Swarm and Evolutionary Computation最新文献

{"title":"Differential evolution based on individual information parameter setting and diversity measurement of aggregated distribution","authors":"Zhenghao Song ,&nbsp;Liangliang Sun ,&nbsp;Natalja Matsveichuk ,&nbsp;Yuri Sotskov ,&nbsp;Shenglong Jiang ,&nbsp;Yang Yu","doi":"10.1016/j.swevo.2024.101793","DOIUrl":"10.1016/j.swevo.2024.101793","url":null,"abstract":"<div><div>Differential Evolution (DE) and its variants have been widely applied to numerical optimization and engineering optimization problems owing to their simple operation, excellent optimization capacity, and high robustness. Many DE variants heavily rely on information from individuals to generate candidate offspring and update control parameters, which limits the search capacity during the later stages of evolution. A more appropriate search scheme is required to enhance DE’s performance by utilizing information from both individuals and distributions. This paper presents a novel DE algorithm based on individual information and diversity measurement of aggregated distribution, termed IDMDE. First, to effectively regulate the search behavior of individuals, a hybrid parameter generation mechanism based on individual information is proposed. This ensures the algorithm always searches for a promising direction and fully utilizes the individuals’ effective information. Second, to avoid the waste of search capacity during parameter updates in many DE variants, a novel parameter update strategy based on individual diversity is proposed, along with a new weighting scheme that utilizes the fitness and position information of individuals. Lastly, to mitigate premature convergence and stagnation during evolution, a diversity measurement mechanism based on aggregated distribution is proposed, which uses the search performance and diversity of individuals to evaluate the evolutionary state. The proposed IDMDE algorithm is evaluated by comparing it with five advanced algorithms on CEC2013, CEC2014, CEC2017, and CEC2022 across different dimensions. Moreover, the experimental results on the truss structure optimization problem confirm its feasibility in real-world optimization.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"92 ","pages":"Article 101793"},"PeriodicalIF":8.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183495","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 effective multi-objective community development algorithm and its application to identify control model of supercritical units","authors":"Mingliang Wu,&nbsp;Dongsheng Yang,&nbsp;Yingchun Wang,&nbsp;Jiayue Sun","doi":"10.1016/j.swevo.2024.101790","DOIUrl":"10.1016/j.swevo.2024.101790","url":null,"abstract":"<div><div>The community development algorithm (CDA) performs well in solving numerical optimization problems and practical engineering applications. To better utilize CDA, this paper combines it with non-dominated sorting and clustering-based special crowding distance to form a multi-objective community development algorithm (MOCDA). In addition, the helper selection mechanism is devised to select the more suitable learning objects for particles. A series of comprehensive examinations prove that MOCDA is better than the other 9 state-of-the-art competitors on the multimodal multi-objective Congress on Evolutionary Computation 2020 and imbalanced distance minimization benchmark problems. Quantitatively, MOCDA leads MMODE_CSCD by 22.44%, demonstrating a strong ability to solve multimodal multi-objective optimization problems. For engineering practice, MOCDA is employed to identify the three-input, three-output control model of supercritical units by regarding the data of multiple time periods as multiple objectives, and the experimental results show that this approach is more effective than the direct summation of the single objective algorithm. During the encoding process, an additional position is added for the solution’s chromosome to control whether or not a delay link works. Experimental results show that this method has a lower root mean square error and significantly reduces the maximum error at the initial moment compared to the encoding scheme with a fixed delay link. Most importantly, the identification accuracy of MOCDA is much higher than that of other algorithms, indicating its superiority in solving challenging multi-objective problems in the real world. The source code of MOCDA is publicly available at: <span><span>https://github.com/Mingliang-Wu/MOCDA.git</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"92 ","pages":"Article 101790"},"PeriodicalIF":8.