Search

Path planning is a fundamental function of mobile robots that have been nowadays used in a variety of areas. Evolutionary algorithms (EAs) have been experimentally showed to be efficient for the problem of mobile robot path planning, an NP-problem. However, there is no theoretical work about EAs on the problem. In this paper, to understand more about why EAs are efficient and to give some advice on how to design more efficient EAs for this problem, we perform the first theoretical analysis of the expected runtime of a simple EA called ( $1+1$ ) EA with three different mutations for the mobile robot path planning problem in two constructed environments. In addition, a new method of producing a path from an uncomplete path is proposed, in which a new Insert process is proposed. Meanwhile, an obvious disadvantage of one of these three mutations is discovered during the expected runtime analysis in the first environment. Based on this discovery, an improved version of this mutation is proposed. In the second constructed environment, the improved mutation is proved to be superior to the others. In detail, the ( $1+1$ ) EA with the improved mutation can find the shortest path in runtime $O(n^{5})$ , while the ( $1+1$ ) EA with any one of the other mutations may be trapped in local optima. Experiments have been performed on the two constructed environments and six randomly generated environments with different sizes and varying obstacle densities.

The control of soft elastomeric actuators via pneumatic systems presents challenges due to system nonlinearities and oscillatory behavior. The control of soft robotics is an underexplored field compared to the control of traditional robotics. This study explores evolutionary algorithms for auto-tuning Proportional-Integral-Derivative (PID) controllers in pneumatic soft robotics. Four optimization algorithms, Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Simulated Annealing (SA), and Surrogate-Based Optimization (SBO), were employed to optimize PID parameters for pneumatic pressure control. Each algorithm was tested with four objective functions: Integral Absolute Error (IAE), Integral Time Absolute Error (ITAE), Integral Square Error (ISE), and Integral Time Square Error (ITSE). Both simulated and experimental studies are conducted to evaluate these algorithms using a pneumatic system designed with affordable on/off valves controlled by Pulse Width Modulation (PWM). Key performance metrics were analyzed: rise time, settling time, overshoot, peak value, and peak time. Results indicate that PSO and GA offer faster response times and moderate overshoot, while SA provides minimal overshoot at the cost of slower response times. These findings can significantly contribute to the practical control of pneumatic systems in soft robotics, offering insights for future optimization and application development.

In environments rich in data, machine learning models often encounter challenges such as data sparsity and overfitting, primarily due to datasets with an excessive number of features. To address these issues, this paper introduces a novel feature selection method employing a Memetic Algorithm (MA) enhanced with a fuzzy fitness function. This method is articulated in three variations: the Fuzzy Fitness Memetic Algorithm with Tabu Search (FFMATS), the Fuzzy Fitness Memetic Algorithm with Hill Climbing (FFMAHC), and a hybrid that combines both techniques, each utilizing specific local search strategies to refine feature selection. When tested across 16 UCI datasets using four different classifiers, these algorithms not only demonstrated competitive accuracy but frequently outperformed existing methods. These results highlight the critical importance of customizing feature selection strategies to meet the specific needs of various datasets and classifiers, ultimately enhancing the practicality and effectiveness of machine learning models.

Permanent Magnet Synchronous Motor (PMSM) drives are widely used for motion control industrial applications and electrical vehicle powertrains, where they provide a good torque-to-weight ratio and a high dynamical performance. With the increasing usage of these machines, the demands on exploiting their abilities are also growing. Usual control techniques, such as field-oriented control (FOC), need some workaround to achieve the requested behavior, e.g., field-weakening, while keeping the constraints on the stator currents. Similarly, when applying the linear model predictive control, the linearization of the torque function and defined constraints lead to a loss of essential information and sub-optimal performance. That is the reason why the application of nonlinear theory is necessary. Nonlinear Model Predictive Control (NMPC) is a promising alternative to linear control methods. However, this approach has a major drawback in its computational demands. This paper presents a novel approach to the implementation of PMSMs’ NMPC. The proposed controller utilizes the native parallelism of population-based optimization methods and the supreme performance of field-programmable gate arrays to solve the nonlinear optimization problem in the time necessary for proper motor control. The paper presents the verification of the algorithm’s behavior both in simulation and laboratory experiments. The proposed controller’s behavior is compared to the standard control technique of FOC and linear MPC. The achieved results prove the superior quality of control performed by NMPC in comparison with FOC and LMPC. The controller was able to follow the Maximal Torque Per Ampere strategy without any supplementary algorithm, altogether with constraint handling.

Human Activity Recognition (HAR) is becoming increasingly important in the fast-evolving landscapes of wearable sensors, smart applications, and the Internet of Things (IoT) paradigms. HAR is rapidly gaining importance, especially in health monitoring, elderly and infant care, fitness tracking, and security. Machine learning (ML) and Deep Learning (DL) methods are used significantly for HAR problems. ML and DL methods often face four significant problems. Firstly, they lack the adaptability to evolve network architectures dynamically, which is vital in complex tasks like HAR. Secondly, classical ML and DL methods could fall short in comprehensively navigating potential solution spaces. Thirdly, finding optimal hyperparameters is a computationally expensive process, and lastly, expert knowledge is required to configure the hyperparameters. This paper proposes the NeuroHAR, a transformative neuroevolutionary method for HAR, to address these problems. NeuroHAR integrates feedforward deep neural networks (FDNN) with evolutionary algorithms. The highlights of NeuroHAR include its dynamic optimization of network architectures and hyperparameters. It is simple to configure and computationally efficient. Due to its adaptable design, it offers a more flexible and robust solution that handles task complexities better than traditional methods. Results are promising, which is evidence of the effectiveness of the proposed NeuroHAR, which outperformed the explicit contender, the state-of-the-art Grid Search approach. NeuroHAR and Grid Search evaluated 900 and 1080 models in Case I. Even with broader hyperparameter ranges in Case II, NeuroHAR still executed 900 models, whereas Grid Search would need over 2 billion models, which proves the computational efficiency of NeuroHAR. Additionally, for HARTH and HAR70Plus imbalanced datasets, the NeuroHAR model achieved a higher prediction accuracy of 89.91% versus Grid Search’s 84.04%. NeuroHAR can facilitate advanced monitoring and analytics, which are crucial for health monitoring, elderly care, and urban management powered by wearable sensors, IoT, and smart applications.

Expensive optimization problems are characterized by the significant amount of time and resources needed to determine the quality of potential solutions. This poses severe limitations for the application of metaheuristic optimization methods, such as evolutionary algorithms, as they usually require evaluating many candidate solutions to deliver satisfactory results. Surrogate model-based strategies have become a popular choice to tackle this type of problem. The key idea of these strategies is to build a model which can approximate and (partially) replace the mechanisms for assessing solution quality, such that the use of this less expensive alternative lowers the overall computational cost of the optimization process. This paper analyzes surrogate model-based strategies in the context of a specific, expensive, combinatorial optimization problem: the configuration of the gas distribution system for an electrostatic precipitator. Focusing on this relevant case study from industry, the aim of this paper is twofold: (i) to investigate the most suitable learning techniques for building the surrogate models and (ii) to explore the advantages of ensemble strategies allowing various surrogate models to collaborate during optimization. This contrasts with previous studies where a single, fixed modeling technique is adopted to address this problem. The experimental evaluation involves eight different learning techniques, three alternative ensemble strategies, and two reference approaches from the literature (previously used to tackle this specific problem). Our results reveal the best modeling techniques at the individual level, while highlighting clear benefits of the simultaneous exploitation of multiple surrogate models when facing this particular optimization challenge.

