m</mi><mi>a</mi><mi>x</mi></mrow></msub></math></span>. To solve this problem, a Hybrid Attraction-Repulsion Optimization Algorithm (HAROA) is proposed, which balances global exploration and local exploitation through a three-phase cooperative search mechanism consisting of the Vortex Diffusion Strategy (VDS), Gravitational Confluence Strategy (GCS), and Turbulent Transition Strategy (TTS). Experimental results on the CEC2017 benchmark suite and intelligent greenhouse scheduling scenarios show that HAROA achieves superior performance over comparative algorithms in terms of solution accuracy and <span><math><msub><mi>C</mi><mrow><mi>m</mi><mi>a</mi><mi>x</mi></mrow></msub></math></span>, together with faster convergence and higher solution stability. Further analyses confirm the effectiveness of the proposed strategies, the transferability of the method, the suitability of the <span><math><msub><mi>C</mi><mrow><mi>m</mi><mi>a</mi><mi>x</mi></mrow></msub></math></span>-oriented formulation, and the applicability of the framework to dynamic task scenarios.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"105 ","pages":"Article 102401"},"PeriodicalIF":8.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147805243","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":"Adaptive swarm intelligence optimization for Unmanned Aerial Vehicle-assisted edge computing","authors":"Fei Teng ,&nbsp;Abdenacer Naouri ,&nbsp;Nabil Abdelkader Nouri ,&nbsp;Osama Abderrahman Gharbi ,&nbsp;Attia Qammar ,&nbsp;Sahraoui Dhelim ,&nbsp;Tianrui Li","doi":"10.1016/j.swevo.2026.102355","DOIUrl":"10.1016/j.swevo.2026.102355","url":null,"abstract":"<div><div>The integration of Unmanned Aerial Vehicles (UAVs) into Mobile Edge Computing (MEC) systems presents a promising paradigm for delivering low-latency, on-demand computational services in dynamic and infrastructure-scarce environments. However, achieving efficient and reliable UAV deployment poses a significant optimization challenge. This challenge necessitates the simultaneous maximization of ground user coverage and robust inter-UAV connectivity, while reducing redundant overlap, which directly minimizes energy consumption and improves overall network efficiency under dynamic operational constraints. To address this complex multi-objective problem, this paper proposes an innovative swarm intelligence-driven optimization framework. The core of this framework is a Gradient-Based Optimization (GBO) algorithm specifically designed for UAV deployment. This algorithm uniquely integrates global exploration capabilities with local refinement mechanisms to navigate the intricate solution space effectively. Furthermore, we introduce a temporal graph modeling approach to capture and predict dynamic UAV–user interactions, enabling adaptive, real-time UAV repositioning in response to changing environmental conditions and user demands. Extensive simulation-based evaluations, conducted within a realistic simulated disaster-affected geographic area, validate the efficacy of our proposed framework. The GBO-driven approach achieves high operational performance: exceeding 85% user coverage, and significantly reducing coverage overlap by 43.7%, with superior convergence characteristics, reaching 95% of its total fitness improvement within just two iterations, indicating extremely fast early-stage convergence. It provides a scalable, adaptive service computing framework for time-critical, resource-constrained UAV-assisted MEC systems.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"105 ","pages":"Article 102355"},"PeriodicalIF":8.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147805245","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":"Adaptive fruit fly optimization-assisted logic-based benders decomposition for distributed parallel precast flowshop scheduling","authors":"Fuli Xiong ,&nbsp;Muming Wu ,&nbsp;Kaihao Zhou","doi":"10.1016/j.swevo.2026.102403","DOIUrl":"10.1016/j.swevo.2026.102403","url":null,"abstract":"<div><div>This paper addresses a distributed precast flowshop scheduling problem that minimizes a weighted combination of delivery time and production–transportation cost in heterogeneous factories with parallel production lines and shared resources. We formulate mixed-integer linear programming (MILP) and constraint programming (CP) models for small-scale instances. For larger instances, we develop two exact decomposition algorithms that exploit the problem’s hierarchical structure. The first algorithm, adaptive fruit fly optimization-assisted logic-based Benders decomposition with scheduling subproblem relaxations (AL_LBBD_SSR), enhances logic-based Benders decomposition with three key strategies: adaptive fruit fly optimization to generate high-quality initial solutions, strong lower bounds to tighten the search space, and subproblem relaxations to accelerate convergence. The second algorithm, branch-and-check with scheduling subproblem relaxations (BCH_SSR), employs a branch-and-check framework strengthened with problem-specific inequalities and effective relaxations. Computational experiments on 180 benchmark instances demonstrate that both algorithms significantly outperform direct MILP and CP approaches, achieving near-optimal solutions with average optimality gaps below 1%. AL_LBBD_SSR excels for large-scale instances with more than 80 orders, while BCH_SSR is more effective for moderate-sized problems with up to 50 orders. These results highlight the potential of combining nature-inspired metaheuristics with exact optimization for complex industrial scheduling problems.