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183601","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":"Approaches to the truck-drone routing problem: A systematic review","authors":"Jie Duan ,&nbsp;He Luo ,&nbsp;Guoqiang Wang","doi":"10.1016/j.swevo.2024.101825","DOIUrl":"10.1016/j.swevo.2024.101825","url":null,"abstract":"<div><div>In recent years, research on collaborative routing between trucks and drones (also known as unmanned aerial vehicles, UAVs) has received considerable attention, with applications gradually emerging across various fields. This paper aims to summarize the theoretical research on the truck-drone routing problem from 2015 to 2024, categorizing, introducing, and comparing the approaches used, with a particular focus on their innovative aspects. It then collects public datasets and extension techniques for various types of problems. Typical application domains and the current state of the art in the field are then presented, laying the groundwork for scenario transition. Finally, based on the discussion, future development trends and research directions in the field are predicted.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"92 ","pages":"Article 101825"},"PeriodicalIF":8.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183604","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":"Metaheuristics should be tested on large benchmark set with various numbers of function evaluations","authors":"Adam P. Piotrowski ,&nbsp;Jaroslaw J. Napiorkowski ,&nbsp;Agnieszka E. Piotrowska","doi":"10.1016/j.swevo.2024.101807","DOIUrl":"10.1016/j.swevo.2024.101807","url":null,"abstract":"<div><div>Numerical metaheuristics are often tested on mathematical problems collected into a benchmark set. There are many benchmark sets, but the number of problems in a particular benchmark rarely exceeds 30, sometimes is much lower. The stopping condition is frequently based on the maximum number of function evaluations, commonly set to a single value, somehow related to the problem's dimensionality. However, the ranking of algorithms may highly depend on the number of allowed function evaluations. As a result, by changing the number of function evaluations, different algorithms may be promoted as the best ones. In the present study, we suggest that metaheuristics should rather be tested using independently four different numbers of maximum function evaluations that differ by orders of magnitude (e.g., 5.000, 50.000, 500.000, and 5.000.000 function calls). We recommend performing tests on both higher- and lower-dimensional versions of 72 problems from three well-known benchmark sets (CEC 2014, CEC 2017, CEC 2022). The ranking of algorithms based on each particular computational budget should be discussed separately. This way, various algorithms may show their strengths in shorter or longer searches and weaknesses in other cases, encouraging more nuanced discussion. We also show that the number of benchmark problems does matter: results based on larger sets of problems are much more frequently statistically significant than results based on a single small benchmark set. There is also a difference in the percentage of statistically significant results between tests performed with lower and higher numbers of allowed function calls.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"92 ","pages":"Article 101807"},"PeriodicalIF":8.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183504","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":"Diversity enhancement-based Differential Evolution with a novel perturbation strategy","authors":"Zhenghao Song ,&nbsp;Liangliang Sun ,&nbsp;Natalja Matsveichuk ,&nbsp;Yuri Sotskov","doi":"10.1016/j.swevo.2024.101822","DOIUrl":"10.1016/j.swevo.2024.101822","url":null,"abstract":"<div><div>Differential Evolution (DE), an advanced population-based optimization algorithm, has been widely adopted to deal with complicated real-parameter optimization problems. However, the original DE lacked sufficient exploration capabilities and did not use population diversity mechanisms to enhance performance. To mitigate this deficiency, a diversity enhancement-based differential evolution with a novel perturbation strategy (DE-NPS) is proposed. Firstly, a bi-stage parameter control strategy is proposed to adjust search behavior for different stages of evolution, thus achieving a better balance between exploration and exploitation. Secondly, a perturbation strategy based on a logarithmic spiral equation is incorporated into crossover operation as a supplementary strategy to the trial vector generation scheme to maintain population diversity. Lastly, a population diversity enhancement strategy based on the covariance matrix is developed to locate and update stagnant individuals in the population. CEC2014, CEC2017, and CEC2022 test suites are used to verify the effectiveness of DE-NPS. The experimental results demonstrate that DE-NPS can obtain highly competitive performance compared to other powerful algorithms regarding optimization accuracy and convergence rate. In addition, DE-NPS is applied to a real-world optimization problem and yields satisfactory results.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"92 ","pages":"Article 101822"},"PeriodicalIF":8.