Search

Nature- and society-inspired metaheuristic algorithms have recently become the most promising technological model. To solve more complex optimization problems and complicated engineering applications, this paper proposes a new people duality psychological tendency and feedback mechanism-based Inherited Optimization Algorithm(IOA), which is inspired by people showing positive-negative duality cognitive tendency and adaptive feedback behavior when selecting information resources with different identity attributes. The IOA algorithm contains the symmetric two exploration phases. The exploitation phase adaptively regulates the dualistic psychological balance of people in inheriting the information resources with better existence value through a feedback regulation mechanism controlled by the profitability awareness to increase population diversity. This paper qualitatively and quantitatively evaluates the optimization performance of IOA on 84 benchmarks, including swarm convergence behavior, effectiveness, convergence, robustness, and significance. The scalability of the IOA is investigated using the CEC2017 suites. The algorithm performance in solving constrained optimization is verified on 8 engineering problems. All statistical results of the IOA are compared with the most promising 12 metaheuristics, which shows that the absolute computational efficiency of IOA on four types of functions is 95%, 96.67%, 80.95%, and 76.92%, respectively, the average rank (rank sum ratio) of IOA is 1.08 (1.19%) among the 13 algorithms, ranking first. The Wilcoxon signed rank test results on the CEC2017 suites show that IOA contains 1437 significance indicators out of 1440 comparisons, with the proportion of significant differences 99.79%, which suggests the proposed IOA maintains efficient search efficiency. [ABSTRACT FROM AUTHOR]

A meta-evolutionary framework called Differential Evolution Ensemble Designer (DEED) has been proposed in this paper to automate the design of DE ensemble algorithms. Given the design components of DE ensembles and a set of optimization problems, DEED evolves effective and robust DE ensemble designs. The design components of DE ensemble algorithms include population management, constituent algorithms in the ensemble, information mixing amongst the sub-populations in the ensemble and the numerical parameters associated with various aspects of the ensemble. DEED employs Dynamic Structured Grammatical Evolution (DSGE) as the meta-evolutionary algorithm. A Backus–Naur form (BNF) grammar has been developed in this paper to represent the design space of DE ensembles and used by DSGE to evolve DE ensemble designs. DEED has been employed to evolve DE ensemble designs for solving 30-dimensional CEC'17 benchmark functions. The evolved designs (both the best design as well as all the final evolved designs) have been validated on CEC'14 and CEC'17 functions at 10, 30 and 50 dimensions and on real-world numerical optimization problems in CEC'11 benchmark suite. The DEED evolved designs have also been tested against the state-of-the-art algorithm configurator - irace. The performance of DEED evolved ensemble designs have been observed to be very competitive against that of manually designed and tuned state-of-the-art DE ensemble algorithms in the literature. DEED has also been demonstrated to evolve both co-operative and competitive style DE ensembles. The simulation experiments demonstrate the effectiveness as well as robustness of the evolved ensemble designs and the reliability of DEED framework in consistently evolving effective DE ensemble designs. • An automated ensemble differential evolution (DE) designer named DEED is presented. • A meta-evolutionary approach to automatically choose ensemble DE design components. • Grammatical evolution forms the meta-evolutionary engine in DEED. • DEED can support researchers and practitioners in the ensemble DE design process. [ABSTRACT FROM AUTHOR]

Automatic fault diagnosis in many systems is performed via the analysis of vibration data in the time or frequency domain. In the literature, many approaches for extracting featured from signals for fault diagnosis have been proposed. In this paper, we apply a variety of transform functions including Fourier, Wavelet, etc. to extract features from the vibration data, which includes the statistical features and some features automatically extracted via Convolutional Neural Networks (CNNs). For each of these feature extraction approaches, a learning algorithm is trained to diagnose the faults and their results are aggregated in an ensemble machine learning algorithm. The weights of the base learner algorithms are optimized via an evolutionary algorithm to achieve the best-weighted voting scheme. The architecture of CNNs has a significant effect on the performance of the algorithm, thus, in this paper, an evolutionary algorithm is proposed to find the best architecture for CNNs in fault diagnosis. The CNNs are trained via gradient descent algorithms which suffer from getting stuck in local optima. To manage this, we propose an evolutionary algorithm that benefits from the speed of gradient descent and the global search of evolutionary algorithms. The proposed algorithm is tested on a number of benchmark problems and the experimental results are presented. • An ensemble algorithm is proposed for fault diagnosis. • Different feature extractors are exploited. • An evolutionary training algorithm is proposed. [ABSTRACT FROM AUTHOR]

The online bin packing problem is a well-known optimization challenge that finds application in a wide range of real-world scenarios. In the paper, we propose a novel algorithm called FuzzyPatternPack(FPP), which leverages fuzzy inference and pattern-based predictions of the distribution of item sizes in online bin packing. In comparison to traditional heuristics like BestFit(BF) and FirstFit(FF), as well as the more recent PatternPack(PaP) and ProfilePacking(PrP) algorithm based on online predictions, FPP demonstrates competitive and superior performance in solving various benchmark problems. Particularly, it excels in addressing problems with evolving distributions, making it a promising solution for real-world applications where the item sizes may change over time. This research unveils the promising potential of employing fuzzy logic to effectively address uncertainty in scheduling and planning problems. • Proposed algorithm utilizes pattern extraction and handles uncertainty effectively. • Employment of a fuzzy classification system adds flexibility in modeling ambiguity. • Superior performance and robustness were seen under evolving input distributions. [ABSTRACT FROM AUTHOR]

The multi-objective multi-task evolutionary algorithm is emerging in the field of evolutionary computating. A great deal of multi-objective multi-tasking algorithms have been proposed recently and proved to be superior in solving many problems. However, in the real world, there are numerous high-dimensional objective problems that need to be solved. With the increasing number of objectives, the slow convergence speed, high computational complexity and reduced population diversity will occur in the existing multi-objective multi-tasking algorithms. There is a growing need to study high-dimensional objective algorithms in multi-tasking environment. To fulfill this research gap, a novel many-objective multi-tasking evolutionary algorithm (MaMTO-ADE) is put forward in this paper. The reference points-based non-dominated sorting method is introduced, which guarantees the diversity of the population in high-dimensional space. And a new offspring generation strategy is proposed to accelerate the population convergence and enables the population to generate high-quality offspring. The performance of MaMTO-ADE is verified on the classical benchmark problems, and the experimental results emphasize the excellent competitiveness of MaMTO-ADE compared to other related algorithms. • A novel many-objective multitask framework is proposed to solve the high-dimension multi-objective problems. • This algorithm uses a new mixture probability model-based knowledge transfer method. • A novel offspring generation strategy is proposed. • The new individual selection method is used for inter-task crossover. [ABSTRACT FROM AUTHOR]

Reducing the number of lower-level function evaluations (LFEs) is crucial to designing a bilevel evolutionary algorithm (BLEA) for bilevel optimization. Direct neighbor solution transfer (DST) is regularly used in BLEA because it can improve the lower-level search efficiency and reduce the number of LFEs. But this method still faces the challenge of misleading searches to make the algorithm converge to the wrong optimum. This paper proposes a bilevel improved multi-operator differential evolution algorithm (BL-IMODE) for bilevel optimization, which uses the knowledge transfer and adaptive coordinate systems approach. Specifically, this paper proposes a lower-level search knowledge transfer (LLS-KT) and a lower-level (LL) optimal solution distribution information transfer (LLO-DIT) mechanism to implement knowledge transfer. In LLS-KT, transferring LL solutions to the LL optimizer is similar to DST, but the LLS-KT transfers multiple solutions and thus has less possibility of misleading searches than DST. Moreover, the LLO-DIT transfers the LL optimal solution distribution information to the LL optimizer for generating promising offspring solutions. Since this method does not directly transfer LL solutions, it avoids misleading searches. Additionally, the adaptive coordinate systems (ACS) approach is used in two-level optimizers to suit different fitness landscapes and further reduce the number of LFEs. In ACS, three differential evolution mutation operators are produced in two coordinate systems. An adaptive strategy is used to select the suitable differential evolution mutation operator to generate promising offspring solutions in the Eigen coordinate or original coordinate system. Experimental studies on three test suites benchmark problems for comparison with seven state-of-the-art algorithms are conducted. The results show BL-IMODE has a significant advantage over the comparative algorithms. Finally, BL-IMODE is applied to tackle the inverse optimal control problem. • We propose a bilevel improved multi-operator differential evolution algorithm. • We propose a knowledge transfer approach to avoid misleading search. • We propose an adaptive coordinate systems approach to improve search efficiency. • We outperform all existing approaches in most test instances. [ABSTRACT FROM AUTHOR]