This paper addresses the challenging problem of Unit Commitment (UC), which involves the optimal scheduling of power generation units while adhering to numerous network operational constraints called security-constrained UC (SCUC). SCUC problem aims to minimize costs subject to turning on economically efficient generators and turning off expensive ones. These operational constraints include load balancing, voltage level at buses, minimum up and down time requirements, spinning reserve, and ramp up and down constraints. The SCUC problem, subject to these operational constraints, is a complex mixed-integer nonlinear problem (MINLP). There has been a growing interest in using evolutionary algorithms (EAs) to tackle large-scale multi-objective MINLP problems in recent two decades. This paper introduces a novel approach to address the SCUC problem, which is further complicated by including network constraints. They are pioneering the integration of single and multi-objective EAs to solve the SCUC problem while incorporating AC network constraints through hybrid binary and real coded operators. The development of an ensemble algorithm that combines mixed real and binary coded operators, extended by a bidirectional coevolutionary algorithm to tackle multi-objective SCUC problems. The paper implements a new formulation based on three conflicting objective functions: cost of energy supplied, startup and shutdown costs of generators, energy loss, and voltage deviation to solve the SCUC problem. Implementing a new formulation also addresses the solution of single and multi-objective SCUC problems using a combination of proposed technical and economic objective functions. The proposed algorithm is rigorously tested on a 10-unit IEEE RTS system and a 6-unit IEEE 30-bus test system, both with and without security constraints, addressing week-ahead and day-ahead SCUC scenarios. Simulation results show that the proposed algorithm finds near-global optimal solutions compared to other state-of-the-art EAs. Additionally, the research demonstrates the effectiveness of the proposed search operator by integrating it with a multi-objective coevolutionary algorithm driven by both feasible and infeasible solutions, showcasing superior performance in solving multi-objective SCUC problems. These results are compared with various recently implemented Multi-Objective Evolutionary Algorithms (MOEAs), demonstrating the superiority of

With the establishment of smart grids, there is a growing demand for metering electric meters in urban areas. Given the diverse sizes and delicate nature of these meters, traditional scheduling solutions are unable to cater to the multifaceted requirements of urban environments, meters loading, and subsequent logistics scheduling. This study presents an intelligent scheduling model for electric meters in an urban context, taking into account various constraints such as urban traffic congestion, three-dimensional packing of metering devices, delivery time windows, and heterogeneous vehicles. To solve this, we design an improved whale optimization algorithm using a hybrid multi-phase heuristic approach (IWOA-HMOHA). Simulation results show that compared with the traditional meter logistics scheduling strategy, the IWOA-HMOHA algorithm reduces the objective function by 5.4%~ 26.1% compared with other similar algorithms. In addition, compared with the traditional first-in-last-out cargo packing method, the vehicle space utilization rate is improved by 12.62%. The proposed models and algorithms demonstrate excellent adaptability to a range of urban constraints, offering valuable insights and a robust framework essential for the development of logistics solutions in urban.

Indoor communication and positioning are significant fields of applications for indoor Internet of Things (IoT) given the rapid growth of IoT in smart cities, smart grids, and smart industries. Visible light positioning (VLP) has become more and more attractive for researchers to provide indoor location-aware IoT services. Additionally, artificial intelligence (AI) has attracted considerable research effort to address the challenges in visible-light communication (VLC) systems. This is an emerging technology in next-generation wireless networks. However, despite the rapid progress, the use of AI in localization, navigation, and position estimation is still underexplored in VLC systems, and various research challenges are still open. This methodological review summarizes the research efforts regarding the use of AI in the field of VLP, to improve the position estimation accuracy in both two-dimensional (2D) and three-dimensional (3D) indoor IoT applications. This treatise also presents open issues and potential future directions for motivating further research in the field. Various databases were utilized in this paper: Scopus, Google Scholar, and IEEE Xplore; obtained 88 papers from 2017 to early 2023. Most (68%) of the AI articles in VLP systems are machine learning (ML) methods applied for localization and position estimation in VLC systems, while the other 32% of the research articles focussed on evolutionary algorithms. ML and evolutionary models may present limitations in terms of complexity and time-consuming nature but offer highly accurate, robust, reliable, and cost-effective results in terms of position estimation over conventional approaches.

In conventional optimization problems, it is assumed that all relevant parametric constraints remain stationary. In contrast, optimization problems encountered in practical applications are dynamic and supervened by uncertainties. The research community has evinced a keen interest in multi-population approaches combined with nature-inspired algorithms to manage dynamic optimization problems efficiently. Applying multi-population approaches to solve dynamic optimization problems engenders specific vital issues, such as reproducing sub-populations in new environments influenced by archival information. Moreover, over-partitioning the population may lead to aberrant utilization of computational resources among the sub-populations. These impediments are addressed using the proposed hybrid multi-population reinitialization strategy, which is a combination of distributed differential evolution algorithmic framework and re-initialization strategy. This scheme relies on simple reinitialization to surmount the dynamism. This framework was assessed on different instances in a moving peak benchmark problem, a proven benchmarking function in the domain of dynamic optimization. Furthermore, this study encompasses a comparative and statistical analysis to validate the effectiveness of the proposed approach in comparison to cutting-edge algorithms in solving dynamic optimization problems efficiently. The experimental results consistently show that the hybrid multi-population reinitialization strategy outperforms conventional Differential Evolution algorithms across various parameter configurations. This hybrid multi-population reinitialization strategy showcases its effectiveness in the successful handling of increased shift lengths and number of peaks, which are pivotal parameters in solving moving peak benchmark function.

The Binary Paint Shop Problem (BPSP) is a combinatorial optimization problem which draws inspiration from the automotive paint shop. Its binary nature, making it a good fit for Quadratic Unconstrained Binary Optimization (QUBO) solvers, has been well studied but its industrial applications are limited. In this paper, in order to expand the industrial applications, QUBO formulations for two generalizations of the BPSP, which are the Multi-Car Paint Shop Problem (MCPSP) and the Multi-Car Multi-Color Paint Shop Problem (MCMCPSP), are proposed. Given the multiple colors, the MCMCPSP is no longer natively binary which increases the problem size and introduces additional constraint factors in the QUBO formulation. Resulting QUBOs are solved using Scatter Search (SS). Furthermore, extensions of the SS that can exploit k-hot constrained structures within the formulations are proposed to compensate the additional complexity introduced by formulating non-binary problems into QUBO. Since no public benchmark database currently exists, random problem instances are generated. Viability of the proposed QUBO solving methods for the MCPSP and MCMCPSP, is highlighted through comparison with an integer-based Random Parallel Multi-start Tabu Search (RPMTS) and a greedy heuristic for the problems. The greedy heuristic has negligible computational requirements and therefore serves as a lower bound on the desired performance. The results for both problems show that better results can be obtained than the greedy heuristic and integer-based RPMTS, by using the novel k-hot extensions of the SS to solve the problems as QUBO.

Currently, almost all direct marketing activities take place virtually rather than in person, weakening interpersonal skills at an alarming pace. Furthermore, businesses have been striving to sense and foster the tendency of their clients to accept a marketing offer. The digital transformation and the increased virtual presence forced firms to seek novel marketing research approaches. This research aims at leveraging the power of telemarketing data in modeling the willingness of clients to make a term deposit and finding the most significant characteristics of the clients. Real-world data from a Portuguese bank and national socio-economic metrics are used to model the telemarketing decision-making process. This research makes two key contributions. First, propose a novel genetic algorithm-based classifier to select the best discriminating features and tune classifier parameters simultaneously. Second, build an explainable prediction model. The best-generated classification models were intensively validated using 50 times repeated 10-fold stratified cross-validation and the selected features have been analyzed. The models significantly outperform the related works in terms of class of interest accuracy, they attained an average of 89.07% and 0.059 in terms of geometric mean and type I error respectively. The model is expected to maximize the potential profit margin at the least possible cost and provide more insights to support marketing decision-making.