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183608","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":"Population-level center-based sampling for meta-heuristic algorithms","authors":"Rasa Khosrowshahli ,&nbsp;Shahryar Rahnamayan ,&nbsp;Amin Ibrahim ,&nbsp;Azam Asilian Bidgoli ,&nbsp;Masoud Makrehchi","doi":"10.1016/j.swevo.2024.101827","DOIUrl":"10.1016/j.swevo.2024.101827","url":null,"abstract":"<div><div>In recent years, the challenge of enhancing the efficiency and effectiveness of meta-heuristic algorithms has gained significant attention. Center-based sampling has shown promise in addressing this challenge, yet its application often requires customization for specific algorithms, limiting its generalizability. This study identifies a gap in the literature regarding the operation-independent application of center-based sampling. To address this, we propose a novel center-based sampling strategy at the population level, which can be seamlessly integrated into any population-based optimization algorithm. Our approach employs a collaborative multi-parent method to generate multiple center-based solutions, thereby increasing diversity and exploiting the solution space more effectively. We introduce two specific strategies: cluster-driven center-based sampling for single-objective optimization and ranking-driven center-based sampling for multi-objective optimization. The performance of these strategies is evaluated using the benchmark functions for the CEC-2017 competition on 5 single- and 6 many-objective evolutionary algorithms, demonstrating <span><math><mrow><mn>40</mn><mtext>%</mtext><mo>∼</mo><mn>100</mn><mtext>%</mtext></mrow></math></span> statistical fitness improvement ratio over parent meta-heuristic algorithms, respectively. These findings highlight the potential of population-level center-based sampling to enhance the performance of meta-heuristic algorithms.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"92 ","pages":"Article 101827"},"PeriodicalIF":8.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Task offloading for Industrial Internet of Things based on evolutionary multiobjective multitasking optimization","authors":"Yan-Yang Cheng ,&nbsp;Zheng-Yi Chai ,&nbsp;Ya-Mei Xia ,&nbsp;Xu Liu ,&nbsp;Gao-Min Yin","doi":"10.1016/j.swevo.2024.101786","DOIUrl":"10.1016/j.swevo.2024.101786","url":null,"abstract":"<div><div>In mobile edge computing(MEC) scenarios, offloading computation-intensive and delay-critical tasks to nearby mobile edge servers with abundant computing resources can reduce the time delay(TD) of tasks and energy consumption(EC) of terminal devices, thus improving application performance and user experience. In order to reduce the TD of tasks and the EC of factory terminal devices(FTDs), MEC is introduced into the Industrial Internet of Things(IIoT). Because TD and EC are conflicting objectives, most existing research models the task offloading problem(TOP) as a multiobjective optimization problem to optimize both TD and EC. However, as the number of FTDs and tasks in the MEC network increases, the scale and computational cost of the TOP increases, and it is very challenging to obtain the optimal task offloading scheme through multiobjective optimization. Multitasking optimization can make use of knowledge transfer between related tasks to promote the solving efficiency of each task. Based on this, we model the TOP in IIoT as a multiobjective TOP (MTOP). A cheap task that is highly similar to MTOP is constructed, and a novel evolutionary multiobjective multitasking framework is developed, which uses the positive knowledge transfer between tasks to optimize TD and EC in the network. Then, an effective evolutionary multiobjective multitasking algorithm is proposed, which includes multi-functional knowledge transfer strategy(MFKT) and adaptive cheap task update strategy(ACTU). MFKT uses decision variables with different functions in the cheap tasks for positive knowledge transfer, so as to improve the performance of MTOP. ACTU dynamically updates the cheap task to maintain knowledge transfer between tasks. In different test instances, the proposed algorithm is compared with the existing multiobjective and multitasking algorithms. The experimental results show that the proposed algorithm is more competitive in terms of TD and EC, it can guarantee the service quality of the IIoT system and has strong robustness and scalability.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"92 ","pages":"Article 101786"},"PeriodicalIF":8.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183628","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 evolutionary task scheduling algorithm using fuzzy fitness evaluation method for communication satellite network","authors":"Xuemei Jiang ,&nbsp;Yangyang Guo ,&nbsp;Yue Zhang ,&nbsp;Yanjie Song ,&nbsp;Witold Pedrycz ,&nbsp;Lining Xing","doi":"10.1016/j.swevo.2024.101830","DOIUrl":"10.1016/j.swevo.2024.101830","url":null,"abstract":"<div><div>Communications satellite network (CSN), as an integral component of the next generation of communication systems, has the capability to offer services globally. Data transmission in this network primarily relies on two modes: inter-satellite communication and satellite-to-ground station communication. The latter directly impacts the successful reception of data by users. However, due to resource and task limitations, finding a satisfactory solution poses a significant challenge. The communication satellite–ground station network scheduling problem (CS-GSNSP) aims to optimize CSN effectiveness by devising a