Swarm and Evolutionary Computation最新文献

{"title":"Bi-stage learning differential evolution for multimodal optimization problems","authors":"Chang-Long Wang ,&nbsp;Zi-Jia Wang ,&nbsp;Yi-Biao Huang ,&nbsp;Dan-Ting Duan ,&nbsp;Zhi-Hui Zhan ,&nbsp;Sam Kwong ,&nbsp;Jun Zhang","doi":"10.1016/j.swevo.2025.101974","DOIUrl":"10.1016/j.swevo.2025.101974","url":null,"abstract":"<div><div>Multimodal optimization problems (MMOPs) require the identification of multiple optimal solutions for decision makers. To address MMOPs, algorithms must enhance the population diversity to find more global optimal regions while simultaneously refine the solution accuracy on each optimum. Therefore, in this paper, we introduces a bi-stage learning differential evolution (BLDE) with two learning stages: the pre-learning <em>Find</em> stage and the post-learning <em>Refine</em> stage. First of all, a bi-stage learning niching technique (BLNT) is proposed, which forms wide niches for full exploration in the pre-learning <em>Find</em> stage, while adaptively adjusts the niche radius for each individual to refine its corresponding solution accuracy in the post-learning <em>Refine</em> stage. Subsequently, a bi-stage learning mutation strategy (BLMS) is developed, enabling each individual to adaptively choose the suitable mutation strategy, achieving effective guidance for evolution. Moreover, different from other DE-based multimodal algorithms with only one selection operator, a bi-stage learning selection strategy (BLSS) is proposed to determine the suitable selection operator in different learning stages and preserve the promising individuals. The widely-used multimodal benchmark functions from CEC2015 competition are employed to evaluate the performance of BLDE. The results demonstrate that BLDE generally outperforms or at least comparable with other state-of-the-art multimodal algorithms, including the champion of CEC2015 competition. Moreover, BLDE is further applied to the real-world multimodal nonlinear equation system (NES) problems to demonstrate its applicability.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"96 ","pages":"Article 101974"},"PeriodicalIF":8.2,"publicationDate":"2025-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144083908","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":"Genetic programming with surrogate evaluation for the electric vehicle routing problem","authors":"Francisco J. Gil-Gala ,&nbsp;Marko Đurasević ,&nbsp;Domagoj Jakobović ,&nbsp;Ramiro Varela","doi":"10.1016/j.swevo.2025.101969","DOIUrl":"10.1016/j.swevo.2025.101969","url":null,"abstract":"<div><div>The focus on environmental sustainability has made the Electric Vehicle Routing Problem (EVRP) an important area of research. Routing Policies (RPs) offer a simple and efficient approach to solving VRPs, providing advantages over methods like metaheuristics by quickly generating solutions. However, designing efficient RPs manually can be time-consuming. Therefore, there is a need to explore hyper-heuristic approaches, particularly Genetic Programming (GP), to automate the design of RPs. However, population-based evolutionary algorithms like GP often require a significant amount of computational resources, especially for fitness calculation. Therefore, surrogate evaluation is essential in enhancing efficiency, especially in GP, where multiple problem instances need to be solved to evaluate each chromosome. In this study, we employ surrogate models within GP to design RPs for EVRP with hard time windows. The experiments show that the RPs designed by GP with surrogate models outperform those produced by standard GP approaches while still requiring less computational time to be generated. Moreover, the RPs designed with GP using surrogate models are also smaller, and consequently, they are also more efficient and easier to interpret.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"96 ","pages":"Article 101969"},"PeriodicalIF":8.2,"publicationDate":"2025-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144083907","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 multi-objective benchmark for UAV path planning with baseline results","authors":"Daison Darlan ,&nbsp;Oladayo S. Ajani ,&nbsp;Anand Paul ,&nbsp;Rammohan Mallipeddi","doi":"10.1016/j.swevo.2025.101968","DOIUrl":"10.1016/j.swevo.2025.101968","url":null,"abstract":"<div><div>Unmanned Aerial Vehicles (UAVs) are increasingly deployed in complex environments for applications such as urban logistics, surveillance, and environmental monitoring. Path planning for UAVs in these settings is a multi-objective optimization problem that must balance conflicting criteria such as safety, efficiency, and regulatory compliance. While heuristic methods like A* and Rapidly-Exploring Random Tree (RRT) have been applied to this problem, they often struggle with convergence speed and solution quality in obstacle-rich, three-dimensional spaces. Multi-Objective Evolutionary Algorithms (MOEAs) have shown strong potential for tackling such challenges, yet the absence of a comprehensive, standardized benchmark for UAV path planning continues to impede meaningful evaluation and comparison across studies. To address this gap, this paper makes three key contributions: (i) a parameterized framework for generating UAV path-planning problems across three representative environment types — urban, suburban, and mountainous — with adjustable scenario difficulty and real-world constraints such as no-fly zones; (ii) a curated benchmark suite comprising 14 diverse and rigorously designed test problems for reproducible and consistent algorithm evaluation; and (iii) baseline performance results for several state-of-the-art MOEAs, offering researchers clear reference points for future comparisons. By providing both a versatile testbed and a standardized evaluation methodology, this work aims to facilitate the development and fair assessment of UAV path-planning algorithms in realistic and challenging environments.