Unlike standard Reinforcement Learning (RL) model, many real-world tasks are non-Markovian, which requires long-term memory and dependency. Hence solving a non-Markovian task is more difficult and challenging than solving a Markovian one. In this paper, we propose a novel RL approach for training non-Markovian tasks expressed in temporal logic LTL f (Linear Temporal Logic over Finite Traces). To this end, an encoding of linear complexity from LTL f into MDPs (Markov Decision Processes) is introduced in order to take advantage of advanced RL algorithms. We further propose an experience classification method based on the automaton structure (theoretically equivalent to LTL f specifications). An automatic reward shaping technique and a prioritized experience replay mechanism are developed to cooperate with the classification method to improve the performance of RL algorithms. We provide empirical evaluations on two widely used benchmark problems, Waterworld and Cartpole , both augmented with complex non-Markovian tasks. The evaluations are conducted with respect to the metrics of training speed, policy quality, convergence rates, computational efficiency and scalability etc. The experimental results show that our approach achieves superior performance over other relevant studies, specially with an average improvement of 133% in convergence rate and a reduction of 11% in training time. • Solving non-Markovian tasks is a challenging and practically research problem. • Reinforcement learning can solve Markovian tasks transformed from non-Markovian tasks. • Prioritized experience replay and reward shaping improve performance in training non-Markovian tasks. [ABSTRACT FROM AUTHOR]

The deviation of micromachined products from their intended design poses a significant challenge for designers. To address this issue, numerical simulations can be employed for predicting such deviations. In this research paper, we present a methodology aimed at preserving accuracy in micromachining products resulting from topology optimization, particularly when intricate geometric features are involved. The accuracy of final products can be difficult to achieve due to the inherent limitations of microfabrication processes, specifically in lithography and etching. To overcome these limitations, our proposed methodology combines density filters and morphological operators to generate lithography masks that closely align with the surface of micromachined silicon. We have validated the efficacy of this methodology by selecting a benchmark problem focused on maximizing the stiffness of a cantilever beam across various volume fractions. The results demonstrate a significant improvement in accuracy while maintaining adherence to the desired design specifications. Additionally, we have conducted simulations of surface micromachining with photolithography to further substantiate the effectiveness of our proposed method in achieving enhanced accuracy for optimized topological structures. These findings hold practical implications for the design of Micro Electronic Mechanical Systems (MEMS) and other similar applications where accuracy is of paramount importance. By providing a means to compensate for the limitations inherent in microfabrication processes, our proposed methodology can contribute to the production of highly accurate and topologically optimized products. • Topology optimization performed with a simulated annealing with crystallization heuristic. • Topology optimization with fabrication constraints. • Topology optimization with density filters and morphological operators. [ABSTRACT FROM AUTHOR]

• An improved Multi-Verse Optimizer named COMVO is proposed. • The modified normalized inflation rate is adjusted to increase the exploration ability of the universe individuals. • The personal best-known universe is introduced to provide the desired search direction. • The comprehensive opposition learning is selected to enhance the exploration accuracy. • COMVO combines a new constraint-handling technique to solve the hydro-thermal-wind hybrid systems. In this paper, a Comprehensive Opposition Multi-Verse Optimizer (COMVO) is proposed to find the optimal solution to the scheduling problem of hybrid hydro-thermal-wind systems. In COMVO, three improvement mechanisms are introduced to enhance the search performance of the original MVO. The proposed method is analyzed and statistically tested against existing state-of-the-art methods in 23 classic benchmark problems and CEC2009 test suite, which demonstrate the superiority of the proposed method. Meanwhile, a constraint-handling technique is developed to solve the complex coupling constraints in the HTW problem and is finally evaluated on a 4 hydro-3 thermal-2 wind system. The results of the comparative analyses show that the proposed method is able to achieve lower costs and fewer emissions than the existing popular methods. In addition, the proposed method reduces emissions by 54%, 81%, 3%, 44%, 17% and 5% in the mean value metrics compared to DA, FFA, GWO, HS, MFO and MVO at 2% wind penetration rate, which fully demonstrates the practicability of the proposed methodology and constraint-handling technique. [ABSTRACT FROM AUTHOR]

The smooth execution of tasks in the cloud manufacturing (CMfg) process can be significantly affected by various real-world uncertainties, resulting in inefficiencies or even failures of the composite manufacturing service (CMS). Nevertheless, current research primarily concentrates on post-exception emergency handling methods, heavily dependent on the planned CMS and real-time status of manufacturing resources, often leading to suboptimal temporary adjustments or even the inability to find a viable solution. Hence, there is an urgent necessity to enhance the robustness of the planned CMS during the service composition and optimal selection (SCOS) phase. In pursuit of this goal, the paper initially presents an adaptive robust service composition and optimization selection (ARSCOS) model. This model dynamically determines whether to await exception resolution or invoke an alternative service, considering the circumstances when a service exception arises and employing task delay time to characterize the robustness of a specific CMS. Subsequently, an enhanced multi-objective artificial hummingbird algorithm (EMOAHA) is devised for this model. Finally, the efficacy of EMOAHA is demonstrated by applying it to nine different scales of civilian unmanned aerial vehicle manufacturing SCOS problems and 17 benchmark functions. The results illustrate that the proposed EMOAHA exhibits superior convergence and diversity in comparison to the other 11 classical intelligent optimization algorithms. [ABSTRACT FROM AUTHOR]

Class imbalance (CI) in classification problems arises when the number of observations belonging to one class is lower than the other. Ensemble learning combines multiple models to obtain a robust model and has been prominently used with data augmentation methods to address class imbalance problems. In the last decade, a number of strategies have been added to enhance ensemble learning and data augmentation methods, along with new methods such as generative adversarial networks (GANs). A combination of these has been applied in many studies, and the evaluation of different combinations would enable a better understanding and guidance for different application domains. In this paper, we present a computational study to evaluate data augmentation and ensemble learning methods used to address prominent benchmark CI problems. We present a general framework that evaluates 9 data augmentation and 9 ensemble learning methods for CI problems. Our objective is to identify the most effective combination for improving classification performance on imbalanced datasets. The results indicate that combinations of data augmentation methods with ensemble learning can significantly improve classification performance on imbalanced datasets. We find that traditional data augmentation methods such as the synthetic minority oversampling technique (SMOTE) and random oversampling (ROS) are not only better in performance for selected CI problems, but also computationally less expensive than GANs. Our study is vital for the development of novel models for handling imbalanced datasets. • Class imbalance (CI) in classification problems arises one class is lower than the other classes. • We present a computational review to evaluate data augmentation and ensemble learning methods for CI problems. • We propose a general framework that evaluates 10 data augmentation and 10 ensemble learning methods for CI problems. • Our objective is to identify the most effective combination for improving classification performance on imbalanced datasets. • The results indicate that the combinations can significantly improve classification performance on imbalanced datasets. [ABSTRACT FROM AUTHOR]

When the evolutionary algorithm is applied to handle high-dimensional expensive multi-objective optimization problems (MOPs), population evolution is crucial since it controls exploration and exploitation and decides if promising candidate solutions could be generated. However, little attention has been paid to this issue, evolution operators based on Genetic Algorithm (GA) and Differential Evolution (DE) are still the two most common approaches, whose convergence on high-dimensional MOPs with a limited number of fitness evaluations remains challenging. In this paper, we propose a decomposition-based local learning strategy to accelerate convergence in the high-dimensional search space of MOPs. Specifically, an individual is updated by learning from one of the best solutions of its corresponding local area based on the multi-objective decomposition approach. Accordingly, a surrogate-assisted evolutionary algorithm is proposed for better solving expensive high-dimensional MOPs. Experimental studies on MOPs with up to 100 decision variables and with 300 fitness evaluations demonstrate the effectiveness of the proposed method. Furthermore, we use the proposed method to solve a 2-objective shape design problem of the blended wing-body underwater glider (BWBUG) with 39 decision variables, and an impressive solution set is obtained. • Efficacy of evolution operators on high-dimensional EMOPs is analyzed. • Propose a novel evolution strategy based on decomposition for high-dimensional EMOPs. • A surrogate-assisted evolutionary algorithm is accordingly developed. • Comprehensive experimental studies on benchmark problems and an engineering application. [ABSTRACT FROM AUTHOR]

