Search

In the rapidly evolving e-commerce landscape, efficient packaging and logistics reduce costs and enhance customer satisfaction. This study addresses the problem of dynamic bin size optimization in e-commerce logistics by proposing a series of intelligent algorithms. Considering real-world constraints such as item separation requirements, a Mixed Integer Programming Model for Multi-Order Multi-Box Open-Dimension Rectangular Packing (MOMB-ODRPP) is formulated. The Stacked Clustering Algorithm (SCA) series, One-Dimensional Fixed Stacked Clustering Algorithm (ODF-SCA), Two-Dimensional Fixed Stacked Clustering Algorithm (TDF-SCA), and Variable Neighborhood Descent Spatial Ordering Algorithm (VND-SOA) series are employed to solve the MOMB-ODRPP model and improve order packing rates and optimize bin sizes. Computational experiments using real-world data from JD's e-commerce operations reveal that the TDF-SCA algorithm series outperforms the ODF-SCA series by approximately 5% in Case 4. In contrast, the VND-SOA-S1 and VND-SOA-S2 algorithms achieve improvements of 0.83% and 0.76%, respectively, over the TDF-SCA-P2 algorithm in Cases 4 and 11. The comparative analysis highlights the practical implications of bin size optimization, with Case 11 providing a more viable option for standardizing bin sizes in e-commerce logistics. • Proposed MOMB-ODRPP model for dynamic bin size optimization in e-commerce. • Developed ODF-SCA, TDF-SCA, and VND-SOA algorithms to solve MOMB-ODRPP. • Improved order packing rates by up to 5% using intelligent algorithms. [ABSTRACT FROM AUTHOR]

This paper is concerned with a variant of the traveling salesman problem (TSP) called the maximum scatter traveling salesman problem (MSTSP). The goal of MSTSP is to find a Hamiltonian cycle that maximizes the minimum length edge of the cycle. This problem has real-world applications in domains such as manufacturing and medical imaging. Both symmetric and asymmetric versions of the problem are considered in this paper. To address this problem, we have developed two evolutionary approaches. Our first approach is based on a genetic algorithm, whereas the second approach is based on a differential evolution algorithm. Initial solution generation procedure, variation operators (crossover and mutation) used in our approaches are designed considering the characteristics of this problem. The solutions obtained through variation operators are improved further through a local search that is specially adapted for MSTSP. For benchmarking, we have compared the performance of our proposed approaches with the state-of-the-art approaches available in the literature. Our approaches obtained better quality solutions in a shorter time for both symmetric and asymmetric versions of the problem, thereby clearly demonstrating their effectiveness in solving MSTSP. • This paper is concerned with the maximum scatter traveling salesman problem (MSTSP). • Two evolutionary approaches are proposed for solving this problem. • First approach is a steady-state genetic algorithm, while the latter is a discrete differential evolution algorithm. • Various components of our approaches are designed considering the characteristics of MSTSP. • Computational results show the effectiveness of our proposed approaches. [ABSTRACT FROM AUTHOR]

Projects face uncertainties during implementation, and these uncertainties often lead to risk factors that disrupt successful project execution and cause a loss of profit. The stochastic project scheduling technique has been an effective avenue for reducing the risks in project management. Nevertheless, previous studies often neglected resource limits for minimizing the financial risks of projects in stochastic project scheduling problems. This study aims to optimize the conditional net present value at risk (CNPVaR), which measures a project's expected loss of net present value (NPV) in a stochastic multi-mode resource-constrained project scheduling problem (SMRCPSP). Accordingly, considering the stochastic activity duration and cash flow, we construct a scenario-based optimization model for the SMRCPSP, where the solution (strategy) of the SMRCPSP is represented by a policy that includes an activity list and execution mode list; a scheduling policy is then employed to map the start times of activities at each decision time based on the solution. Three simulation-based hybrid genetic algorithms embedded with different iteration alternatives according to activity-based (AB) or resource-based (RB) scheduling policies are used. Furthermore, computational experiments are conducted to evaluate the proposed procedures through 480 instances. With respect to the CNPVaR, expected NPV, and algorithm stability metrics, the results of the experiments demonstrate that the genetic algorithm (GA) with local search based on the RB scheduling policy performed best compared with two other hybrids of the GA, the existing vibration damping optimization (VDO) method and a pure GA for large-sized projects. However, the VDO outperforms the other approaches according to the AB scheduling policy for small-scale projects. Additionally, a case study is conducted; the findings demonstrate that the risk attitude (confidence level) of project managers and the discount factor have impacts on the CNPVaR and expected NPV, but the CNPVaR is insensitive to the number of simulations. • The simulation-based CNPVaR optimization of the SMRCPSP is constructed. • Three simulation-based hybrid genetic algorithms are developed for the SMRCPSP. • Numerical experiments are conducted to test the performance of the procedures. • Case study is operated by the proposed model and the GA-LS. [ABSTRACT FROM AUTHOR]