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"105 ","pages":"Article 102403"},"PeriodicalIF":8.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147804675","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":"Co-evolution with hierarchical decomposition for Vehicle Routing Problem with Drones","authors":"Ruonan Zhai ,&nbsp;Xuejun Zhang ,&nbsp;Yi Mei ,&nbsp;Tong Guo ,&nbsp;Wenbo Du","doi":"10.1016/j.swevo.2026.102402","DOIUrl":"10.1016/j.swevo.2026.102402","url":null,"abstract":"<div><div>Drone delivery technology has made rapid advances and has seen growing real-world adoption in recent years. To address the inherent limitations of drones in payload and endurance, the collaborative delivery system between trucks and drones, formulated as the Vehicle Routing Problem with Drones (VRP-D), has attracted increasing research interest. In this paper, we introduce a novel Co-evolution with Hierarchical Decomposition (CH-SaBO) framework for VRP-D. Firstly, we propose a co-evolutionary scheme to decompose the original VRP-D into multiple subproblems, each with one truck, effectively reducing the dimensionality of the problem. We then design a hierarchical decomposition strategy for problem decomposition. First, a top-down procedure progressively partitions the solution into finer elements—route, operation, subroute, path, and arc. Then, a recursive bottom-up construction of inter-layer correlation matrices captures the temporal and spatial dependencies among these elements. Finally, k-medoids clustering is applied to the correlation matrices to further group the elements. For subproblem boundaries, we design a simple yet effective overlapping strategy to enhance solution quality. Each subproblem is then independently optimized using a surrogate-assisted bi-level optimizer, enabling efficient search within smaller, well-structured solution spaces. Extensive experiments on 51 benchmark instances demonstrate that CH-SaBO significantly outperforms state-of-the-art algorithms in both solution quality and computational efficiency. Further analyses confirm the scalability of CH-SaBO, as well as the effectiveness of its hierarchical decomposition and overlapping strategies.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"105 ","pages":"Article 102402"},"PeriodicalIF":8.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147805244","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":"Hybrid evolutionary framework with multiple fragmentation strategies and reinforcement learning for multi-objective molecule generation","authors":"Meijia Song ,&nbsp;Lian Shen ,&nbsp;Heming Jia ,&nbsp;Juan Liu ,&nbsp;Xiangrong Liu","doi":"10.1016/j.swevo.2026.102406","DOIUrl":"10.1016/j.swevo.2026.102406","url":null,"abstract":"<div><div>While computational fragment-based drug discovery (FBDD) has made significant strides, many approaches still face challenges including reliance on predefined fragment libraries, limited interpretability in deep generative models, and difficulties in complex multi-property optimization. To address these issues, we propose a novel evolutionary framework. Our method initializes with sample molecules that are fragmented and fed into a policy-decoupled architecture. This architecture employs reinforcement-learning-guided crossover and mutation operators to recombine and modify fragments. This approach expands the latent fragment space without relying on predefined libraries. By employing a grid-based fragment-masked crossover, our method enables combinatorial exploration and extends beyond conventional fragmentation patterns. In comparative experiments, our method outperforms recent state-of-the-art methods on most Practical Molecular Optimization (PMO) benchmarks and target-protein docking tasks. Additionally, it achieves a low average synthetic accessibility (SA) score and maintains a structural novelty rate above 90%.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"105 ","pages":"Article 102406"},"PeriodicalIF":8.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147860013","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":"Dynamic auxiliary reference vector-based many-objective evolutionary algorithm with adaptive multi-population collaboration","authors":"Zhanjie Wang ,&nbsp;Yifei Yao ,&nbsp;Du Cheng ,&nbsp;Renyun Liu ,&nbsp;Xiaojing Feng ,&nbsp;Zhenwei Dong","doi":"10.1016/j.swevo.2026.102389","DOIUrl":"10.1016/j.swevo.2026.102389","url":null,"abstract":"<div><div>Reference-vector-based many-objective evolutionary algorithms exhibit limited performance when addressing problems with irregular Pareto fronts. To overcome this limitation, this paper proposes a dynamic reference vector adjustment strategy that incorporates auxiliary vectors to enhance adaptability to complex Pareto front geometries. This strategy combines uniformly distributed reference vectors with adaptively adjusted auxiliary vectors, which effectively improves the approximation capability for irregular fronts. To further balance convergence and diversity, an adaptive multi-population evolutionary framework is designed, in which subpopulations with different search tendencies are defined and computational resources are dynamically allocated to achieve effective coordination between exploration and exploitation. In addition, a dimension-aware environmental selection mechanism is introduced, which adaptively switches selection strategies according to the number of objectives, thereby enabling more refined control over the trade-off between convergence and diversity. Based on these components, a unified algorithmic framework, termed A-RVEA-LS, is constructed. Comparative experiments on benchmark problems including DTLZ, MaF, and WFG against nine advanced algorithms demonstrate that A-RVEA-LS exhibits significantly superior overall performance and robustness in the majority of test cases.