Combined heat and power economic dispatch (CHPED) problem is one of the most widely handled, optimization problem by researchers in modern power systems. CHPED problem is a complicated, non-continuous, and non-convex optimization problem due to the constraints. Moreover, considering the valve-point loading effect (VPLE), transmission losses (TLs), and prohibited operating zones (POZs) of power-only units as constraints, the complexity of CHPED problem increases. Therefore, a powerful optimization algorithm needs to be introduced to find global solution that meets all constraints. In this paper, a novel adaptive fitness-distance balance based artificial rabbits optimization (AFDB-ARO) is developed to solve CHPED problems. AFDB-based guiding mechanism was implemented to enhance the exploration capability of ARO and to strengthen exploitation-exploration balance. A comprehensive experimental study was realized to prove the performance of the proposed algorithm on the CHPED and benchmark problems. In experimental study between AFDB-ARO variants and ARO on 40 benchmark problems, according to Wilcoxon analysis results, all AFDB-ARO variants outperformed the base ARO, and the best AFDB-ARO variant won victory in 20 of 40 problem and achieved similar results in other 20 problem. In other experimental study, AFDB-ARO algorithm was implemented on the CHPED systems with 4-, 5-, 7-, 24-, 48-, 96-, and 192-units, and fifteen case studies were considered using these systems, VPLE, TLs, and POZs. One of the important points of this study was that POZs were considered for the first time in 96- and 192-units system. The results show that AFDB-ARO achieved the best optimal solution in ten of fifteen cases, was same in one case, and obtained almost same results in four cases compared to the literature. Moreover, the stability of the AFDB-ARO and base ARO algorithms in solving the CHPED problem were tested by performing stability analysis. While the mean success rate, mean iteration number, and mean search time were obtained 87.62%, 353.63, and 2.91 sec of AFDB-ARO, respectively, ARO managed to find the optimal solution in two cases. Thus, the superior performance of AFDB-ARO algorithm is confirmed by experimental studies and analysis against ARO algorithm. The source codes of the AFDB-ARO algorithm (proposed method) can be accessed at this link: https://www.mathworks.com/matlabcentral/fileexchange/136846-afdb-aro-an-improved-aro-algorithm-for-optimization-problem. [ABSTRACT FROM AUTHOR]

The main challenge in addressing constrained multi-objective optimization problems (CMOPs) lies in achieving a balance among convergence, diversity, and feasibility. To address this issue, this paper proposes a constrained multi-objective evolutionary algorithm with Pareto estimation via neural network named CMOEA-PeNN. In order to exploit and explore the decision space, the proposed algorithm employs a dual-population mechanism, which is trained with a self-organizing map (SOM). Firstly, the population distribution structure in decision space is mapped to objective space while preserving neighborhood information, and then the neuron weight is utilized to estimate the Pareto front (PF). Secondly, a novel approach is devised to preserve the feasibility of the population and enhance the estimation of the Pareto front by SOM. The achievement scalarizing function (ASF) is employed to choose promising solutions. This strategy could guide the population toward the optimal solution while exploring the small feasible regions. Finally, the performance of CMOEA-PeNN is compared with five state-of-the-art constrained multi-objective evolutionary algorithms (CMOEAs) on three widely used benchmark problems and a real-world problem. The experimental results show that CMOEA-PeNN could archive competitive performance in solving CMOPs. [ABSTRACT FROM AUTHOR]

This paper develops a new adaptive surrogate model method for structural reliability analysis, which is constructed by approximating the limit state function (LSF) using stochastic configuration network (SCN) ensemble model. The proposed ensemble model integrates three SCN base models and sets weights for each prediction result, reducing outliers and greatly improving the output accuracy and stability of the algorithm. In constructing a surrogate model for reliability analysis, a new adaptive sampling strategy is designed, including a novel learning function. The new strategy enables the most informative sample points are selected and significantly reduce the number of required sample point. Without further calls to the LSF, the structural failure probability is quickly evaluated using Monte Carlo simulation in terms of the constructed SCN ensemble surrogate model. Three benchmark problems, including a highly nonlinear problem with small failure probability and two multi-variable engineering structural problems, are used to verify the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

In many practical many-objective optimization problems, the computational cost of real evaluation is often high. To solve such problems, surrogate-assisted evolutionary algorithms (SAEAs) are effective methods that use low-cost surrogate models to approximate expensive real evaluation and select better-performing individuals for further evaluation. However, existing surrogate models accumulate prediction errors with increasing numbers of objectives, and the computational cost of constructing surrogate models also increases sharply. This paper proposes a composite surrogate-assisted evolutionary algorithm for expensive many-objective optimization. The crux of this method involves categorizing the surrogate models into two distinct types: a global model, constructed based on all costly real evaluations, and local models tailored to subsets derived from clustering all evaluations. Additionally, a new filling sampling criteria is proposed to guide the selection of solutions, considering the requirements of convergence, diversity, and uncertainty. Experimental results on multiple benchmark problems demonstrate that the proposed method is competitive compared to other popular surrogate-assisted evolutionary algorithms in solving expensive many-objective optimization problems within limited computational budgets. • Composite surrogate model combines global & local models for selection. • Filling sampling criterion guides solution selection. • The proposed algorithm is competitive in solving expensive MaOPs. • Experimental results on multiple benchmark problems demonstrate the effectiveness. [ABSTRACT FROM AUTHOR]

Expensive multi-objective combinatorial optimization problems have constraints in the number of objective function evaluations due to time, financial, or resource restrictions. As most combinatorial problems, they are subject to a high number of duplicated solutions. Given the fact that expensive environments limit the number of objective function evaluations, the existence of duplicated solutions heavily impacts the optimization process due to poor diversity and low convergence speed. This paper proposes the Novel-First Tabu Search, a greedy-strategy mechanism that uses Knowledge-Assisted Local Search methods to preserve the population diversity and increase the exploration and exploitation ability of MOEA/D. Experiments are conducted on constrained, unconstrained, multimodal, deceptive, linear, convex, and non-convex Pareto Front multi-objective combinatorial optimization benchmark problems. This paper also conducts an experiment on the real-world, expensive problem of Well Placement Optimization using a benchmark case based on the Namorado oil field, located in the Campos Basin, Brazil. The experimental results and performance comparison with state-of-the-art algorithms demonstrate that the proposed design significantly preserves diversity and increases convergence without violating the constraint in the number of objective function evaluations. • A greedy strategy that uses knowledge-assisted local search methods is developed. • The greedy strategy is combined with the MOEA/D algorithm. • The method is evaluated on five well-known multi-objective combinatorial problems. • The method is evaluated on the real-world problem of Well Placement Optimization. • It achieves faster convergence in comparison with state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

• A more structured metric is proposed to promote the balance between convergence and diversity in many-objective optimization. • A distance-based diversity maintenance scheme is used to each non-dominated front to maintain population diversity. • Make full use of the neighborhood information in mating selection, which significantly improves the efficiency of the algorithm. • At most one solution in each sub-region of current Pareto front is selected in selection operation, which improves the efficiency of the algorithm. Maintaining a good balance between convergence and diversity in many-objective optimization is a key challenge for most Pareto dominance-based multi-objective evolutionary algorithms. In most existing multi-objective evolutionary algorithms, a certain fixed metric is used in the selection operation, no matter how far the solutions are from the Pareto front. Such a selection scheme directly affects the performance of the algorithm, such as its convergence, diversity or computational complexity. In this paper, we use a more structured metric, termed augmented penalty boundary intersection, which acts differently on each of the non-dominated fronts in the selection operation, to balance convergence and diversity in many-objective optimization problems. In diversity maintenance, we apply a distance-based selection scheme to each non-dominated front. The performance of our proposed algorithm is evaluated on a variety of benchmark problems with 3 to 15 objectives and compared with five state-of-the-art multi-objective evolutionary algorithms. The empirical results demonstrate that our proposed algorithm has highly competitive performance on almost all test instances considered. Furthermore, the combination of a special mate selection scheme and a clustering-based selection scheme considerably reduces the computational complexity compared to most state-of-the-art multi-objective evolutionary algorithms. [ABSTRACT FROM AUTHOR]