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"96 ","pages":"Article 101968"},"PeriodicalIF":8.2,"publicationDate":"2025-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144071959","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":"Multifaceted collaborative evolutionary constrained multimodal multiobjective optimization","authors":"Zeyi Wang,&nbsp;Songbai Liu,&nbsp;Lijia Ma,&nbsp;Qiuzhen Lin,&nbsp;Jianyong Chen","doi":"10.1016/j.swevo.2025.101951","DOIUrl":"10.1016/j.swevo.2025.101951","url":null,"abstract":"<div><div>In addressing constrained multimodal multiobjective optimization problems (CMMOPs), this paper proposes a multifaceted collaborative evolutionary algorithm (MCEA) designed to balance feasibility, convergence, and diversity in both the objective and decision spaces. Existing approaches often focus solely on maintaining population diversity or feasibility, neglecting the intricacies of CMMOPs, which require simultaneous consideration of multiple conflicting goals. Our MCEA framework features a local–global collaborative search strategy that employs dynamic clustering for effective exploration and exploitation of diverse decision space regions. Additionally, a parent–offspring collaborative transfer strategy facilitates knowledge sharing between populations, enhancing convergence early in the evolutionary process and preserving diversity in later stages. Furthermore, we customize an objective-search space collaborative selection strategy that filters solutions based on population diversity across both spaces. Extensive experiments on thirty-one benchmark CMMOPs demonstrate that MCEA significantly outperforms state-of-the-art algorithms on more than half of the test problems, as measured by IGD, IGDX, RPSP, and HV performance indicators. Furthermore, MCEA effectively locates multiple Pareto subsets, showcasing its ability to balance convergence, diversity, and feasibility in solving CMMOPs. This work underscores the importance of a comprehensive approach to tackling the complexities of CMMOPs and provides valuable insights for future research in this domain.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"96 ","pages":"Article 101951"},"PeriodicalIF":8.2,"publicationDate":"2025-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144083906","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 estimation of distribution algorithm for agile earth observing satellite task scheduling problem","authors":"Chunchun Ma ,&nbsp;Panxing Huang ,&nbsp;Xiaoze Liu ,&nbsp;Chu-ge Wu ,&nbsp;Rui Xu","doi":"10.1016/j.swevo.2025.101971","DOIUrl":"10.1016/j.swevo.2025.101971","url":null,"abstract":"<div><div>Agile Earth Observing Satellites (AEOSs) represent a new generation of Earth observation satellites, widely used for various observation tasks. To efficiently utilize the visible and observing durations of the orbiting AEOS, the AEOS scheduling problem (AEOSSP) is formulated to maximize the overall observation profit while satisfying the complex operational constraints. In this paper, a hybrid Estimation of Distribution Algorithm (EDA) that incorporates three knowledge-oriented local search operators is proposed to efficiently solve the AEOSSP. The multiple multidimensional knapsack problem with conflicts (MMdKPC) is first modeled and used to formulate AEOSSP. An EDA probability model as well as its updating and sampling mechanisms, is designed to generate solutions to explore the solution space and generate potential solutions. In addition, based on the characteristics of MMdKPC, three knowledge-oriented local search operators are developed to improve the solution. Based on the benchmark instances and simulation data provided sampled from Satellite Tool Kit, the comparison simulation experiments are carried out. The results validate the effectiveness of three knowledge-oriented local search operators, respectively. Additionally, the proposed hybrid EDA performs better compared to the existing state-of-the-art algorithms in terms of overall observation profit.