Automatic circle detection is one of the most important construction elements to design more complex industrial image tasks such as target detection, manufacture inspection and process control. Most of the literature concerning multi-circle detection can be categorized under deterministic or stochastic perspectives. Deterministic approaches combine geometric and histogram information to identify circle shapes. However, deterministic techniques are unable to detect circles in presence of noise, shape variance, or occlusion. On the other hand, stochastic methodologies such as metaheuristic algorithms have been also proposed as alternatives to deterministic approaches for circle detection. Although these techniques are in general more robust, they allow detecting only one circle per execution, translating the detection problem into a unimodal optimization problem. Under such conditions, it is necessary to execute multiple times the algorithm to identify all the circles contained in the image. In this paper, the multi-circle detection problem has been reformulated as a multimodal optimization problem. The proposed approach adopts the Multimodal Flower Pollination Algorithm (MFPA) to detect all the circular instances contained in the image. Experimental results suggest that the proposed approach significantly improves the multi-circle detection problem.

This study aimed to minimize the tumor cell population using minimal medicine for chemotherapy treatment, while maintaining the effector-immune cell population at a healthy threshold. Therefore, a mathematical model was developed in the form of ordinary differential equations (ODE), and the solution to the Multi-Objective Optimal Control Problem (MOOCP) was obtained using Multi-Objective Optimization algorithms. In this study, the interaction of the tumor cell and effector cell populations with chemotherapy was investigated using Pure MOOCP and Hybrid MOOCP methods. The handling of constraints and the Pontryagin Maximum Principle (PMP) differ among these methods. Swarm Intelligence (SI) and Evolutionary Algorithms (EA) were used to process the results of these methods. The numerical outcomes of SI and EA are displayed via the Pareto Optimal Front. In addition, the solutions from these algorithms were further analyzed using the Hypervolume Indicator. The findings of this study demonstrate that the Hybrid Method outperforms Pure MOOCP via Multi-Objective Differential Evolution (MODE). MODE produces a point on the Pareto Optimal Front with a minimal distance to the origin, where the distance represents the best solution.

This article investigates the application of swarm and evolutionary algorithms, namely the SOMA, DE, and GA, for optimizing the F-transform-based image compression. To do this, we introduce the cost function, evaluating the approximation of decompressed images to the original image, concerning the parameters that control the approximation quality of the F-transform. This function is then minimized by the selected algorithms to find optimal settings for image compression and decompression. We design experiments to compare the performance of the original F-transform method and the methods optimized by SOMA, DE, and GA on a dataset of 10 pictures. In all considered cases, the results obtained with the optimized method completely surpass those obtained by the original one. We also apply a statistical test (called Wilcoxon signed-rank test) for ranking the performance of selected algorithms in this issue. The results show that the SOMA and DE perform well in cases where the compressed image sizes are small. However, the GA algorithm shows outperformance in comparison with the others in more complicated cases where the compressed image size is bigger. The outperformance of the GA is in terms of decompression quality and computation time. Finally, we provide a visual comparison between the original F-transform-based method and the method optimized by the GA, tested on a $128\times 128$ picture. The decompressed image by the latter is much sharper and more detailed than that obtained by the former.

Hundreds of variants of Swarm Intelligence or Evolutionary Algorithms are proposed each year and numerous competitions and comparisons between algorithms may suggest rapid improvement in the field. However, such comparisons are often done between a limited number of methods and are based on averaged ranks of algorithms. This way they measure whether one method is on average ranked better than the others, without giving any information on how much improvement is in fact obtained. In this study we show a general comparison between 69 algorithms, starting from methods proposed in the 1960’s up to variants developed in the early 2020’s, on single-objective static numerical problems. Algorithms are compared on searching for a minimum of 30 different 50-dimensional mathematical functions, and on 22 real-world problems. We focus on the relative improvement achieved by various algorithms over a single-solution based method proposed in 1960 by Howard Rosenbrock. We find that the general improvement of Evolutionary Algorithms over Rosenbrock’s algorithm is relatively limited. It is high for the artificial benchmarks, for which many Evolutionary Algorithms find solutions 10 times closer to the global optimum in terms of fitness than Rosenbrock’s algorithm, but much lower for real-world problems. Improvement is also higher when performance averaged over many runs is compared, but lower when the best results from multiple runs are analyzed. In the last case, only the best Evolutionary Algorithms are able to find solutions of a “typical” real-world problem that are 2–3 times better in terms of fitness than those found by Rosenbrock’s algorithm. The relative improvement of recently proposed algorithms is not much better than the improvement achieved by algorithms proposed over a decade ago.

Multicriteria sorting involves assigning the objects of decisions (actions) into $a$ priori known ordered classes considering the preferences of a decision maker (DM). Two new multicriteria sorting methods were recently proposed by the authors. These methods are based on a novel approach called interval-based outranking which provides the methods with attractive practical and theoretical characteristics. However, as is well known, defining parameter values for methods based on the outranking approach is often very difficult. This difficulty arises not only from the large number of parameters and the DM’s lack of familiarity with them, but also from imperfectly known (even missing) information. Here, we address: i) how to elicit the parameter values of the two new methods, and ii) how to incorporate imperfect knowledge during the elicitation. We follow the preference disaggregation paradigm and use evolutionary algorithms to address it. Our proposal performs very well in a wide range of computational experiments. Interesting findings are: i) the method restores the assignment examples with high effectiveness using only three profiles in each limiting boundary or representative actions per class; and ii) the ability to appropriately assign unknown actions can be greatly improved by increasing the number of limiting profiles.

Feature selection has been considered as an effective method to solve imbalanced classification problems. It can be formulated as a multiobjective optimization problem (MOP) aiming to find a small feature subset while achieving a high classification accuracy. With traditional MOP, the focus is on deriving an optimal solution (i.e., a feature subset), while ignoring the diversity in solution space (e.g., there could exist multiple feature subsets achieving the same accuracy). Providing more options for feature selection would be beneficial since some features can be more difficult to obtain than others. In this work, we treat feature selection as a multimodal MOP (MMOP) whose goals are to find an excellent Pareto front in objective space and as many equivalent Pareto optimal solutions (feature subsets) as possible in feature space. Note that though several multimodal multiobjective evolutionary algorithms (MMEAs) have been proposed, their use of a convergence-first selection criterion could cause the loss of solution diversity in an objective and feature space. To address the issue, a novel competition-driven mechanism is designed to assist the existing multimodal MMEAs in locating more equivalent feature subsets and a desired Pareto front. The effectiveness of the proposed mechanism is first verified on all 22 MMOPs from CEC2019. Then, the proposed method is applied to feature selection in imbalanced classification problems and a real-world application, i.e., credit card fraud detection. Experimental results show that the proposed mechanism can not only provide more equivalent feature subsets but also improve classification accuracy. [ABSTRACT FROM AUTHOR]