A critical issue of constrained optimization evolutionary algorithms (COEAs) is how to balance the minimization of the objective function and the reduction of the constraint violation during the optimization process. This paper proposes a differential evolution (DE) algorithm with an adaptive penalty coefficient mechanism and a search history exploitation mechanism (called APCSH) to address this issue. Specifically, APCSH includes a learning stage and an evolving stage. In the learning stage, the correlation between the constraint violation and objective function values is calculated. Then, in the evolving stage, the correlation is used with a sigmoid function to determine the upper and lower bounds of the penalty coefficient, followed by a ranking strategy to assign the penalty coefficient to each solution. Furthermore, an exponentially weighted average method is used to update the search direction of solutions to promote convergence. Additionally, the scaling factor and crossover rate of DE are tuned automatically based on the successful search history. Many numerical experiments are conducted on 18 CEC2010 benchmark problems, 28 CEC2017 benchmark problems, and 4 heat pipe constraint optimization problems to demonstrate the advantages of the proposed algorithm over some state-of-the-art COEAs. [ABSTRACT FROM AUTHOR]

We introduce a multidimensional multiblock clustering (MDMBC) algorithm in this paper. MDMBC can generate overlapping clusters with similar values along clusters of dimensions. The parsimonious binary vector representation of multidimensional clusters lends itself to the application of efficient meta-heuristic optimization algorithms. In this paper, a hill-climbing (HC) greedy search algorithm has been presented that can be extended by several stochastic and population-based meta-heuristic frameworks. The benefits of the algorithm are demonstrated in a bi-clustering benchmark problem and in the analysis of the Leiden higher education ranking system, which measures the scientific performance of 903 institutions along four dimensions of 20 indicators representing publication output and collaboration in different scientific fields and time periods. • Multiblock clustering algorithm generates overlapping clusters along clusters of dimensions. • Generalized parsimonious binary vector-based representation for higher-dimensions. • The proposed algorithms are demonstrated via an open multidimensional dataset. [ABSTRACT FROM AUTHOR]

Abstract: This paper discusses the vehicle routing problem with simultaneous pickup and delivery (VRPSPD), in which both pickup and delivery tasks simultaneously occur at various customer locations. The objective is to design a set of minimum distance routes with the minimum number of vehicles satisfying all the customers. This paper proposes a heuristic algorithm consisting of a route construction procedure, a route improvement procedure, and a solution perturbation procedure. A new sweep-based route construction method is developed to generate better initial solutions. In the route improvement procedure, a series of inter- and intra-route improvement algorithms are applied. The perturbation procedure perturbs a solution by removing some routes and stops from the solution and reinserting them into the solution. The last procedure is used to escape from local optima. Computational experiments on various benchmark instances are performed to evaluate our algorithm against the previously proposed approaches. New best solutions for many benchmark instances are found. [Copyright &y& Elsevier]

• This paper addresses Multiobjective Vehicle Routing Problem with Time Windows. • A Nondominated Sorting Genetic Algorithm II approach is proposed for the problem. • The proposed approach uses objective-specific greedy variation operators. • Computational results demonstrate the effectiveness of the proposed approach. Vehicle routing problem with time windows (VRPTW) is a pivotal problem in logistics domain as it possesses multiobjective characteristics in real-world applications. Literature contains a general multiobjective VRPTW (MOVRPTW) with five objectives along with MOVRPTW benchmark instances that are derived from real-world data. In this paper, we have proposed a nondominated sorting genetic algorithm II (NSGA-II) based approach with objective-specific variation operators to address the MOVRPTW. In the proposed NSGA-II approach, the crossover and mutation operators are designed by exploiting the problem characteristics as well as the attributes of each objective. The performance of the proposed approach is evaluated on the standard benchmark instances of the problem and compared with the state-of-the-art approach available in literature. The computational results demonstrate the superiority of our approach over the state-of-the-art approach for the MOVRPTW. [ABSTRACT FROM AUTHOR]

• Propose a new representation scheme of hierarchical BRB using MTS. • Propose a MTS-BRB modeling procedure for a new BRB expert system. • Propose a MTS-BRB inferencing procedure for a new BRB expert system. • Propose a MTS-BRB learning procedure for a new BRB expert system. • Verify the MTS-BRB expert system using four benchmark problems. Belief rule-base (BRB) expert system is one of recognized and fast-growing approaches in the areas of complex problems modeling. However, the conventional BRB has to suffer from the combinatorial explosion problem since the number of rules in BRB expands exponentially with the number of attributes in complex problems, although many alternative techniques have been looked at with the purpose of downsizing BRB. Motivated by this challenge, in this paper, multilayer tree structure (MTS) is introduced for the first time to define hierarchical BRB, also known as MTS-BRB. MTS-BRB is able to overcome the combinatorial explosion problem of the conventional BRB. Thereafter, the additional modeling, inferencing, and learning procedures are proposed to create a self-organized MTS-BRB expert system. To demonstrate the development process and benefits of the MTS-BRB expert system, case studies including benchmark classification datasets and research and development (R&D) project risk assessment have been done. The comparative results showed that, in terms of modelling effectiveness and/or prediction accuracy, MTS-BRB expert system surpasses various existing, as well as traditional fuzzy system-related and machine learning-related methodologies. [ABSTRACT FROM AUTHOR]

Differential Evolution Algorithm (DE) is a commonly used stochastic search method for solving real-valued numerical optimization problems. Unfortunately, DE's problem solving success is very sensitive to the internal parameters of the artificial numerical genetic operators (i.e., mutation and crossover operators) used. Although several mutation and crossover methods have been developed for DE, there is not still an analytical method that can be used to select the most efficient mutation and crossover method while solving a problem with DE. Therefore, selection and parameter tuning processes of artificial numerical genetic operators used by DE are based on a trial-and-error process which is time consuming. The development of modern DE versions has been focused on developing fast, structurally simple and efficient genetic operators that are not sensitive to the initial values of their internal parameters. Problem solving successes of the Universal Differential Algorithms (uDE) are not sensitive to the structure and internal parameters of the related artificial numerical genetic operators used, unlike DE. In this paper a new uDE, Bezier Search Differential Evolution Algorithm , BeSD, has been proposed. BeSD's mutation and crossover operators are structurally simple, fast, unique and produce highly efficient trial patterns. BeSD utilizes a partially elitist unique mutation operator and a unique crossover operator. In this paper, the experiments were performed by using the 30 benchmark problems of CEC2014 with Dim =30, and one 3D viewshed problem as a real world application. The problem solving success of BeSD was statistically compared with five top-methods of CEC2014, i.e., CRMLSP, MVO, WA, SHADE and LSHADE by using Wilcoxon Signed Rank test. Statistical results exposed that BeSD's problem solving success is better than those of the comparison methods in general. • A new parameter-free crossover operator. • A new multi-component mutation operator. • A new Universal Differential Evolution Algorithm. • Real-valued evolutionary search for numerical function optimization. • Evolutionary 3D viewshed analysis. [ABSTRACT FROM AUTHOR]

Flower pollination algorithm (FPA) is a recent addition to the field of nature inspired computing. The algorithm has been inspired from the pollination process in flowers and has been applied to a large spectra of optimization problems. But it has certain drawbacks which prevents its applications as a standard algorithm. This paper proposes new variants of FPA employing new mutation operators, dynamic switching and improved local search. A comprehensive comparison of proposed algorithms has been done for different population sizes for optimizing seventeen benchmark problems. The best variant among these is adaptive-Lévy flower pollination algorithm (ALFPA) which has been further compared with the well-known algorithms like artificial bee colony (ABC), differential evolution (DE), firefly algorithm (FA), bat algorithm (BA) and grey wolf optimizer (GWO). Numerical results show that ALFPA gives superior performance for standard benchmark functions. The algorithm has also been subjected to statistical tests and again the performance is better than the other algorithms. [ABSTRACT FROM AUTHOR]

Classification tasks are of great importance in machine learning. However, class imbalance is a universal problem that needs to be solved in classification and can greatly affect the performance of machine learning classifiers. Developing from the basic belief rule base (BRB) system, the hierarchical belief rule-based system can integrate expert knowledge and has the potential to alleviate the negative effect of class imbalance. To utilize BRB to solve the imbalanced multi-classification task and avoid the combinational explosion problem, a novel hierarchical BRB structure based on the extreme gradient boosting (XGBoost) feature selection method, abbreviated as HFS-BRB is proposed in this paper in order to deal with any number of classes. In the hierarchical BRB structure, there is one main-BRB in the first level and several sub-BRBs in the second level. The XGBoost technique is used for feature selection in the modelling process of each abovementioned BRB model. The output of the main BRB represents the approximated classification between confusable classes. Then, these samples were transmitted to a certain sub-BRB for binary classification to make a precise prediction. Thus, a multi-classification problem can be transformed into several binary classification problems. The class imbalance is alleviated. To test the effectiveness of the proposed method, seven classical benchmark problems for imbalanced classification and a real asteroid orbit classification were performed. [ABSTRACT FROM AUTHOR]

