Swarm and Evolutionary Computation最新文献

{"title":"Attention-enhanced RBF-assisted constrained multi-objective optimization via improvement value-driven adaptive global and local search","authors":"Hongchao Ouyang,&nbsp;Zan Yang,&nbsp;Senhai Liu,&nbsp;Hui Huang,&nbsp;Xianzhong Yu,&nbsp;Wei Xiong,&nbsp;Chengping Zhong","doi":"10.1016/j.swevo.2026.102517","DOIUrl":"10.1016/j.swevo.2026.102517","url":null,"abstract":"<div><div>Surrogate-assisted constrained multi-objective evolutionary algorithms (SACMOEAs) have demonstrated favorable applications in real-world scenarios, whereas most existing SACMOEAs are confined to the combination or collaboration ways between the surrogate models and multi-objective evolutionary algorithms, neglecting the improvement on the internal architecture of surrogate models as well as the adaptive refinement of mutation parameters within the evolutionary framework. Therefore, this paper proposes an attention-enhanced RBF-assisted constrained multi-objective optimization algorithms, where the improvement value-driven adaptive global and local search framework (IVAGLS) is designed. Specifically, drawing inspiration from the construction mechanism of polar coordinate system, the attention information (reflected by angle value) between individuals and the Euclidean distance information are simultaneously used to construct the corresponding basis functions, thereby enhancing the adaptability of the RBF model to the complex landscapes. Moreover, the global and local search phases are sequentially conducted where the improvement values of the aggregate tchebycheff function in the two phases are employed to adaptively adjust the mutation parameters. Subsequently, for selecting best candidate, the constraint violation and TE values are alternately compared to determine the best individual. For obtaining the exact offspring, two different acquisition functions are designed in a periodical mechanism to determine excellent individuals with diverse potential. Therefore, the IVAGLS achieves the adaptive adjustment for parameters and selection mechanisms within the sequential global and local evolution framework, thus the adaptability to complex problems can be enhanced greatly. Systematical experimental results on two benchmark test suites demonstrate that the IVAGLS is competitive in solving expensive multi-objective constrained optimization problems.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"108 ","pages":"Article 102517"},"PeriodicalIF":9.6,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148856316","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":"Multi-strategy adaptive differential evolution with dimension-based pre-selection and state-based population adjustment for numerical optimization","authors":"Yuhao Zhu,&nbsp;Mengnan Tian,&nbsp;Xueqing Yan,&nbsp;Xinduan Wang","doi":"10.1016/j.swevo.2026.102518","DOIUrl":"10.1016/j.swevo.2026.102518","url":null,"abstract":"<div><div>In this study, we propose an innovative dimension-based multi-strategy adaptive differential evolution algorithm, termed DMDE, by leveraging the dimensional information and search state of population to determine the offspring among various candidate solutions and dynamically adjust the population reduction module, respectively. First, to avoid the estimation bias caused by Euclidean distance in the process of selecting offspring in the composite search strategies, a new dimension-based pre-selection method is designed. Herein, the neighbor information of candidate solution based on each dimension is taken into account, and the median of their fitness values is leveraged to decide the offspring for target individual due to its strong tolerance for outliers. Meanwhile, a state-based adaptive population adjustment mechanism is further developed to dynamically regulate the algorithm’s search capability. In this mechanism, we record the fitness improvements of superior and inferior individuals to gauge the difficulty of optimization process, and we employ two distinct population reduction modules tailored to different optimization scenarios. Through these approaches, the proposed algorithm can more accurately identify suitable candidate solution as offspring and achieve a better trade-off between exploration and exploitation. Finally, the performance of DMDE is validated by benchmarking it against 21 renowned or up-to-date methods on CEC’2017 test suite within different dimensions and applying to two distinct practical application scenarios. Numerical outcomes reveal that the new algorithm exhibits significantly superior performance compared to its counterparts.