Acquisition functions for surrogate-assisted many-objective optimization require a delicate balance between convergence and diversity. However, the conflicting nature between many objectives may lead to an imbalance between exploration and exploitation, resulting in a low efficiency in search for a set of optimal solutions that can well balance convergence and diversity. To meet this challenge, we propose an adaptive model management strategy assisted by two sets of reference vectors, one set of adaptive reference vectors accounting for convergence while the other set of fixed reference vectors for diversity. Specifically, we first propose a new acquisition function that calculates an amplified upper confidence bound (AUCB). Two optimization processes are performed in parallel to optimize the acquisition function, each based on one of the two sets of reference vectors. Then, we select one promising candidate solution according to diversity or convergence from the nondominated solutions obtained by the two optimization processes. The experimental results on four suites of test functions as well as six real-world application problems demonstrate the competitive performance of the proposed reference vector-assisted adaptive model management strategy, in comparison with seven state-of-the-art surrogate-assisted evolutionary algorithms (SAEAs). [ABSTRACT FROM AUTHOR]

This work presents an automated design methodology for time-mode proportional and integral (PI) controllers aimed for an on-chip switched-capacitor (SC) dc/dc converter system. The basis of this design is the use of evolutionary optimization algorithms to find the near-optimal set of sizings for the time-mode PI controller. It is motivated due to the difficulty faced when tuning the controller parameters at a circuit level, which arise as a result of the presence of modeling inaccuracies and the small region for the linearized model where it is defined. Moreover, this design proposes the required modifications for the original design presented for the inductor-based dc/dc converter. These modifications are necessary to operate the SC dc/dc converter in slow switching limit (SSL). The addition of a pulse-width-modulated (PWM)-to-pulse frequency-modulated (PFM) conversion block is presented and elaborated in this article. The controller is codesigned using the differential evolution algorithm for the circuit level implementation to mitigate the issues prior mentioned. The optimized controller is then tested in a simulation environment using TSMC $0.18~\mu \text{m}$ technology. The results of the optimized controller were superior to those of a conventional controller. The optimized system achieved an overall efficiency of 79.1%. [ABSTRACT FROM AUTHOR]

Nonlinear equation systems (NESs) are ubiquitous, and solving them is an important yet challenging task in numerical computation. Most of the NESs usually contain multiple roots. To locate multiple roots, the use of evolutionary algorithms attracts more attention recently. For the purpose of finding different roots within a limited computational budget in a single run simultaneously, in this article, a hybrid niching-based differential evolution with two archives, referred to as HNDE/2A, is proposed. It can be featured as: 1) two niching techniques, i.e., crowding and speciation, are combined to balance the diversity and convergence; 2) a root archive is used to save the found roots during the run. Additionally, if a root is saved into the root archive, it will be reinitialized immediately to further promote the population diversity; and 3) an inferior offspring archive is presented to utilize the useful information of the inferior offspring. To evaluate the performance of our proposal, 30 problems are chosen as the test suite. Moreover, HNDE/2A is also used to solve two real-world problems. Compared with other algorithms, HNDE/2A yields better results in terms of the root ratio and the success rate with a less computational budget. [ABSTRACT FROM AUTHOR]

Traditionally, visualizing benchmark problems is an integral task in the domain of evolutionary algorithms development. Researchers get inspired for new search heuristics by challenges observed in functional landscapes. Moreover, landscape characteristics, features, and even terminology to describe them are derived from visualizations. And most importantly, benchmark designers need visualizations for identifying diverse problems that potentially challenge different aspects of optimization algorithms. As easy as it is to visualize single-objective problems, until recently there were hardly any approaches for gaining similar insights for multi-objective problems. Also, there have been no seamlessly accessible tools to support such visualizations. This article presents a comprehensive overview of the available visualization techniques from literature, including two interactive techniques to visualize 3-D problems, as well as two novel techniques which are suitable to scale some visualization properties to even higher-dimensional spaces. All presented techniques are integrated into a single tool, the moPLOT-dashboard, which enables users to perform landscape analyses in an interactive manner. Finally, the value of the tool and the visualizations is demonstrated in a series of usage scenarios on well-known benchmark problems. [ABSTRACT FROM AUTHOR]

In the field of optimization and machine learning, the statistical assessment of results has played a key role in conducting algorithmic performance comparisons. Classically, null hypothesis statistical tests have been used. However, recently, alternatives based on Bayesian statistics have shown great potential in complex scenarios, especially when quantifying the uncertainty in the comparison. In this work, we delve deep into the Bayesian statistical assessment of experimental results by proposing a framework for the analysis of several algorithms on several problems/instances. To this end, experimental results are transformed to their corresponding rankings of algorithms, assuming that these rankings have been generated by a probability distribution (defined on permutation spaces). From the set of rankings, we estimate the posterior distribution of the parameters of the studied probability models, and several inferences concerning the analysis of the results are examined. Particularly, we study questions related to the probability of having one algorithm in the first position of the ranking or the probability that two algorithms are in the same relative position in the ranking. Not limited to that, the assumptions, strengths, and weaknesses of the models in each case are studied. To help other researchers to make use of this kind of analysis, we provide a Python package and source code implementation at https://zenodo.org/record/6320599. [ABSTRACT FROM AUTHOR]

Quality-diversity (QD) algorithms refer to a class of evolutionary algorithms designed to find a collection of diverse and high-performing solutions to a given problem. In robotics, such algorithms can be used for generating a collection of controllers covering most of the possible behaviors of a robot. To do so, these algorithms associate a behavioral descriptor (BD) to each of these behaviors. Each BD is used for estimating the novelty of one behavior compared to the others. In most existing algorithms, the BD needs to be hand-coded, thus requiring prior knowledge about the task to solve. In this article, we introduce: autonomous robots realizing their abilities, an algorithm that uses a dimensionality reduction technique to automatically learn BDs based on raw sensory data. The performance of this algorithm is assessed on three robotic tasks in simulation. The experimental results show that it performs similar to traditional hand-coded approaches without the requirement to provide any hand-coded BD. In the collection of diverse and high-performing solutions, it also manages to find behaviors that are novel with respect to more features than its hand-coded baselines. Finally, we introduce a variant of the algorithm which is robust to the dimensionality of the BD space. [ABSTRACT FROM AUTHOR]

Band selection (BS) is a widely used dimensionality reduction technique for hyperspectral images. However, most of existing evolutionary algorithms focus on searching a globally optimal band subset under a fixed size, and their obtained band subsets may still contain a large number of redundant bands. In order to simultaneously obtain multiple optimal band subsets with different sizes, this article proposes an unsupervised multitask artificial bee colony (ABC) BS algorithm based on variable-size clustering (MBBS-VC). First, a variable-size band clustering method based on worst class decomposition is developed, based on which the BS problem can be modeled as a multitask optimization problem. Next, a multitask multimicrogroup bee colony algorithm with variable coding length is proposed to simultaneously search multiple optimal band subsets with different sizes. Moreover, several new strategies, including the intergroup collaboration strategy based on bidirectional neighborhood learning and the multimeasure integration judgment (MIJ) mechanism, are designed to improve the performance of MBBS-VC. In this article, the hyperspectral BS problem is transformed into a multitask optimization problem for the first time. Finally, compared with 15 classical BS algorithms on several commonly used datasets, experimental results verify the superiority of the proposed BS algorithm. [ABSTRACT FROM AUTHOR]

This work assesses the efficacy of evolutionary algorithms (EAs) using an intuitive multidimensional scaling (MDS) visualization of the evolution of a population. We propose the use of landmark MDS (LMDS) to overcome computational challenges inherent to visualizing many-objective and complex problems with MDS. For the benchmark problems we tested, LMDS is akin to MDS visually, whilst requiring less than 1% of the time and memory necessary to produce an MDS visualization of the same objective space solutions, leading to the possibility of online visualizations for multi- and many-objective optimization evaluation. Using multi- and many-objective problems from the DTLZ and WFG benchmark test suites, we analyze how Landmark MDS visualizations can offer far greater insight into algorithm performance than using traditional algorithm performance metrics such as hypervolume alone, and can be used to complement explicit performance metrics. Ultimately, this visualization allows the visual identification of problem features and assists the decision maker in making intuitive recommendations for algorithm parameters/operators for creating and testing better EAs to solve multi- and many-objective problems. [ABSTRACT FROM AUTHOR]