The balance between multiple objectives and various constraints is the key to solving constrained multi-objective optimization problems (CMOPs). When dealing with CMOPs with complex feasible regions, some evolutionary algorithms suffer from great challenges in converging to the constrained Pareto front (CPF) with well-distributed feasible solutions. To address this issue, this paper proposes a dual-population based bidirectional coevolution algorithm, called DBC-CMOEA, which aims to converge to the CPF using promising solutions explored from both feasible and infeasible regions. To do so, DBC-CMOEA maintains two populations and an archive, where the dual-population is complementary in the search process and the archive is used to retain promising feasible and infeasible solutions, thus facilitating information exchange between these two populations. For updating the archive, a nondominated sorting procedure and an angle-based selected scheme are conducted to store infeasible and feasible solutions, as they can help to maintain the diversity of the search and find more feasible regions. To evolve the CPF from the bidirectional side of the feasible region, a novel mating selection strategy is used to choose appropriate mating parents. In comparison with some related constraint multi-objective optimization algorithms on a number of benchmark problems, experimental results show that the proposed algorithm performs better than the state-of-the-art constrained multi-objective evolutionary optimizers. • A cooperative dual-population algorithm is proposed for solving CMOPs. • Constraints are handled in three different ways to explore the search space. • A brand-new selection strategy is applied to select the mating parents. • A novel archive update strategy is designed to retain the promising information. [ABSTRACT FROM AUTHOR]

• An enhanced MOEA/D with a self-organizing collaborative scheme is proposed. • Different evolutionary strategies are collaboratively selected at different stages. • A neighborhood reconstruction scheme is designed to enhance information interaction. • Many benchmark and engineering problems are used for performance evaluation. The multi-objective evolutionary algorithm based on decomposition (MOEA/D) decomposes a multi-objective optimization problem (MOP) into multiple single-objective subproblems using an aggregation function and optimizes them together using a collaborative approach. MOEA/D exhibits extraordinary optimization ability in solving MOPs. However, the algorithm's performance is hampered by the single evolutionary operator, fixed control parameters, and fixed neighborhood architecture used in primal MOEA/D. To obtain the balance between convergence and diversity, a self-organizing collaborative scheme (SCS) is designed to achieve the best combination of differential evolution (DE) operator, control parameters, and neighborhood size in this paper. The SCS is integrated into MOEA/D to form the MOEA/D-SCS. In MOEA/D-SCS, firstly, the evolutionary process is separated into three stages: early, middle, and late, and different indicators are utilized to screen elite and inferior solutions for the early and late stages. Secondly, different evolutionary strategies are developed for the three stages and the various types of solutions to help boost the algorithm's efficiency. Then, to enhance convergence and diversity in different evolutionary stages, a neighborhood reconstruction strategy is proposed to assign different neighborhood sizes according to the neighborhood information. Finally, the capability of MOEA/D-SCS was tested using three standard test suites, ZDT, DTLZ, and UF, and a MOP of sewage treatment. The simulation results demonstrate that the proposed MOEA/D-SCS can achieve better optimization results and show stronger optimization ability in most cases compared to state-of-the-art evolutionary algorithms, such as MOEA/D-DE, MOEA/D-STM, MOEA/D-FRRMAB, MOEA/D-DRA, NSGA-II, and dMOPSO, especially in problems with multiple peaks, complex Pareto fronts, and practical engineering optimization problems. [ABSTRACT FROM AUTHOR]

Component Layout Optimization (CLO) is attracting growing attention as electronic equipment rapidly advances toward integration, miniaturization, and high function. Heat pipe-constrained Component Layout Optimization (HCLO) seeks to maximize the thermal performance of electronic equipment by optimizing its component layout, where heat dissipation relies heavily on heat pipes. HCLO poses its peculiar challenges with Many-dimensions, Many-constraints and Many-optima. Although many approaches have been proposed for various CLO applications, they do not consider heat pipe-related constraints and therefore underperform on HCLO. In this paper, we propose a Bi-population Clan-based Genetic Algorithm (BCGA) tailored for HCLO. BCGA introduces (i) A bi-population strategy to decompose HCLO into two interrelated sub-problems, which alleviates the challenge of Many-dimensions and Many-constraints; (ii) A clan-based framework inspired by human evolution to handle Many-optima, under which improved GA operators are also designed. The performance of BCGA is evaluated on a suite of HCLO benchmark problems of varying complexity. Compared with other algorithms recently proposed for CLO and HCLO, BCGA can deliver much more satisfying layout solutions within only half of the computation time. • It is a challenge to optimize component layout with heat dissipation on heat pipes. • Bi-population strategy can alleviate the intractability by decomposing HCLO. • Clan-based framework inspired by human evolution is proposed to tackle Many-optima. • BCGA outperforms the competing algorithms in both effectiveness and efficiency. [ABSTRACT FROM AUTHOR]

This paper proposes a hybrid version of marine predator algorithm (MPA) and naked mole-rat algorithm (NMRA) to aggregate the strengths of both algorithms. The new proposed algorithm is named as MpNMRA and designed to overcome the inherent drawbacks of MPA (slow exploitation) and NMRA (limited exploration). The algorithm adds the basic structure of MPA to the worker phase of NMRA, while keeping all the major parameters of both the algorithm. The major parameters of both the algorithms are subjected to four different mutation strategies namely, exponential, linear, simulated annealing and logarithmic mutation strategies. The concept of simulated annealing-based mutation is found to be best for most of the parameters, whereas in some cases exponentially decreasing weights provide better results. Leveraging on the best mutation strategies for all the parameters, the proposed MpNMRA is tested on CEC2005, CEC2014 and CEC 2019 benchmark problems. The experimental results demonstrate that MpNMRA provides best results when compared to other algorithms in the literature on higher-dimensional problems. This work also considers solving three real-world optimization problems and training of a multi-layer perceptron using the proposed algorithm. Statistical results obtained from Wilcoxon's rank-sum test, Freidman's test and computational complexity further proves that the proposed algorithm is highly efficient and provide superior results. 1 1 The link of source-code will be added after publishing. • Proposes new hybrid version of Marine predator naked mole-rat algorithm. • Four new mutation operators are added for adding the adaptive properties. • MpNMRA is tested on CEC 2005 and CEC 2014 benchmark functions. • MpNMRA is used to solve engineering design problems. • Experiments show that MpNMRA is efficient and effective. [ABSTRACT FROM AUTHOR]

Development of approaches for reducing the prediction error has been an active research field in collaborative filtering recommender systems since the accuracy of the prediction plays a crucial role in user purchase preferences. Unlike the conventional collaborative filtering methods which directly use the computed user-to-user similarity values, this paper presents a genetic algorithm approach for refining them before using in the prediction process. The approach was found to yield promising results according to the statistical analysis performed on a variety numbers of neighbours for various similarity metrics including Pearson’s Correlation, Extended Jaccard Coefficient and Vector Cosine Similarity along with a metric that assigns random weights to be used as a benchmark. Results show that the evolutionary approach has significantly reduced the prediction error using the evolved weights and Vector Cosine Similarity has shown the best performance. [ABSTRACT FROM AUTHOR]

• This paper proposes an enhanced whale optimizer (WOA) for global search. • Strategy of random replacement is created to enhance the convergence speed of WOA. • Strategy of double adaptive weight is introduced to improve the ability of global search of WOA. • The excellent performance is validated on benchmark problems and engineering problems. Whale optimization algorithm (WOA) is a newly developed meta-heuristic algorithm, which is mainly based on the predation behavior of humpback whales in the ocean. In this paper, a reinforced variant called RDWOA is proposed to alleviate the central shortcomings of the original method that converges slowly, and it is easy to fall into local optimum when dealing with multi-dimensional problems. Two strategies are introduced into the original WOA. One is the strategy of random spare or random replacement to enhance the convergence speed of this algorithm. The other method is the strategy of double adaptive weight, which is introduced to improve the exploratory searching trends during the early stages and exploitative behaviors in the later stages. The combination of the two strategies significantly improves the convergence speed and the overall search ability of the algorithm. The advantages of the proposed RDWOA are deeply analyzed and studied by using typical benchmark examples such as unimodal, multi-modal, and fixed multi-modal functions, and three famous engineering design problems. The experimental results show that the exploratory and exploitative tendencies of WOA and its convergence mode have been significantly improved. The RDWOA developed in this paper is a promising improved WOA variant, and it has better efficacy compared to other state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