plan that maximizes link construction time while considering constraints associated with satellite operation modes. The large number of tasks and numerous constraints in the problem result in a time-consuming evaluation of fitness function values. To address this issue, we propose a fuzzy fitness evaluation method (FFEM) that employs fuzzy or real evaluation methods based on individual similarity degrees. Additionally, we introduce an evolutionary algorithm based on FFEM, called evolutionary algorithm based on FFEM (FFEEA), for iteratively searching high-quality network construction schemes. In FFEEA, an adaptive crossover approach is used for efficient population search. Finally, extensive experiments are conducted to demonstrate that our proposed fuzzy fitness evaluation method and other improvement strategies significantly enhance satellite network service time. The study introduces a novel approach to enhance the efficiency of solving combinatorial optimization problems, such as CS-GSNSP, by mitigating the complexity associated with fitness evaluation.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"92 ","pages":"Article 101830"},"PeriodicalIF":8.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143182971","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 dynamic constrained multi-objective task assignment for heterogeneous crowdsensing","authors":"Jianjiao Ji ,&nbsp;Yinan Guo ,&nbsp;Wentao Wang ,&nbsp;Xiao Yang ,&nbsp;Dunwei Gong","doi":"10.1016/j.swevo.2024.101788","DOIUrl":"10.1016/j.swevo.2024.101788","url":null,"abstract":"<div><div>Efficient task allocation is a crucial issue of Mobile crowdsensing (MCS). Generally, only homogeneous mobile users like human are selected as the participants, causing a difficulty to meet the spatiotemporal coverage demand on human-unreachable regions. To overcome this drawback, unmanned aerial vehicles are introduced to form heterogeneous MCS, which can be formulated into a dynamic constrained multi-objective task allocation model. Taking the maximum average sensing quality of all tasks and the maximum average remaining budget for each subtask as the optimization objectives, an improved decomposition-based multi-objective evolutionary algorithm is presented to find the optimal allocation scheme. Specifically, the problem is first decomposed into a set of dynamic constrained scalar subproblems. For each subproblem, a stochastic configuration network (SCN)-based initialization is developed to produce the promising population, in which SCNs learn the probabilities of mobile users being allocated to each task. Following that, a reinforcement learning-based autonomous evolutionary strategy is adopted to recommend the most appropriate solvers in terms of the state of current population. A hybrid population update mechanism is then employed to form the high-quality offspring, with the purpose of balancing the feasibility, convergence and diversity. The extensive experiments on 20 dynamic instances are conducted to demonstrate the effectiveness of proposed algorithm compared to other task allocation algorithms.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"92 ","pages":"Article 101788"},"PeriodicalIF":8.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143183629","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 spiking neural network based proximal policy optimization method for multi-point imaging mission scheduling of earth observation satellite","authors":"Wei Yao ,&nbsp;Xin Shen ,&nbsp;Guo Zhang ,&nbsp;Zezhong Lu ,&nbsp;Jiaying Wang ,&nbsp;Yanjie Song ,&nbsp;Zhiwei Li","doi":"10.1016/j.swevo.2025.101867","DOIUrl":"10.1016/j.swevo.2025.101867","url":null,"abstract":"<div><div>Owing to the escalating demand for earth observation, a solitary satellite will be required to undertake an expanded array of missions, thereby rendering the scheduling of multipoint earth observation satellite imaging missions increasingly intricate. Herein, we propose a proximal policy optimization (PPO) algorithm based on a spiking neural network (SNN) to solve the multipoint satellite mission scheduling problem. Initially, we preprocess the mission–transition time by incorporating satellite attitude constraints and conceptualize the mission planning process as a Markov decision. Then, our methodology integrates SNN with PPO to effectively handle a high-dimensional state space by leveraging temporal information–processing capabilities. The SNN-based actor-critic network is trained to enhance the scheduling policy via PPO. Our method exhibits superior performance across various satellite orbits, satellite attitude maneuver speeds, and mission scales. In comparison with heuristic methods and traditional reinforcement learning techniques, our method shows a swifter convergence rate and an increased success rate of observation, coupled with superior convergence speed, robustness, and stability.</div></div>","PeriodicalId":48682,"journal":{"name":"Swarm and Evolutionary Computation","volume":"94 ","pages":"Article 101867"},"PeriodicalIF":8.2,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143176221","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}