Large-scale, high-quality data sets are central to the development of advanced machine learning techniques that increase the effectiveness of existing optimization methods or even inspire novel ones. Especially in the engineering domain, such high-quality data sets are rare due to confidentiality concerns and generation costs, be it computational or manual efforts. We, therefore, introduce the OSU-Honda Automobile Hood Dataset (CarHoods10k), an industry-grade 3-D vehicle hood data set of over 10 000 shapes along with mechanical performance data that were validated against real-world hood designs by industry experts. CarHoods10k offers researchers and practitioners the unique opportunity to develop novel methods on realistic data with relevance to real-world vehicle design. To illustrate central use cases, we first apply methods from geometric deep learning to learn a compact latent representation for design space exploration. Second, we use machine learning models to predict mechanical hood performance from the learned latent representation. We thus demonstrate the effectiveness of machine learning for building metamodels, which are used in design optimization whenever possible to replace costly engineering simulations. Third, we integrate CarHoods10k in a topology optimization approach based on evolutionary algorithms to demonstrate its capability to search for high-performing structures, while maintaining manufacturability constraints. [ABSTRACT FROM AUTHOR]

Comprehensive benchmarking of clustering algorithms is rendered difficult by two key factors: 1) the elusiveness of a unique mathematical definition of this unsupervised learning approach and 2) dependencies between the generating models or clustering criteria adopted by some clustering algorithms and indices for internal cluster validation. Consequently, there is no consensus regarding the best practice for rigorous benchmarking, and whether this is possible at all outside the context of a given application. Here, we argue that synthetic datasets must continue to play an important role in the evaluation of clustering algorithms, but that this necessitates constructing benchmarks that appropriately cover the diverse set of properties that impact clustering algorithm performance. Through our framework, HAWKS, we demonstrate the important role evolutionary algorithms play to support flexible generation of such benchmarks, allowing simple modification and extension. We illustrate two possible uses of our framework: 1) the evolution of benchmark data consistent with a set of hand-derived properties and 2) the generation of datasets that tease out performance differences between a given pair of algorithms. Our work has implications for the design of clustering benchmarks that sufficiently challenge a broad range of algorithms, and for furthering insight into the strengths and weaknesses of specific approaches. [ABSTRACT FROM AUTHOR]

Intensity-modulated radiotherapy (IMRT) is one of the most popular techniques for cancer treatment. However, existing IMRT planning methods can only generate one solution at a time and, consequently, medical physicists should perform the planning process many times to obtain diverse solutions to meet the requirement of a clinical case. Meanwhile, multiobjective evolutionary algorithms (MOEAs) have not been fully exploited in IMRT planning since they are ineffective in optimizing the large number of discrete variables of IMRT. To bridge the gap, this article formulates IMRT planning into a large-scale combinatorial many-objective optimization problem and proposes a coevolutionary algorithm to solve it. In contrast to the existing MOEAs handling high-dimensional search spaces via variable grouping or dimensionality reduction, the proposed algorithm evolves one population with fine encoding for local exploitation and evolves another population with rough encoding for global exploration. Moreover, the convergence speed is further accelerated by two customized local search strategies. The experimental results verify that the proposed algorithm outperforms state-of-the-art MOEAs and IMRT planning methods on a variety of clinical cases. [ABSTRACT FROM AUTHOR]

Scientific workflow scheduling involves the allocation of workflow tasks to particular computational resources. The generation of optimal solutions to reduce run-time, cost, and energy consumption, as well as ensuring proper load balancing, remains a major challenge. Therefore, this work presents a Multi-Swarm Particle Swarm Optimization (MS-PSO) algorithm to improve the scheduling of scientific workflows in cloud-fog environments. MS-PSO seeks to address the canonical PSO’s problem of premature convergence, which leads it to suboptimal solutions. In MS-PSO, particles are divided into several swarms, with each swarm having its own cognitive and social learning coefficients. This work also develops a weighted sum objective function for the workflow scheduling problem, based on four objectives: makespan, cost, energy and load balancing for cloud and fog tiers. The FogWorkflowSim Toolkit is used in the evaluation process, with the objectives serving as performance metrics. The MS-PSO approach is compared with the canonical PSO, Genetic Algorithm (GA), Differential Evolution (DE) and GA-PSO. The following scientific workflows are used in the simulations: Montage, Cybershake, Epigenomics, LIGO and SIPHT. MS-PSO outperforms the canonical PSO on all scientific workflows and under all performance metrics. It competes fairly well against the other approaches and it is more stable and reliable. It only ranks second to PSO, in terms of execution time. In future, multiple species, incorporating population update mechanisms from several algorithmic frameworks (MS-PSO, DE, GA), will be used for scientific workflow scheduling. Hybdridization of the realized algorithm with dynamic approaches will also be investigated.

With the increasing uptake of electric vehicles (EVs), the need for efficient scheduling of EV charging is becoming increasingly important. A charging station operator needs to identify charging/discharging power of the client EVs over a time horizon while considering multiple objectives, such as operating costs and the peak power drawn from the grid. Evolutionary algorithms (EAs) are a popular choice when faced with problems involving multiple objectives. However, since the objectives and constraints of this problem can be expressed using linear functions, it is also possible to come up with improvised multi-objective formulations which can be solved with exact techniques such as mixed-integer linear programming (MILP). With both approaches having their potential strengths and pitfalls, it is worth investigating their use to inform the algorithmic choices, which this study aims to address. In doing so, it makes a number of contributions to the topic, including extension of an existing EV charging problem to a multi-objective form; observing some interesting properties of the problem to improve both the MILP and EA solution approaches; and comparing the performance of MILP and EA. The study provides some useful insights into the problem, initial results and quantitative basis for selecting solution approaches, and highlights some areas of further development.

We present a novel method for generating cycling training routes from geographical property graphs based on an Evolutionary Algorithm. The algorithm operators of crossover and mutation are adjusted for use in the Property Graph domain. Data fusion of geographical data from the OpenStreetMap, EU-DEM digital surface model, and existing training records is performed as a basis of the intersections-paths property graph. The proposed approach allows route generation based on their starting and ending points in the property graph and their distance and ascent. A property graph of all intersections and cycling roads is shown and generated for the regions of Podravje and Pomurje. The property graph used in the proposed algorithm’s feasibility demonstration is shown. This is done by presenting four different cases of routes generated with our algorithm. The algorithm allows for generating classic cycling routes of A to B nature, routes with more than two fixed points, and cyclical training routes. The research is concluded by offering further directions on route generation research.