• Bernstein polynomials based parameter-free crossover. • A new universal/parameter-free Differential Evolution. • Real-valued numerical function optimization. • Evolutionary Image Vectorization. • Evolutionary Digital Terrain Model Simplification. The standard Differential Evolution Algorithm (sDE) is a stochactic search method commonly used in evolutionary computing. The problem solving success of sDE is highly sensitive to the genetic operators used and the initial values of the parameters of these operators. Since a universal Differential Evolution Algorithm (uDE) is not sensitive to the structure and parameter values of the genetic operators used, it is parameter-free in practice and easier to control than sDE. uDE does not need a trial-and-error process when selecting the genetic operators and initial values of intrinsic parameter of related genetic operators to solve the problem, unlike the sDE. Therefore, the use and adaptation of a uDE to solve different types of numerical engineering problems is easy and time-consuming compared to sDE. In this paper, a new uDE, Bernstain-Search Differential Evolution Algorithm (BSD), is introduced. BSD is new and easily controllable, simple structured, non-recursive, highly efficient, fast and practically parameter-free uDE. BSD have a too feasible random crossover and mutation process and does not have a control-parameter setting process, contrary to sDE and its improved variants. In this paper, 30 benchmark problems of CEC'2014, 60 classic benchmark problems, image evolution problems for 12 test images and one Triangulated Irregular Network (TIN) refinement problem were used in the experiments performed to investigate the problem solving success of BSD, statistically. Four tested methods (i.e. , ABC, JADE, CUCKOO, WDE) were used in the conducted experiments. Problem solving successes of BSD and tested methods were statistically compared by using Wilcoxon Signed Rank Test piecewisely. Results obtained from the performed tests showed that in general, problem solving success of BSD is statistically better than the tested methods that have been used in this paper. [ABSTRACT FROM AUTHOR]

Moth–Flame Optimization (MFO) is widely utilized to solve optimization problems in different fields since it has a simple structure and easy implementation. However, MFO cannot effectively balance exploration and exploitation and often suffers from the lack of population diversity in the search process, especially in solving some complex engineering optimization problems. To overcome the above problems, in this paper, a multi-swarm improved moth–flame algorithm (MIMFO) is proposed. In MIMFO, firstly, the population is grouped and dynamically reorganized through chaotic grouping mechanism and dynamic regrouping mechanism, which can improve the grouping quality and diversity of the population. Secondly, spiral search and linear search are carried out for the two sub-swarms to improve the search efficiency and balance exploration and exploitation. In addition, Gaussian mutation is used to generate flame, which can accelerate convergence and enhance the exploration. The MIMFO is verified on 13 benchmark problems with 30, 500, 1000, 2000 dimensions and CEC 2014 test functions. The results show that the MIMFO is superior to other swarm intelligence algorithms and MFO variants in finding the global optimum and convergence performance. Finally, MIMFO is used to solve 57 engineering constraint optimization problems, and the results show that MIMFO can solve real-world engineering problems. • A multi-swarm improved moth–flame algorithm (MIMFO) is proposed. • Chaotic grouping and dynamic regrouping are used to improve population diversity. • Two sub-swarms are searched by spiral search and line search respectively. • Gaussian mutation is used to mutate the optimal flame. • 43 test problems and 57 engineering problems are used to evaluate the MIMFO. [ABSTRACT FROM AUTHOR]

• A velocity-based butterfly optimization algorithm is proposed. • A modified position updated equation is designed. • The refraction-based learning is introduced to enhance the diversity. • The extensive results show the competitiveness of the proposed algorithm. Throughout the last decade, high-dimensional function optimization problems have received substantial research interest in the field of intelligence computation. Butterfly optimization algorithm is a new meta-heuristic technique that has been proven to be more competitive than other optimization methods on low- dimensional problems. The basic butterfly optimization algorithm and its variants are not used to solve high-dimensional problems. Therefore, this paper develops an improved version of butterfly optimization algorithm to deal with the high- dimensional optimization problems. Inspired by particle swarm optimization, a modified position updated equation is designed by introducing velocity item and memory item in the local search phase to guide the search for candidate individuals. Moreover, a novel refraction-based learning strategy is introduced into butterfly optimization algorithm to effectively enhance diversity and exploration. The effectiveness of the proposed algorithm is tested on 40 benchmark problems with 100, 1000, 10,000 dimensions. The statistical results show that our algorithm has better performance than the basic butterfly optimization algorithm, its variants, and other population-based approaches to deal with the high-dimensional optimization problems. Finally, the high-dimensional feature selection and fault identification of wind turbine problems are solved and the comparisons show that the proposed algorithm outperforms better than most methods in terms of classification accuracy and number of the optimal feature subset. [ABSTRACT FROM AUTHOR]

Since the multi-objective ant colony optimization algorithm consumes a massive cost of time and computation resources, improving its convergence performance is essential. This paper proposes a historical experience-guided pheromone updating approach to improve the efficiency and enhance the optimization quality, which uses a learning automata to choose suitable pheromone updating methods adaptively according to the searching history. The learning automata performs different kinds of pheromone updating schemes, including a novel intragroup evolutionary information-guided strategy and two widely accepted strategies. A comparative experiment tests the proposed algorithm on multi-objective benchmark traveling salesman problems with a high-dimensional search space, as well as the primer design problem, which is practical in the biology area. The experimental results indicate that the proposed algorithm has a competitive performance on both the benchmark problems and the practical problem. • An intragroup evolutionary information-guided pheromone updating strategy. • A learning automata-based adaptive pheromone updating strategy. • Comparison experiments on benchmark traveling salesman problems (TSPs). • The proposed algorithm is used in solving the degenerate primer design problem. [ABSTRACT FROM AUTHOR]

• First attempt is made on Wiener-Hammerstein model estimation by the proposed method. • Simulations carried out on numerical examples and also on benchmark plants. • Various performance metrics are used to measure the estimated parameter accuracy. • The result of proposed method is compared with other employed benchmark algorithms. • Consistency of results obtained by the adopted methods is verified by Hypothesis test. Block oriented models are recurrently employed to describe the dynamic characteristics of nonlinear systems. This paper adopts an efficacious block-oriented model called cascaded Wiener-Hammerstein (W-H) structure to model a nonlinear system. The parameters of the W-H model are identified using a nature-inspired evolution algorithm-supported Kalman filter (KF). In the proposed technique, the utility of the employed optimisation algorithm named Harris Hawks optimiser (HHO) is to acquire the optimal global solution of initial state KF parameters for the W-H system identification problem. Then, the conventional KF algorithm utilises these optimised KF parameters to beget the near-global estimated W-H system parameters. The effectiveness of the proposed HHO-based KF algorithm is verified through a comparison study with other variants of particle swarm optimisation (PSO) and harmony search (HS) algorithms-supported KF methods on two numerical W-H problems with different nonlinearity levels and two practical benchmark plants, namely, real electronic diode switching network and flutter clearance of F-18 systems research aircraft. The simulation results ensure that the proposed HHO-KF approach offers significantly improved results over other competing methods in terms of various well known standard metrics. The hypothesis test verifies the stability and consistency of the proposed identification method. Moreover, the proposed approach provides a faster convergence rate and the lowest estimation precision compared with other reported techniques. [ABSTRACT FROM AUTHOR]