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"105 ","pages":"Article 102389"},"PeriodicalIF":8.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147805183","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 Clustering-Based Dynamic Adaptive Search Strategy based on Competitive Swarm Optimizer for Constrained Multi-Objective Optimization","authors":"Lei Yang ,&nbsp;Haiqin Yao ,&nbsp;Dongbin Zheng ,&nbsp;Daoyuan Zheng ,&nbsp;Kangshun Li ,&nbsp;Ke Li","doi":"10.1016/j.swevo.2026.102409","DOIUrl":"10.1016/j.swevo.2026.102409","url":null,"abstract":"<div><div>In recent years, numerous algorithms for constrained multi-objective optimization based on the Competitive Swarm Optimizer (CSO) have demonstrated remarkably rapid convergence. However, after the swarm converges to the Constrained Pareto Front (CPF), these algorithms often encounter significant challenges in effective local exploitation. To enhance fine-grained search performance in the vicinity of the CPF, this paper proposes a clustering-based dynamic adaptive search strategy. The proposed strategy partitions the swarm into multiple sub-swarms via a clustering process, with the number of clusters adaptively adjusted throughout evolution. It dynamically modulates the swarm’s search dynamics — switching between cohesive global convergence and partitioned local exploitation — to effectively navigate constrained landscapes while maintaining both high convergence and well-distributed diversity along the CPF. Furthermore, a diversity enhancement strategy is designed to effectively balance solution diversity and constraint feasibility, enabling the swarm to achieve more uniform convergence toward the CPF and enhancing search effectiveness within the feasible solution region. Based on the proposed methodology, a novel algorithm for solving constrained multi-objective optimization problems (CMOPs) is developed. Extensive experimental results on 37 benchmark instances and 16 real-world engineering applications demonstrate that the proposed algorithm significantly outperforms nine state-of-the-art approaches, achieving either superior or highly competitive performance.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"105 ","pages":"Article 102409"},"PeriodicalIF":8.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147952562","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":"Hybrid evolutionary meta-optimization assisted by recurrent neural networks for interpretable parametric symbolic models","authors":"Flavio Adalberto Gomes ,&nbsp;Wanderson Rainer H. Araujo ,&nbsp;Marcio R. Cunha Reis ,&nbsp;Geovanne Pereira Furriel ,&nbsp;Guilherme Alberto S. Ribeiro ,&nbsp;Viviane M. Gomes Pacheco ,&nbsp;Marcella Scoczynski R. Martins ,&nbsp;Clóves Gonçalves Rodrigues ,&nbsp;Antonio Paulo Coimbra ,&nbsp;Wesley Pacheco Calixto","doi":"10.1016/j.swevo.2026.102413","DOIUrl":"10.1016/j.swevo.2026.102413","url":null,"abstract":"<div><div>The modeling of complex nonlinear systems poses persistent challenges for evolutionary optimization due to high-dimensional search spaces, fragile convergence behavior, and the difficulty of balancing exploration and exploitation under computational constraints. To address these limitations, this study investigated a hybrid evolutionary meta-optimization strategy assisted by recurrent neural networks, designed to guide and stabilize the search process during the identification of analytical models. The proposed framework integrates symbolic structure discovery with continuous parametric refinement, enabling the simultaneous optimization of coefficients and real-valued exponents within an evolutionary setting. A recurrent neural network was embedded in the meta-optimization loop to adaptively influence the evolutionary dynamics, improving convergence stability and search efficiency while preserving structural diversity. Comprehensive search trajectories were recorded and subsequently analyzed using Shapley Additive Explanations, allowing the association of optimized parameters with dominant physical mechanisms. The method was evaluated on synthetic benchmarks, electromechanical systems, thermodynamic maps, and physicochemical surfaces, consistently preserving global morphological features and structural coherence. In all cases, stable generalization behavior was observed, with normalized relative errors typically of the order of <span><math><mrow><mn>1</mn><msup><mrow><mn>0</mn></mrow><mrow><mo>−</mo><mn>4</mn></mrow></msup></mrow></math></span> and structural similarity indices exceeding 0.85, along with a statistically consistent separation from reference methods. The results demonstrate that neural-assisted evolutionary meta-optimization constitutes a viable strategy for improving convergence robustness and analytical model discovery in complex system identification problems.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"105 ","pages":"Article 102413"},"PeriodicalIF":8.5,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147952560","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}