The Salp Swarm Algorithm (SSA) outperforms well-known algorithms such as particle swarm optimizers and grey wolf optimizers in complex optimization challenges. However, like most meta-heuristic algorithms, SSA suffers from slow convergence and stagnation in the best local solution. In this study, a Salp swarm algorithm (SSA) is combined with a local escaping operator (LEO) to overcome some inherent limitations of the original SSA. SSALEO is a novel search technique that accounts for population diversity, the imbalance between exploitation and exploration, and the SSA algorithm’s premature convergence. By implementing LEO in SSALEO, the search slowdown in SSA is eliminated, and the local search efficiency of swarm agents is improved. The proposed SSALEO method is tested using the CEC 2017 benchmark with 50 and 100 decision variables, seven CEC2008lsgo test functions with 200, 500, and 1000 decision variables, and its performance was compared to other metaheuristic algorithms (MAs) and advanced algorithms, including seven Salp swarm variants. The comparisons show that SSA greatly benefits from LEO by enhancing the quality and accelerating its solutions’ convergence rate. The SSALEO was then assessed using a benchmark set of seven well-known constrained design challenges in various engineering domains defined in the CEC 2020 conference benchmark. Friedman and Wilcoxon rank-sum statistical tests are also used to examine the results. According to experimental data and statistical tests, the SSALEO algorithm is very competitive and often superior to the algorithms used in the studies. Further, the proposed approach can be viewed as a special LSGO optimizer whose performance exceeds that of specialized state-of-the-art algorithms like CMA-ES and SHADE.

The differential evolution (DE) is a well known population-based evolutionary algorithm that has shown capabilities for solving real-world problems such as resource allocation, multicast routing, and localization of target nodes. However, the accuracy of the DE, like other evolutionary algorithms, depends on the settings of its control parameters. The localization of target nodes is highly nonlinear and multi-modal, which may trap the DE in a local optimum. A local optimum may be avoided by a proper selection of the control parameters. One of the key control parameters is the population size (PS), which affects directly the localization accuracy and computational complexity. Finding an adequate PS throughout the evolution process is a challenging task. Even if an adequate PS is found it may not be the adequate PS anymore when the scenario of a problem changes. Although several approaches have been proposed for adapting the PS, they have not been evaluated when solving the localization problem. In this paper, a comprehensive comparison in terms of accuracy and computational demand is conducted among the state-of-the-art PS adaptation techniques when employed with the DE for solving the localization problem of target nodes in various scenarios. We also propose three new PS adaptation techniques, namely, exponential, parabolic, and logistic reduction. The results from extensive numerical simulations show that, after setting the initial PS properly, there is no technique that outperforms the others in practically all the scenario of the localization problem. Additionally, the DE with the proposed techniques provides competitive localization accuracy with considerably less computational complexity. Specifically, The proposed approaches reduce the computational demand by approximately 50 % over the standard DE in all the scenarios considered here.

The emergence of next-generation Internet-of-Things (NG-IoT) applications introduces several challenges for the sixth-generation (6G) mobile networks, such as massive connectivity, increased network capacity, and extremely low-latency. To countermeasure the aforementioned challenges, ultra-dense networking has been widely identified as a possible solution. However, the dense deployment of base stations (BSs) is not always possible or cost-efficient. Drone-base-stations (DBSs) can facilitate network expansion and efficiently address the requirements of NG-IoT. In addition, due to their flexibility, they can provide on-demand connectivity in emergency scenarios or address temporary increases in network traffic. Nevertheless, the optimal placement of a DBS is not a straightforward task due to the limited energy reserves and the increased signal quality degradation in air-to-ground links. To this end, swarm intelligence approaches can be attractive solutions for determining the optimal position of the DBS in the three-dimensional (3D) space. In this work, we explore well-known swarm intelligence approaches, namely the Cuckoo Search (CS), Elephant Herd Optimization (EHO), Grey Wolf Optimization (GWO), Monarch Butterfly Optimization (MBO), Salp Swarm Algorithm (SSA), and Particle Swarm Optimization (PSO) and investigate their performance and efficiency in solving the aforementioned problem. In particular, we investigate the performance of three scenarios in the presence of different swarm intelligence approaches. Additionally, we carry out non-parametric statistical tests, namely the Friedman and Wilcoxon tests, in order to compare the different approaches.

Many precious lives are lost world-wide due to road accidents. To counter this issue, the ultimate solution is Vehicular ad hoc networks (VANETs). Due to the high mobility of vehicles and varying network topology in VANETs, efficient communication among the vehicular nodes is of extreme importance. To enhance communication proficiency in VANETs, clustering is a renowned procedure. Therefore, a clustering algorithm based on Moth-Flame Optimization (MFO), titled AMONET, is projected which can effectively work in high mobility nodes scenario of VANETs. AMONET is based on bio inspired procedure and creates optimized clusters for reliable and efficient communication. Our algorithm is assessed experimentally with well-known procedures such as Ant Colony Optimization (ACO), Comprehensive Learning Particle Swarm Optimization (CLPSO) and Multi Objective Particle Swarm Optimization (MOPSO). Several experiments are conducted to measure the comparative efficiency of these procedures. The average cumulative results for all the grid sizes are 27.1% for AMONET while 36.3% for ACO, 54.9% for CLPSO and 58.7% for MOPSO. The results signify that AMONET produces near ideal results, covers the entire network and generates least number of clusters. It is an efficient technique to accomplish vehicular clustering with the purpose of improving the network’s overall performance and consequently reducing the routing cost of the vehicular network.

The generation and transmission capacities of many power systems in the world are significantly increasing due to the escalating global electricity demand. Therefore, the adequacy evaluation of power systems has become a computationally challenging and time-consuming task. Recently, population-based intelligent search methods such as Genetic Algorithms (GAs) and Binary Particle Swarm Optimization (BPSO) have been successfully employed for evaluating the adequacy of power generation systems. In this work, the authors propose a novel Evolutionary Swarm Algorithm (ESA) for the adequacy evaluation of composite generation and transmission systems. The random search guiding mechanism of the ESA is based on the underlying philosophies of GAs and BPSO. The main objective of the ESA is to find out the most probable system failure states that significantly affect the adequacy of composite systems. The identified system failure states can be directly used to estimate the system adequacy indices. The proposed ESA-based framework is used to evaluate the adequacy of the IEEE Reliability Test System (RTS). The estimated annualized and annual adequacy indices such as Probability of Load Curtailments (PLC), Expected Duration of Load Curtailments (EDLC), Expected Energy Not Supplied (EENS) and Expected Frequency of Load Curtailments (EFLC) are compared with those obtained using Sequential Monte Carlo Simulation (SMCS), GA and BPSO. The results show that the accuracy, computational efficiency, convergence characteristics, and precision of the ESA outperform those of GA and BPSO. Moreover, compared to SMCS, the ESA can estimate the adequacy indices in a more time-efficient manner.

Evolutionary algorithms have shown great successes in various real-world applications ranging in molecule to astronomy. As a mainstream evolutionary algorithm, differential evolution (DE) possesses the characteristics of simple algorithmic structure, easy implement, and efficient search performance. Nevertheless, it still suffers from the issues of local optimal trapping and premature of evolution problems. In this study, we innovatively improve the performance of DE by incorporating a full utilization of information feedback, which includes the population’s holistic information and the direction of differential vectors. The proposed permutation-archive information directed differential evolution (PAIDDE) algorithm is verified on a set of 29 benchmark numerical functions and 22 real-world optimization problems. Extensive experimental and statistical results show that PAIDDE can significantly outperform other 12 state-of-the-art algorithms in terms of solution qualities. Additionally, the computational complexity, solution distribution, convergence speed, search dynamics, and population diversity of PAIDDE are systematically analyzed.