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"96 ","pages":"Article 101971"},"PeriodicalIF":8.2,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143947233","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":"Probabilistic bootstrap-based evolutionary algorithm for three-objective wind farm turbine position optimization","authors":"Chen Zhang ,&nbsp;Haotian Li ,&nbsp;Xiuxian Li ,&nbsp;Jiujun Cheng ,&nbsp;Zhenyu Lei ,&nbsp;Shangce Gao","doi":"10.1016/j.swevo.2025.101972","DOIUrl":"10.1016/j.swevo.2025.101972","url":null,"abstract":"<div><div>Amid the worsening energy crisis, wind farm layout optimization (WFLO) to increase power generation, reduce costs, and mitigate potential environmental impacts is of great significance. This paper formulates three-objective wind farm layout optimization (TWFLO) which is rarely considered, aiming to effectively utilize existing information to optimize power output, land usage, and costs. We propose a new algorithm (MOEA/D-P) based on probability distributions to guide turbine placement and improve the performance of a Multiobjective Evolutionary Algorithm Based on Decomposition (MOEA/D). This algorithm addresses the issue of neglecting valuable information during layout optimization. Additionally, we make improvements to the traditional land-use function to avoid situations where non-convex layouts result in an area calculation of zero. The MOEA/D-P is tested on six different initial layouts and compared with five algorithms under two wind conditions. Results are evaluated using inverted generational distance, hypervolume, and scatter plot distributions. The impact of initial probability distribution on algorithm performance is discussed under four simple wind conditions. The results show that MOEA/D-P outperforms the other five algorithms in terms of performance.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"96 ","pages":"Article 101972"},"PeriodicalIF":8.2,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144068021","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":"Terminal normalization in genetic programming for dynamic flexible job shop scheduling","authors":"Binzi Xu ,&nbsp;Xinyu Cao ,&nbsp;Shuwen Zhang ,&nbsp;Jiangping Shen ,&nbsp;Chaohua Wang ,&nbsp;Songhua Wang ,&nbsp;Dengchao Huang ,&nbsp;Chun Wang ,&nbsp;Yi Mei ,&nbsp;Yan Wang","doi":"10.1016/j.swevo.2025.101970","DOIUrl":"10.1016/j.swevo.2025.101970","url":null,"abstract":"<div><div>The dynamic flexible job shop scheduling problem (DFJSSP) is a challenging NP-hard problem that has been studied for decades. Among various optimization algorithms, genetic programming hyper-heuristic (GPHH) excels in learning dispatching rules (DRs) for online scheduling. However, an overlooked phenomenon in GPHH is that terminals with larger magnitude may overshadow smaller ones, potentially limiting GPHH’s effectiveness. The min–max normalization of terminal values may resolve this issue. Thus, this study aims to investigate its impact on GPHH and its potential benefits. Since the terminal ranges used for normalization cannot be obtained directly and in advance within the current GPHH framework, a range archive (RA) enhanced mechanism is proposed for normalization, including three strategies: RA-lite, RA-all, and simulation-RA. These strategies are evaluated on a multi-objective DFJSSP targeting mean tardiness and total energy consumption. Experimental results on small function set demonstrate the presence of inconsistent terminal ranges, in which case normalization can effectively address this issue and enhance the optimization performance of GPHH. While experimental results on large function set indicate that multiplication and division achieve a similar effect, suggesting limited benefit from normalization. However, further analysis reveals that normalization, particularly simulation-RA, can produce more interpretable DRs without compromising performance.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"96 ","pages":"Article 101970"},"PeriodicalIF":8.2,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144068169","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 multi-objective evolutionary algorithm for adjustable-speed job shop scheduling with multi-trip automated guided vehicles constraints","authors":"Ming Lu ,&nbsp;Chengjie Zhang ,&nbsp;Gongfu Su ,&nbsp;Long Chen ,&nbsp;Ying Zou ,&nbsp;Pei Li","doi":"10.1016/j.swevo.2025.101950","DOIUrl":"10.1016/j.swevo.2025.101950","url":null,"abstract":"<div><div>Currently, escalating energy consumption and the resulting environmental crisis are propelling the manufacturing industry into an unprecedented ecological predicament. In traditional job shop scheduling research, most studies overlook the transportation system. Furthermore, existing production scheduling problems with transportation resource constraints consider only single-trip scheduling of Automated Guided Vehicles (AGVs), leading to resource inefficiency. Therefore, this study investigates the adjustable-speed energy-efficient job shop scheduling with mutil-trip AGV constraints (AEJSP-MAC), proposing a multi-trip rescheduling heuristic algorithm based on the initial scheduling of single-trip. Given the NP-hard nature of the problem, a Hybrid Meta-heuristic Multi-Objective Evolutionary Algorithm (HMOMA) is proposed. HMOMA introduces three performance enhancement strategies: firstly, an effective initialization strategy is employed to generate high-quality initial solutions; secondly, crossover and mutation strategies targeting global exploration are adopted; thirdly, a novel local search framework based on Adaptive Large Neighborhood Search (ALNS) is integrated into the multi-objective evolutionary algorithm. Three destruction strategies are designed based on the depth and width of destruction, with three repair strategies developed to the problem’s characteristics. Through three sets of experiments, the effectiveness of the multi-trip AGV model, the applicability and superiority of MOALNS, and the high potential of HMOMA in solving the AEJSP-MAC.