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"108 ","pages":"Article 102518"},"PeriodicalIF":9.6,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148856317","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":"An imitation-based framework for large-scale sparse multi-objective optimization","authors":"Yi-Han Chen,&nbsp;Min-Rong Chen,&nbsp;Xiang Liu,&nbsp;Jian Weng","doi":"10.1016/j.swevo.2026.102525","DOIUrl":"10.1016/j.swevo.2026.102525","url":null,"abstract":"<div><div>Many real-world problems can be modeled as large-scale sparse multi-objective optimization problems (large-scale sparse MOPs). Recent studies on such problems have developed algorithms that significantly outperform conventional multi-objective evolutionary algorithms (MOEAs). However, these state-of-the-art algorithms typically treat decision variables as independent entities and optimize them in isolation. Moreover, they do not fully exploit the information embedded in non-dominated solutions, which hinders further performance gains. This paper introduces an imitation-based framework, dubbed SIF. The framework leverages value combinations of binary variables and distribution characteristics of real variables from non-dominated solutions to generate additional promising offspring solutions. Existing large-scale sparse MOEAs can be integrated with SIF to achieve enhanced performance. Experimental results on the SMOP benchmark problems and three real-world applications validate the effectiveness of the proposed SIF framework.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"108 ","pages":"Article 102525"},"PeriodicalIF":9.6,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148953567","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 Heuristic-Guided Genetic Algorithm for heterogeneous agricultural machinery scheduling with time windows","authors":"Ying Bi,&nbsp;Jinfei Zhao,&nbsp;Yanchao Guo,&nbsp;Caitong Yue,&nbsp;Jing Liang,&nbsp;Mengnan Liu,&nbsp;Xiaodong Zhu","doi":"10.1016/j.swevo.2026.102486","DOIUrl":"10.1016/j.swevo.2026.102486","url":null,"abstract":"<div><div>Effective scheduling of heterogeneous agricultural machinery (e.g., harvesters, tractors, and seeders) is critical for timely operations during busy farming seasons. However, most existing work focuses on homogeneous agricultural machinery scheduling, which is relatively simpler than heterogeneous agricultural machinery scheduling. To address this, the paper provides the first formal definition and mathematical model of the Heterogeneous Agricultural Machinery Scheduling Problem with Time Windows (HAMSP-TW). The model uniquely integrates multiple practical constraints previously treated in isolation, including heterogeneous machinery coordination, spatial transfer times between dispersed plots, multi-process job sequences, and plot-specific earliest start time windows. To solve HAMSP-TW, this paper proposes a Heuristic-Guided Genetic Algorithm (HGA) with problem-driven components, i.e., a two-part encoding scheme capturing machinery assignment and operation sequencing, a hybrid rule-based initialization strategy seeding high-quality, diverse solutions, and a heuristic-guided mutation operator leveraging domain knowledge for effective search. Experiments on 18 instances (60–400 plots) show the proposed HGA approach significantly outperforms benchmark algorithms, with advantages particularly pronounced in large-scale scenarios. Ablation studies confirm the substantial individual contribution of each proposed component. Gantt chart analysis reveals that optimized schedules are able to reduce idle time and improve machinery utilization, directly translating to shorter operation completion times.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"107 ","pages":"Article 102486"},"PeriodicalIF":9.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148657379","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":"Mechanism-level coupling in evolutionary computation: Social learning control via Value Shaping and Decision Biasing","authors":"Peng Gao,&nbsp;Bin Zhang,&nbsp;Jingya Wang,&nbsp;Ziyuan Wang,&nbsp;Chenglong Li,&nbsp;Jinglei Tan,&nbsp;Yifu Zeng","doi":"10.1016/j.swevo.2026.102505","DOIUrl":"10.1016/j.swevo.2026.102505","url":null,"abstract":"<div><div>Population-based metaheuristics are widely used in engineering artificial intelligence to automate parameter configuration and support multi-criteria decision making under black-box, nonconvex, and evaluation-limited objectives. In high-dimensional continuous optimization, they are often hindered by wasted evaluations from unfocused perturbations and by premature stagnation induced by unstable propagation of social information. We propose an evolutionary-computation-driven hybrid framework that couples structured trial generation with a selection-centered feedback controller. Selection-verified elite improvements govern what information propagates, a learned dimension-wise gate controls where it is injected, and payoff feedback schedules Value Shaping versus Decision Biasing. The controller is portable across heuristic backbones. Extensive evaluations on CEC2017/CEC2022 and constrained engineering design problems consistently demonstrate strong overall performance, robustness, and scalability. On CEC2017, the method achieves the best average rank across all tested dimensions, and its average rank improves by 34.2% from 10D to 100D. Engineering design studies further indicate stable feasible solutions with very low run-to-run variation on representative cases.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"107 ","pages":"Article 102505"},"PeriodicalIF":9.