Due to the diversity of power consumption patterns, the false positive rate (FPR) of data-driven electricity theft detection (ETD) methods is too high to meet practical needs, which severely restricts the engineering application of data-based methods. To reduce FPR of ETD methods based on advanced metering infrastructure (AMI), a deep neural network with low FPR (LFPR-DNN) is proposed in this paper. First, a deep model is constructed based on one-dimensional convolution and residual network, which can automatically extract features from consumption data. Then, a two-stage training scheme is used to train the network. In the first stage, the conventional gradient descent algorithm is used to update the network weights. To minimize the impact of data imbalance on detection performance, focal loss is used. Besides, grid search is used to optimize the hyper-parameters of the model. In the second stage, with FPR as the optimization objective, the particle swarm optimization (PSO) algorithm is used to train the network. Finally, the proposed LFPR-DNN is verified by using the open Irish data set. Compared to other state-of-the-art classifiers, LFPR-DNN has the lowest FPR with 0.29% and the highest AUC with 99.42%. The FPR is reduced by an order of magnitude, which verifies the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

The sensor nodes in wireless sensor networks have the deficiency of limited energy, and the multihop transmission of information will lead to a premature paralysis of nodes near the sink. The use of the mobile sink can balance the energy consumption and greatly prolong the lifetime. Therefore, this article studies the scheduling strategy of multiple mobile sinks and proposes a heuristic strategy based on interval type-2 fuzzy rough neural network. The energy and lifetime of sensor nodes, as well as location information of the mobile sink and special nodes are taken as input features. Through neural network learning, the outputs determine whether to move, moving direction, moving distance, and residence time, which can complete the scheduling task. The scheduling problem is regarded as a multiobjective optimization problem, and the network lifetime, the moving path length, and the network interpretability are optimized at the same time, so as to obtain a lightweight network with good interpretability and performance. Based on the parallel multiobjective evolutionary algorithm, a multiobjective neural evolutionary framework is constructed. This framework can balance multiple objectives and complete complex scheduling tasks. Compared with static sinks, random-moving sinks, sinks with manually designed strategy, gene expression programming-based sinks, as well as the other state-of-the-art multiobjective evolutionary algorithms, the proposed framework can achieve superior results. [ABSTRACT FROM AUTHOR]

This article studies a distributed heterogeneous hybrid flow shop scheduling problem under nonidentical time-of-use electricity tariffs (DHHFSP-NTOU). The makespan and the total electricity charge are considered as the optimization objectives from the view of production and management. The DHHFSP-NTOU considers different processing capabilitie and time-of-use electricity tariffs for each factory. The mixed-integer linear programming (MILP) model of DHHFSP-NTOU is established. To solve the DHHFSP-NTOU, this article proposes an ant colony optimization behavior-based multiobjective evolutionary algorithm based on decomposition (ACO_MOEA/D). A problem-specific ant colony behavior is presented to construct offspring individuals. Eight neighborhoods within the factory and between factories are adopted to improve the quality of the individuals in the archive set. A right-shift movement is used to reduce the electricity charge. A large number of numerical experiments and comprehensive investigations are carried out to test the efficiency and effectiveness of ACO_MOEA/D. The experimental results show that each component (e.g., ant colony behavior, neighborhoods move operators, right-shift movement) contributes to the performance of ACO_MOEA/D. The comparisons with several related algorithms show the superiority of ACO_MOEA/D for solving the DHHFSP-NTOU. Note to Practitioners—From the managers’ insights, the electricity charge is a large cost in the production. The scheduling is an economical approach to reduce the electricity charge. For the time-of-use (TOU) tariffs, the managers can adjust the schedule to reduce the idle time or move some operations to the interval period with a lower electric price. This article studies a distributed heterogeneous hybrid flow shop scheduling problem under nonidentical TOU (UTOU) electricity. This model can be used in many manufacturing enterprises that have several heterogeneous factories. This article proposes an ant colony optimization behavior-based multiobjective evolutionary algorithm based on decomposition (ACO_MOEA/D) to minimize the makespan and the total electricity charge. The ACO_MOEA/D can provide the economy and high-efficiency schedules for practitioners. The computational results confirm its effectiveness and efficiency. [ABSTRACT FROM AUTHOR]

This article addresses the dynamic vehicle routing problem (DVRP). DVRP is a challenging variation of the classic vehicle routing problem in which some customers are not known in advance. The objective is to incorporate new customers into the schedule as they become known while still attempting to minimize the cost of serving all customers without violating the problem constraints. This work proposes an effective population-based approach that integrates various algorithmic components to address DVRP. The approach combines a local search algorithm with various evolutionary operators (crossover and mutation) in an adaptive manner. To promote diversity, the proposed approach utilizes a population of solutions and uses a quality-and-diversity strategy to retain only promising solutions. The well-known 21 DVRP benchmark instances are utilized to test the performance of the proposed approach. An experimental comparison is carried out to assess the contribution of the integrated components. Results demonstrate that the integrated components significantly improve search performance. It is also shown that the proposed approach produces new best results for several instances when compared with the best methods reported in the literature. Note to Practitioners—This work deals with dynamic vehicle routing problems (DVRPs). In DVRP, only limited information is available at the start, and new information is revealed over time. It proposes an effective hybrid approach to tackle this problem. The proposed approach combines evolutionary operators and a local search approach in an adaptive manner to exploit the benefits of each algorithmic component. The proposed approach produces several new best-known solutions. The experimental results demonstrate that the proposed approach is very effective in dealing with DVRP and can help decision-makers in designing best-routing solutions. [ABSTRACT FROM AUTHOR]

The spread of the COVID-19 disease has prompted a need for immediate reaction by governments to curb the pandemic. Many countries have adopted different policies and studies are performed to understand the effect of each of the policies on the growth rate of the infected cases. In this article, the data about the policies taken by all countries at each date, and the effect of the policies on the growth rate of the pandemic are used to build a model of the pandemic’s behavior. The model takes as input a set of policies and predicts the growth rate of the pandemic. Then, a population-based multiobjective optimization algorithm is developed, which uses the model to search through the policy space and finds a set of policies that minimize the cost induced to the society due to the policies and the growth rate of the pandemic. Because of the complexity of the modeling problem and the uncertainty in measuring the growth rate of the pandemic via the models, an ensemble learning algorithm is proposed in this article to improve the performance of individual learning algorithms. The ensemble consists of ten learning algorithms and a metamodel algorithm that is built to predict the accuracy of each learning algorithm for a given data record. The metamodel is a set of support vector machine (SVM) algorithms that is used in the aggregation phase of the ensemble algorithm. Because there is uncertainty in measuring the growth rate via the models, a landscape smoothing operator is proposed in the optimization process, which aims at reducing uncertainty. The algorithm is tested on open access data online and experiments on the ensemble learning and the policy optimization algorithms are performed. [ABSTRACT FROM AUTHOR]

Multiobjective optimization evolutionary algorithms (MOEAs) have received significant achievements in recent years. However, they encounter many difficulties in dealing with many-objective optimization problems (MaOPs) due to the weak selection pressure. One possible way to improve the ability of MOEAs for these MaOPs is to balance the convergence and diversity in the high-dimensional objective space. Based on this consideration, this article proposes a novel generic two-stage (TS) framework for MaOPs. The entire evolutionary search process is divided into two stages: in the first stage, a new subregion dominance and a modified subregion density-based mating selection mainly purse the convergence and in the second stage, a novel level-based Pareto dominance cooperates with the traditional Pareto dominance that mainly promotes diversity. Integrated into NSGA-II, the TS NSGA-II, referred to as TS-NSGA-II, is proposed. To extensively evaluate the performance of our approach, 29 benchmark problems were used as the test suite. The experimental results demonstrate our approach obtained superior or competitive performance compared with eight state-of-the-art many-objective optimization evolutionary algorithms. To study its generality, the proposed TS strategy was also combined with four other advanced methods for MaOPs. The results show that it can also improve the performance of these four methods in terms of convergence and diversity. [ABSTRACT FROM AUTHOR]