Solving constrained optimization problems (COPs) has been gathering attention from many researchers. In this paper, we defined the best fitness value among feasible solutions in current population as gbest . Then, we converted the original COPs to multi-objective optimization problems (MOPs) with one constraint. The constraint set the function value f ( x ) should be less than or equal to gbest ; the objectives are the constraints in COPs. A reverse comparison strategy based on multi-objective dominance concept is proposed. Compared with usual strategies, the innovation strategy cuts off the worse solutions with smaller fitness value regardless of its constraints violation. Differential evolution (DE) algorithm is used as a solver to search for the global optimum. The method is called multi-objective optimization based reverse strategy with differential evolution algorithm (MRS-DE). The experimental results demonstrate that MRS-DE can achieve better performance on 22 classical benchmark functions compared with several state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Fruit fly optimization algorithm (FOA) is one of the recent evolutionary computation approaches. This paper presents an effective and improved FOA (IFOA) for optimizing numerical functions and solving joint replenishment problems (JRPs). In the proposed IFOA, a new method of maintaining the population diversity is developed to enhance the exploration ability. Fruit flies with better fitness values use vision to fly toward a new location, and the others fly randomly in initial search space based on swarm collaboration. In addition, a new parameter to avoid the acquisition of local optimal solution is introduced to implement intelligent searching. Random perturbation is added to the updated initial location to jump out of the local optimum. Comparisons are carried out using 18 benchmark functions to verify the performance of the IFOA. Experimental results show that IFOA has better comprehensive performance than the original FOA, differential evolution algorithm, and particle swarm optimization algorithm. The IFOA is also utilized to solve the typical JRPs that have been proven as non-deterministic polynomial hard problems. Comparative examples reveal that the proposed IFOA can find better solutions than the current best algorithm; thus, it is a potential tool for various complex optimization problems. [ABSTRACT FROM AUTHOR]

High utility itemset mining problem involves the use of internal and external utilities of items (such as profits, margins) to discover interesting patterns from a given transactional database. It is an extension of the basic frequent itemset mining problem and is proven to be considerably hard and intractable. This is due to the lack of inherent structural properties of high utility itemsets that can be exploited. Several heuristic methods have been suggested in the literature to limit the large search space. This paper aims to improve the state-of-the-art and proposes a high utility mining method that employs novel pruning strategies. The utility of the proposed method is demonstrated through rigorous experimentation on several real and synthetic benchmark sparse and dense datasets. A comparative evaluation of the method against a state-of-the-art method is also presented. Our experimental results reveal that the proposed method is very effective in pruning unpromising candidates, especially for sparse transactional databases. [ABSTRACT FROM AUTHOR]

Sine cosine algorithm (SCA) is a recently developed and widely used metaheuristic to perform global optimization tasks. Due to its simplicity in structure and reasonable performance, it has been utilized to solve several real-world applications. This paper proposes an alternate version of the SCA by adopting the greedy approach of search, crossover and exponentially decreased transition control parameter to overcome the issues of low exploitation, insufficient diversity and premature convergence. The proposed algorithm, called ECr-SCA, is validated and compared with the original SCA using computational time, diversity, performance index, statistical and convergence analysis on a set of 23 standard benchmark problems. Later, the proposed ECr-SCA is compared with seventeen other algorithms including improved versions of the SCA and state-of-the-art algorithms. Furthermore, the ECr-SCA is used to train multi-layer perceptron and the results are compared with variants of SCA and other metaheuristics. Overall comparison based on several different metrics illustrates the significant improvement in the search strategy of the SCA by the proposal of the ECr-SCA. • The enhanced sine cosine algorithm (ECr-SCA) is proposed for global optimization. • A greedy search and modified transition parameter are used to increase exploitation. • The multi-parent crossover is embedded to enhance diversity and global search. • The excellent performance of the ECr-SCA is validated on benchmark problems. • The ECr-SCA is also used to train multilayer perceptron. [ABSTRACT FROM AUTHOR]

One of the key issues in solving constrained multi-objective optimization problems (CMOPs) is balancing the three indicators of convergence, diversity, and feasibility. We believe at different stages of the evolution, different indicators should be emphasized. This paper proposes a two-stage constrained multi-objective evolutionary algorithm (CMOEA) with different emphases on the three indicators. In Stage-I, the Pareto nondominated sorting and the unbiased two-objective model are used to evaluate the three indicators. The purpose of Stage-I is to obtain solutions with good distribution and to prevent the population from falling into local optima. After Stage-I, almost all of the individuals in the population are distributed in the vicinity of all feasible areas. The goal of Stage-II is to quickly converge the population to the Pareto front (PF). Thirty benchmark CMOPs and four real-world problems were used to evaluate the performance of our algorithm. Experimental results indicate that our approach achieved significantly better results or was at least competitive to the compared eight state-of-the-art CMOEAs on most of the benchmark problems. • A TSTI method is proposed for the CMOPs. • Two stages are used in TSTI with different goals. • TSTI obtains better results compared with six related methods. [ABSTRACT FROM AUTHOR]

The Grey wolf optimizer (GWO) is a recently introduced popular swarm-intelligence-based metaheuristic algorithm, compared to other algorithms, it has shown competitive performance. Despite its popularity, the conventional GWO suffers from slow convergence rate and tendency to stuck in local optima. Therefore, there is a chance of improvement in the search mechanism of the GWO through different operators. To improve the performance of the GWO, this paper proposes a new variant of the GWO called Mutation-driven Modified Grey wolf optimizer and denoted by MDM-GWO. The MDM-GWO combines a new update search mechanism, modified control parameter, mutation-driven scheme, and greedy approach of selection in the search procedure of the GWO. The performance of the proposed MDM-GWO is evaluated on 23 well-known standard benchmark problems of wide varieties of complexities and four real-world engineering design problems. The numerical results, statistical tests, convergence, and diversity curves, and comparisons among several algorithms show the superiority of the proposed MDM-GWO. • A new mutation-driven modified grey wolf optimizer (GWO) is proposed. • Search mechanism of GWO is modified using multi-parent crossover. • The control parameter ' a ' is redefined to be non-linearly decreasing with iterations. • Levy-flight based mutation scheme is used to enhance the global search ability. [ABSTRACT FROM AUTHOR]

As a popular global optimization algorithm, artificial bee colony (ABC) has strong search ability and simple concept. However, ABC has some deficiencies. The exploitation ability of ABC is not as strong as its exploration ability. The original roulette selection in the onlooker bee search will gradually lose its effect with increasing of iterations. In order to tackle the above problems, an efficient ABC based on adaptive search strategy and random grouping mechanism (called ASRGABC) is proposed in this paper. Firstly, an adaptive search strategy is designed by comparing the success rate of the current and previous iterations. According to the changes of the success rate, a suitable search strategy is adaptively selected. Then, a random grouping mechanism is proposed to replace the original roulette selection. The whole population is randomly divided into several groups. The onlooker bees are allowed to follow the best solution in each group. Based on the random grouping, the search strategy is modified. Thirdly, opposition-based learning is employed to enhance the scout bee phase. To verify the performance of ASRGABC, 22 classical benchmark problems and 28 CEC 2013 benchmark problems are tested. Experimental results show ASRGABC obtains better performance than thirteen other ABC variants. • An adaptive search strategy is designed by monitoring the changes of the success rate. • A random grouping mechanism is proposed to replace the original roulette selection. • Opposition-based learning is employed to enhance the scout bee phase. • Experiments are verified on 22 classical problems and 28 CEC 2013 complex problems. • The proposed approach obtains competitive results when compared with thirteen other ABCs. [ABSTRACT FROM AUTHOR]

• We are the first to address the NP-hard Budgeted Maximum Coverage Problem. • We propose an algorithm that combines reinforcement learning with local search. • We introduce the first set of benchmark instances for the problem. • The proposed algorithm outperforms the comparison algorithms and the CPLEX solver. The Budgeted Maximum Coverage Problem (BMCP) is a general model with a number of real-world applications. Given n elements with nonnegative profits, m subsets of elements with nonnegative weights and a total budget, the BMCP aims to select some subsets such that the total weight of the selected subsets does not exceed the budget, while the total profit of the associated elements is maximized. BMCP is NP-hard and thus computationally challenging. We investigate for the first time an effective practical algorithm for solving this problem, which combines reinforcement learning and local search. The algorithm iterates through two distinct phases, namely a tabu search phase and a probability learning based perturbation phase. To assess the effectiveness of the proposed algorithm, we show computational results on a set of 30 benchmark instances introduced in this paper and present comparative studies with respect to the approximation algorithm, the genetic algorithm and the CPLEX solver. [ABSTRACT FROM AUTHOR]