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148730826","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":"Reinforcement learning-driven multi-objective optimization for energy-efficient flexible job shop scheduling considering steam consumption in textile production","authors":"Lijun Pan,&nbsp;Xingjie Wang,&nbsp;Jing-fang Chen,&nbsp;Bin He,&nbsp;Jun Li","doi":"10.1016/j.swevo.2026.102498","DOIUrl":"10.1016/j.swevo.2026.102498","url":null,"abstract":"<div><div>In textile production workshops, electricity and steam are critical energy resources, which are consumed simultaneously and coordinated supply. Production scheduling requires the integrated optimization of power and steam consumption, as well as operational efficiency. This study focuses on developing effective scheduling strategies that can shorten production cycles and reduce energy consumption simultaneously. First, an energy-efficient flexible job shop scheduling model considering steam consumption (EEFJSP-CSC) is established to minimize the maximum completion time, total steam consumption, and total power usage. We present a break-even duration strategy to determine the shutdown of steam equipment in the model. Secondly, we propose a state-guided Q-learning-driven multi-objective evolutionary algorithm (SGQL-MOEA) to solve EEFJSP-CSC. During the decoding phase, a five-stage energy-saving decoding method is proposed by using differentiated right-lean strategies and critical path slack optimization techniques to minimize steam waste. During the evolutionary phase, a state-guided Q-learning-driven search mechanism is proposed to adaptively select suitable operators. Specifically, a state representation approach is designed using the improvements of hypervolume and fluctuation of population crowding distance, and an exponential decay function is designed to guide the learning of the agent. Finally, synthetic benchmark tests demonstrate that our algorithm produces high-quality Pareto solution-sets consistently across various scale scenarios. Its convergence performance and distribution characteristics surpass existing algorithms. Ablation experiments reveal that the state function-guided reinforcement learning search mechanism enhances algorithmic performance.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"107 ","pages":"Article 102498"},"PeriodicalIF":9.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148730829","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 unified evolutionary framework for structural–topological mixed-variable optimization in STAR-RIS-aided systems","authors":"Bai Yan,&nbsp;Jin Zhang,&nbsp;Qi Zhao","doi":"10.1016/j.swevo.2026.102501","DOIUrl":"10.1016/j.swevo.2026.102501","url":null,"abstract":"<div><div>Beamforming design in simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS)-aided communications exhibits the characteristics of structured–topological mixed-variable optimization (STMVO), involving bounded, discrete, and periodic variables with structured coupling constraints. A theoretical analysis of these structural properties reveals that existing Euclidean-update-based evolutionary baselines may be insufficient when these structural and topological properties are not explicitly considered. From a system-level perspective, these challenges arise from mismatches between optimization dynamics and variable structure, rather than from specific algorithms. Motivated by these insights, we propose ST-EF, a unified evolutionary framework for STMVO. ST-EF introduces three major innovations: (i) a unified representation that embeds heterogeneous variables into a normalized space while preserving topology-specific adjacency; (ii) topology-aware repair operators that respect boundary and periodic conditions, ensuring unbiased geometric consistency; and (iii) topology-aligned deviation mapping with margin correction that maintains balanced sampling over discrete variables and structured coupling constraints, thereby preventing premature collapse. Extensive experiments on STAR-RIS beamforming tasks demonstrate that ST-EF achieves improved robustness and competitive runtime scalability over representative baselines, highlighting its potential as a principled paradigm for solving structured mixed-variable problems.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"107 ","pages":"Article 102501"},"PeriodicalIF":9.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148730225","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":"Pareto evolutionary algorithm based on data mining for a kind of task assignment problem considering agent reliability","authors":"Zuocheng Li,&nbsp;Chang Li,&nbsp;Chang Liu,&nbsp;Rongjuan Luo,&nbsp;Junqing Li","doi":"10.1016/j.swevo.2026.102470","DOIUrl":"10.1016/j.swevo.2026.102470","url":null,"abstract":"<div><div>The task assignment problem (TAP) is a well-known combinatorial optimization problem with rich applications. Most existing TAPs have minimized cost criteria, and are often unsuitable for complex decision-making. This work focuses on a specific multi-objective TAP considering the reliability of agents (MOTAP/RA). The objective is to optimize the total assignment cost and total agent reliability. To solve the problem, we present a Pareto evolutionary