Constrained multiobjective optimization problems (CMOPs) involve both conflicting objective functions and various constraints. Due to the presence of constraints, CMOPs’ Pareto-optimal solutions are very likely lying on constraint boundaries. The experience from the constrained single-objective optimization has shown that to quickly obtain such an optimal solution, the search should surround the boundary of the feasible region from both the feasible and infeasible sides. In this article, we extend this idea to cope with CMOPs and, accordingly, we propose a novel constrained multiobjective evolutionary algorithm with bidirectional coevolution, called BiCo. BiCo maintains two populations, that is: 1) the main population and 2) the archive population. To update the main population, the constraint-domination principle is equipped with an NSGA-II variant to move the population into the feasible region and then to guide the population toward the Pareto front (PF) from the feasible side of the search space. While for updating the archive population, a nondominated sorting procedure and an angle-based selection scheme are conducted in sequence to drive the population toward the PF within the infeasible region while maintaining good diversity. As a result, BiCo can get close to the PF from two complementary directions. In addition, to coordinate the interaction between the main and archive populations, in BiCo, a restricted mating selection mechanism is developed to choose appropriate mating parents. Comprehensive experiments have been conducted on three sets of CMOP benchmark functions and six real-world CMOPs. The experimental results suggest that BiCo can obtain quite competitive performance in comparison to eight state-of-the-art-constrained multiobjective evolutionary optimizers. [ABSTRACT FROM AUTHOR]

Autonomous underwater vehicles (AUVs) have been widely implemented to explore marine resources or perform marine missions; meanwhile, complex mission planning has been expected to enhance the intelligence level, improve energy efficiency, and expand AUVs' application areas. The mission planning model needs to consider energy consumption, mission quality, and mission quantity of long-term working in the marine environment. In this article, one unified and robust model was proposed based on the above requirements in the complex marine environment to improve the automaticity of AUV. However, solving this mission planning model is a nontrivial problem, especially when high robustness, high efficiency, fast response, and near-optimal results are required. Therefore, we propose a novel hyperheuristic algorithm based on evolutionary strategy (ES-HH). The proposed ES-HH combines a metaheuristic framework with a selection function to evaluate the performance of low-level heuristic operators online. This evolutionary strategy endows the hyperheuristic algorithm with the online learning feature, giving the ES-HH algorithm better computing efficiency, robustness, and near-optimal results. The experiment results show that the ES-HH algorithm can achieve better convergence and higher robustness than other algorithms, such as ant colony optimization and biogeography-based optimization algorithms. Compared with ACO and BBO algorithms, the proposed ES-HH algorithm can improve the mission completion rate by 3.42% and reduce the average energy consumption of a single task by 8.02%. [ABSTRACT FROM AUTHOR]

Various evolutionary algorithms (EAs) have been proposed to address feature selection (FS) problems, in which a large number of fitness evaluations are needed. With the rapid growth of data scales, the fitness evaluation becomes time consuming, which makes FS problems expensive optimization problems. Surrogate-assisted EAs (SAEAs) have been widely used to solve expensive optimization problems. However, the SAEAs still face difficulties in solving expensive FS problems due to their high-dimensional discrete decision variables. To address this issue, we propose an SAEA with parallel random grouping for expensive FS problems, in which three main components consist. First, a constraint-based sampling strategy is proposed, which considers the influence of the constraint boundary and the number of selected features. Second, a high-dimensional FS problem is randomly divided into several low-dimensional subproblems. Surrogate models are then constructed in these low-dimensional decision spaces. After that, all the subproblems are optimized in parallel. The process of random grouping and parallel optimization continues until the termination condition is met. Finally, a final solution is chosen from the best solution in the historical search and the best solution in the last population using a random, distance-, or voting-based method. Experimental results show that the proposed algorithm generally outperforms traditional, ensemble, and evolutionary FS methods on 14 datasets with up to 10 000 features, especially when the required number of real fitness evaluations is limited. [ABSTRACT FROM AUTHOR]

This article deals with a practical distributed hybrid differentiation flowshop scheduling problem (DHDFSP) for the first time, where manufacturing products to minimize makespan criterion goes through three consecutive stages: 1) job fabrication in first-stage distributed flowshop factories; 2) job-to-product assembly based on specified assembly plan on a second-stage single machine; and 3) product differentiation according to customization on one of the third-stage dedicated machines. Considering the characteristics of multistage and diversified processing technologies of the problem, building new powerful evolutionary algorithm (EA) for DHDFSP is expected. To achieve this, we propose a general EA framework called distributed co-evolutionary memetic algorithm (DCMA). It includes four basic modules: 1) dual population (POP)-based global exploration; 2) elite archive (EAR)-oriented local exploitation; 3) elite knowledge transfer (EKT) among POPs and EAR; and 4) adaptive POP restart. EKT is a general model for information fusion among search agents due to its problem independence. In execution, four modules cooperate with each other and search agents co-evolve in a distributed way. This DCMA evolutionary framework provides some guidance in algorithm construction of different optimization problems. Furthermore, we design each module based on problem knowledge and follow the DCMA framework to propose a specific DCMA metaheuristic for coping with DHDFSP. Computational experiments validate the effectiveness of the DCMA evolutionary framework and its special designs, and show that the proposed DCMA metaheuristic outperforms the compared algorithms. [ABSTRACT FROM AUTHOR]

In offline data-driven multiobjective optimization, no new data are available during the optimization process. Approximation models, also known as surrogates, are built using the provided offline data. A multiobjective evolutionary algorithm can be utilized to find solutions by using these surrogates. The accuracy of the approximated solutions depends on the surrogates and approximations typically involve uncertainties. In this article, we propose probabilistic selection approaches that utilize the uncertainty information of the Kriging models (as surrogates) to improve the solution process in offline data-driven multiobjective optimization. These approaches are designed for decomposition-based multiobjective evolutionary algorithms and can, thus, handle a large number of objectives. The proposed approaches were tested on distance-based visualizable test problems and the DTLZ suite. The proposed approaches produced solutions with a greater hypervolume, and a lower root mean squared error compared to generic approaches and a transfer learning approach that do not use uncertainty information. [ABSTRACT FROM AUTHOR]

Expensive multiobjective optimization problems pose great challenges to evolutionary algorithms due to their costly evaluation. Building cheap surrogate models to replace the expensive real models has been proved to be a practical way to reduce the number of costly evaluations. Supervised learning techniques from the community of machine learning have been widely applied to build either regressors, which approximate the fitness values of candidate solutions, or classifiers, which estimate the categories of candidate solutions. Considering the characteristics of the data produced in optimization, this article proposes a new surrogate model, called a relation model, for evolutionary multiobjective optimization. Instead of estimating the qualities of candidate solutions directly, the relation model tries to estimate the relationship between a pair of solutions $\langle \mathbf {x}, \mathbf {y}\rangle $ , i.e., $\mathbf {x}$ dominates $\mathbf {y}$ , $\mathbf {x}$ is dominated by $\mathbf {y}$ , or $\mathbf {x}$ is nondominated with $\mathbf {y}$ in the case of multiobjective optimization. To implement this idea, first a balanced training set is prepared, then a classifier is built based on the training data set to learn the relationship, and finally, the classifier with a voting-scoring strategy is applied to estimate the relationship between the candidate solutions and parent solutions. By this way, the promising candidate solutions are recognized and evaluated by the real models. The new approach is applied to three well-known benchmark suites and two real-world applications, and the experimental results suggest that the proposed method outperforms some state-of-the-art methods based on regression and classification models on the given instances. [ABSTRACT FROM AUTHOR]