• A performance-based algorithm (HGS) is proposed for global search and optimization in real world. • HGS simulates the logic of the collaborative interactions based on individual hunger. • The extensive results on benchmark problems and real datasets have been investigated. • The proposed HGS is applied to engineering optimization to reduce the consumption. A recent set of overused population-based methods have been published in recent years. Despite their popularity, most of them have uncertain, immature performance, partially done verifications, similar overused metaphors, similar immature exploration and exploitation components and operations, and an insecure tradeoff between exploration and exploitation trends in most of the new real-world cases. Therefore, all users need to extensively modify and adjust their operations based on main evolutionary methods to reach faster convergence, more stable balance, and high-quality results. To move the optimization community one step ahead toward more focus on performance rather than change of metaphor, a general-purpose population-based optimization technique called Hunger Games Search (HGS) is proposed in this research with a simple structure, special stability features and very competitive performance to realize the solutions of both constrained and unconstrained problems more effectively. The proposed HGS is designed according to the hunger-driven activities and behavioural choice of animals. This dynamic, fitness-wise search method follows a simple concept of "Hunger" as the most crucial homeostatic motivation and reason for behaviours, decisions, and actions in the life of all animals to make the process of optimization more understandable and consistent for new users and decision-makers. The Hunger Games Search incorporates the concept of hunger into the feature process; in other words, an adaptive weight based on the concept of hunger is designed and employed to simulate the effect of hunger on each search step. It follows the computationally logical rules (games) utilized by almost all animals and these rival activities and games are often adaptive evolutionary by securing higher chances of survival and food acquisition. This method's main feature is its dynamic nature, simple structure, and high performance in terms of convergence and acceptable quality of solutions, proving to be more efficient than the current optimization methods. The effectiveness of HGS was verified by comparing HGS with a comprehensive set of popular and advanced algorithms on 23 well-known optimization functions and the IEEE CEC 2014 benchmark test suite. Also, the HGS was applied to several engineering problems to demonstrate its applicability. The results validate the effectiveness of the proposed optimizer compared to popular essential optimizers, several advanced variants of the existing methods, and several CEC winners and powerful differential evolution (DE)-based methods abbreviated as LSHADE, SPS_L_SHADE_EIG, LSHADE_cnEpSi, SHADE, SADE, MPEDE, and JDE methods in handling many single-objective problems. We designed this open-source population-based method to be a standard tool for optimization in different areas of artificial intelligence and machine learning with several new exploratory and exploitative features, high performance, and high optimization capacity. The method is very flexible and scalable to be extended to fit more form of optimization cases in both structural aspects and application sides. This paper's source codes, supplementary files, Latex and office source files, sources of plots, a brief version and pseudocode, and an open-source software toolkit for solving optimization problems with Hunger Games Search and online web service for any question, feedback, suggestion, and idea on HGS algorithm will be available to the public at https://aliasgharheidari.com/HGS.html. [ABSTRACT FROM AUTHOR]

• Proposed novel Opposition-based Laplacian Equilibrium Optimizer. • Applied techniques improve macro and micro search abilities. • Performance is verified over a set of 23 benchmark problems. • Applied in image segmentation using Multi-level thresholding. This paper proposes a modified version of freshly developed Equilibrium Optimizer (EO) for segmentation of gray-scale images using multi-level thresholding. Laplace distribution based random walk is utilized to update the concentration of search agents around equilibrium candidates (best solution) towards to attain optimal position (equilibrium state) for achieving better diversification of search space. An opposition based learning (OBL) mechanism is then applied with hybridization of the varying acceleration coefficient to the best solution for accelerating exploitation at a later phase of each iteration. The performance of proposed Opposition-based Laplacian Equilibrium Optimizer (OB-L-EO) is validated using test suites containing benchmark problems of wide varieties of complexities. Various analyses are conducted including Wilcoxon ranksum test for statistical significance, convergence curves and distance between solution before and after applying modification strategies. Finally, the proposed OB-L-EO is employed for image segmentation by utilizing Otsu's interclass variance function to obtain optimum threshold values for image segmentation. The performance of the proposed algorithm is verified by determining mean value of interclass variance and peak signal to noise ratio (PSNR). The obtained results are then compared and analysed with other metaheuristics algorithms to show superiority of proposed OB-L-EO. [ABSTRACT FROM AUTHOR]

• A feedback-based prediction strategy based DMOEA namely MOEA/D-FPS is proposed. • A new method for decision variable classification is introduced. • Correction feedback and effectiveness feedback mechanisms are suggested. • MOEA/D-FPS is efficient at solving various dynamic multi-objective problems. Prediction methods are widely used to solve dynamic multi-objective optimization problems (DMOPs). The key to the success of prediction methods lies in the accurate tracking of the new location of the Pareto set (PS) or Pareto front (PF) in a new environment. To improve the prediction accuracy, this paper proposes a novel feedback-based prediction strategy (FPS), which consists of two feedback mechanisms, namely correction feedback (CF) and effectiveness feedback (EF). CF is used to correct an initial prediction model. When the environment changes, CF constructs a representative individual to reflect the characteristics of the current population. The predicted solution of this individual in the new environment is calculated based on the initial prediction model. Afterward, a step size exploration method based on variable classification is introduced to adaptively correct the prediction model. EF is applied to enhance the effectiveness of re-initialization in two stages. In the first stage, half of the individuals in the population are re-initialized based on the corrected prediction model. In the second stage, EF re-initializes the rest of the individuals in the population using two rounds of roulette method based on the re-initialization effectiveness feedback of the first stage. The proposed FPS is incorporated into a dynamic multi-objective optimization evolutionary algorithm (DMOEA) based on decomposition resulting in a new algorithm denoted as MOEA/D-FPS. MOEA/D-FPS is compared with six state-of-the-art DMOEAs on twenty-two different benchmark problems. The experimental results demonstrate the effectiveness and efficacy of MOEA/D-FPS in solving DMOPs. [ABSTRACT FROM AUTHOR]

• Composite mutation strategy (CMS) and restart strategy (RS) are embedded into SSA. • CMS is able to boost the exploitation and exploration trends of SSA. • RS is capable of assisting salps in getting away from local optimum. • The proposed method is verified on IEEE CEC2017 benchmarks. • The proposed method is also tested on IEEE CEC2011 and engineering problems. This research proposes a reinforced salp swarm algorithm (SSA) variant with an ensemble mutation strategy and a restart mechanism, which is named CMSRSSSA for short, to enhance exploration and exploitation capacity of SSA and conquer the restriction of a single search mechanism of the SSA in tackling continuous optimization problems. In this variant, an ensemble/composite mutation strategy (CMS) can boost the exploitation and exploration trends of SSA, as well as restart strategy (RS) is capable of assisting salps in getting away from local optimum. To investigate the performance of the proposed optimizer, firstly, IEEE CEC2017 benchmark problems are used to estimate the capability of the presented CMSRSSSA in solving continuous optimization problems in comparison to other advanced algorithms; furthermore, IEEE CEC2011 real-world benchmark problems and constrained engineering optimization problems are also utilized to assess the performance of CMSRSSSA for practical ideas. Experimental and statistical results reveal that the CMSRSSSA outperforms all the competitors, including winners of the related IEEE CEC competition; therefore, it will be able to be treated as a promising method in resolving both constrained and unconstrained optimization problems. For post-publication supports and guides on the idea of the paper, please be in touch with the hosting website: http://aliasgharheidari.com. [ABSTRACT FROM AUTHOR]

• A new multi-objective optimization method used OBL strategy, WOA and DE algorithms • It combines DE and the OBL to improve the performance of the WOA • The MWDEO results outperformed all other algorithms in most of the test problems • 32 multi-objective test problems are used in the experiments and CEC2017 problems The optimization problems are divided into a single objective and multi-objective. Single objective optimization has only one objective function; whereas, multi-objective optimization has multiple objective functions that generate the Pareto set; therefore, solving a multi-objective problem is a challenging problem. This paper presents a new multi-objective optimization method (called MWDEO) based on improved whale optimization algorithm (WOA) by combining the differential evolution (DE) algorithm and the opposition-based learning (OBL). The MWDEO uses the WOA to perform a global exploration, whereas DE is used to exploit the search space; while the OBL is applied to improve the exploration and exploitation by generating the opposite values. The proposed algorithm is evaluated using 32 multi-objective test problems besides a set of benchmark problems of CEC'2017. The experimental results are compared with nine state-of-the-art multi-objective methods. The analysis of the results showed that the proposed MWDEO outperformed all other algorithms in most of the test problems which indicates that the proposed MWDEO is competitive and effective in solving different types of multi-objective problems. [ABSTRACT FROM AUTHOR]