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"96 ","pages":"Article 101950"},"PeriodicalIF":8.2,"publicationDate":"2025-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143942087","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":"Phasor particle swarm optimization-based control for Quadcopter systems: An event-triggered impulsive super twisting approach","authors":"Ahsan Nawaz Jadoon ,&nbsp;Abdullah Mughees ,&nbsp;Mohammad Nashit Shah ,&nbsp;Mujahid Nawaz Jadoon","doi":"10.1016/j.swevo.2025.101955","DOIUrl":"10.1016/j.swevo.2025.101955","url":null,"abstract":"<div><div>Achieving precise and robust trajectory tracking in quadcopter systems remains a significant challenge due to inherent system nonlinearities, external disturbances, and actuator constraints. Conventional control strategies, such as classical sliding mode control (SMC) and proportional–integral–derivative (PID) controllers, often suffer from high-frequency chattering and limited adaptability to dynamic environments. To address these limitations, this study presents a novel control framework that synergistically integrates event-triggered impulsive super twisting terminal sliding mode control (ETISTT-SMC) with phasor particle swarm optimization (PPSO). The proposed approach enhances control precision while mitigating chattering effects, offering superior performance over traditional techniques. Unlike conventional SMC-based methods that rely on fixed control gains, ETISTT-SMC introduces event-triggered and impulsive control mechanisms, reducing unnecessary control updates and enhancing computational efficiency. PPSO, a refined swarm intelligence optimization algorithm, is employed to dynamically tune the controller parameters, ensuring optimal performance across varying flight conditions. This integration enables the quadcopter to achieve improved transient response, higher tracking accuracy, and greater robustness against disturbances. Extensive numerical simulations validate the efficacy of the proposed framework. Comparative analyses demonstrate that ETISTT-SMC achieves exceptional yaw stabilization, while SMC-PPSO and ETISTT-PPSO provide smoother and more responsive roll and pitch control. Furthermore, Lyapunov stability analysis rigorously establishes system stability under diverse operational conditions. Evaluations on complex 3D trajectory scenarios further confirm the robustness of the proposed methodology. Beyond theoretical and simulation-based validation, the modularity and adaptability of the control framework make it well-suited for real-world applications, including autonomous aerial surveillance, precision agriculture, disaster response, and infrastructure inspection. This research lays a strong foundation for future experimental validation and real-world deployment, paving the way for more efficient and resilient UAV control systems.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"96 ","pages":"Article 101955"},"PeriodicalIF":8.2,"publicationDate":"2025-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143947234","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":"Dual-population collaborative prediction with reinforcement learning adjusting for dynamic multi-objective optimization","authors":"Li Yan ,&nbsp;Yinjin Wu ,&nbsp;Yaxin Li ,&nbsp;Boyang Qu ,&nbsp;Jing Liang ,&nbsp;Kunjie Yu ,&nbsp;Dezheng Zhang","doi":"10.1016/j.swevo.2025.101981","DOIUrl":"10.1016/j.swevo.2025.101981","url":null,"abstract":"<div><div>Numerous studies have paid significant attention to dynamic multi-objective optimization problems (DMOPs). Prediction-based methods, as an efficient dynamic response strategy, can estimate the positions of the new Pareto sets and generate some individuals around them. However, current prediction-based approaches focus on estimating the moving trends or change patterns of the solutions. The diversity and distribution of the predicted population are not considered sufficiently, affecting the quality of solutions. Further, it is crucial to balance the convergence and diversity of the predicted solutions that can determine the final performance of the algorithm. Therefore, this paper proposes a dual-population collaborative prediction approach with reinforcement learning adjusting for solving DMOPs. Based on the framework of dual-population collaboration, two prediction strategies, cluster-based multiple prediction and manifold prediction based on knee points, which respectively focus on convergence and diversity, are designed to generate two predicted subpopulations. Further, to balance the diversity and convergence of the final predicted population, a reinforcement learning-based population generation mechanism is presented to adjust and determine the proportion of the two subpopulations in the final predicted population by evaluating each strategy’s performance. We performed experiments on the CEC2018 test problems to validate its performance. Experimental results demonstrate its effectiveness and superiority.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"96 ","pages":"Article 101981"},"PeriodicalIF":8.2,"publicationDate":"2025-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143941992","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}