algorithm based on data mining (PEADM). The primary contributions of this work are two-fold: a mathematical model of MOTAP/RA and a novel search model of PEADM. For the mathematical model, we propose for the first time the MOTAP/RA model based on which problem-specific heuristics are introduced. For the PEADM framework, we introduce two populations to capture valuable algorithm knowledge: a main population (MP) and an auxiliary population (AP). We update MP using an improved nondominated sorting method while extracting pertinent frequent patterns from MP through a dedicated data mining approach. Then, we generate new individuals of AP based on the mined frequent patterns and so improve MP based on AP. Results of experiments on 180 instances with up to 310 tasks show that PEADM competes favorably with state-of-the-art solution methods.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"107 ","pages":"Article 102470"},"PeriodicalIF":9.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148642255","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 decomposition-based reinforcement learning approach for production and distribution integrated scheduling problem of perishable products","authors":"Yuan-yuan Yang,&nbsp;Bin Qian,&nbsp;Li-jun Wang,&nbsp;Jun-qing Li,&nbsp;Rong Hu","doi":"10.1016/j.swevo.2026.102478","DOIUrl":"10.1016/j.swevo.2026.102478","url":null,"abstract":"<div><div>Production and distribution integrated scheduling for perishable products (PDISP_PP) is an NP-hard problem in modern supply chains, where the fundamental challenge lies in balancing production efficiency, distribution responsiveness, and product freshness. It is relevant to cold-chain food, pharmaceutical, fresh agricultural products, and make-to-order delivery systems, where even small delays between production completion and vehicle arrival can reduce freshness, impair service, and increase operating costs. This study proposes a decomposition-based reinforcement learning incorporating problem properties (DRLIPP). DRLIPP first constructs high-quality vehicle routes using a hybrid large neighborhood search (HLNS) and then decomposes the production into route-induced flow-shop scheduling subproblems solved by a double deep Q-network-based policy. This architecture reduces the coupling between routing and production sequencing while retaining the integrated objective of production and distribution scheduling. Computational experiments demonstrate that DRLIPP provides competitive running times and better solution quality compared with representative metaheuristics, alternative decoupling strategies, action variants, and classical scheduling rules.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"107 ","pages":"Article 102478"},"PeriodicalIF":9.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148642692","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":"Deep reinforcement learning assisted adaptive guided vectors for large-scale multimodal multi-objective optimization","authors":"Min Zuo,&nbsp;Guoqing Li,&nbsp;Yu Xin","doi":"10.1016/j.swevo.2026.102485","DOIUrl":"10.1016/j.swevo.2026.102485","url":null,"abstract":"<div><div>Multimodal multi-objective optimization problems (MMOPs) aim to identify multiple equivalent Pareto solution sets (PSs) in the decision space with the same Pareto front (PF) in the objective space. As the number of decision vectors increases, traditional multimodal multi-objective evolutionary algorithms (MMEAs) often become ineffective because distance-based diversity measures lose their discriminative ability in large-scale MMOPs (LMMOPs) with high-dimensional decision vectors. To address this challenge, we propose an adaptive guided vector assisted large-scale multimodal multi-objective evolutionary using deep reinforcement learning, termed as AGVLMMEA, for solving LMMOPs. A deep Q-network is designed in AGVLMMEA to adaptively select distinct guided vectors for guiding the direction of evolutionary search. Subsequently, a dimension-analysis two complementary operator including a global exploration operator and a local exploitation operator is designed, where a global exploration operator performs full-dimensional variation to improve global exploration and a local refinement operator focuses on important dimensions identified through guided-vector analysis to amplify local search. Finally, a similarity-based dynamic dividing and merging evolutionary mechanism (DAM) is incorporated into a multi-population co-evolutionary framework, and it dynamically adjusts the number of subpopulations to preserve multiple equivalent PSs. Experimental results on 15 LMMOPs test problems demonstrate that AGVLMMEA consistently achieves superior decision space diversity while maintaining competitive convergence in the objective space. Compared with seven state-of-the-art algorithms, AGVLMMEA more effectively discovers and preserves multiple equivalent PSs, particularly in high-dimensional space.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"107 ","pages":"Article 102485"},"PeriodicalIF":9.6,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148642694","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}