For warehouse operations, efficiently scheduling the available resources is crucial to improve the productivity and customer satisfaction. This paper proposes a simulation-based evolutionary algorithm for order scheduling and multi-robot task assignment in a robotic mobile fulfillment system. The algorithm proactively deals with the effects of the processing time variability by evaluating schedules based on both its system performance as well as its robustness under uncertain conditions. The algorithm implements an efficient resource allocation method and a variance reduction technique to reduce the overall computational burden. The experimental results show that the techniques to reduce the computational time are effective and can significantly reduce the amount of simulations required for the fitness evaluation. If a candidate schedule is allocated insufficient simulation replications it can lead to an inaccurate estimate of its long-term average performance. This could lead to an average performance loss of 7.3 %. Furthermore, the proactive scheduler is able to generate schedules that are more robust compared to deterministically generated. A reduction in the average operational cost of about 5 % can be reached, compared to a deterministically generated schedule. The paper reveals the relevance of identifying and modeling uncertainty when designing schedules in an operational system, rather than looking for optimal schedules for ideal scenarios. • We present a proactive evolutionary algorithm for the stochastic RMFS scheduling problem. • We consider the integrated scheduling problem with processing variability. • We implement efficient evaluation, resource allocation and variance reduction procedures. • We evaluate the performance of the algorithm on instances with increasing due date tightness. • The proactive algorithm can generate robust and good quality schedules under uncertainty. [ABSTRACT FROM AUTHOR]

Our work aims to investigate methods for solving the mixed-model assembly line balancing problem (MALBP) under uncertainty with the objective of minimizing the number of workstations. Specifically, we model task processing time as fuzzy stochastic variables (FRVs) due to the inherent uncertainties and variations in the manufacturing environment. Additionally, we introduce a ranking method for FRVs and propose a mathematical model to address MALBP. The recently developed Red Fox Optimization (RFO) algorithm is also discretized for the first time to support solving this problem. Finally, matheuristic algorithms combine a metaheuristic such as the popular Genetic Algorithm (GA), Particle Swarm Optimization (PSO), or the Discretized Red Fox Optimization (DRFO) algorithm with the Mixed-Integer Programming (MIP) model to generate the best solution in a reasonable time. Our comparative results demonstrate that the GA-MIP combination outperforms the others in both objective value and computational time. • We are the first to consider the Assembly Line Balancing Problem with fuzzy random processing time. • a new method to rank fuzzy stochastic processing times is introduced by comparing interval values. • The problem is formulated as chance-constrained programming model. • The recently developed Red Fox Optimization (RFO) algorithm is also discretized for the first time to support solution process. • A matheuristic combines a metaheuristic and the MIP model to search for better solutions for large-sized problems. • The metaheuristic algorithm to generate the starting solution for the MIP model to shorten finding the optimal solution. • Our results are nearly the same quality as the exact solution of the MIP model in a much shorter time for small-sized instances. MGA matheuristic yields significantly better performance than others in both terms of objective value and computational time for large-sized problems. [ABSTRACT FROM AUTHOR]

In the petrochemical industry, product type conversion can cause capacity loss and produce transition products, leading to waste. Our study aims to minimize these transition products and tardiness in meeting customer due dates. We introduce the Engineering Experience Heuristic (EEH), which utilizes scheduling reference sheets to identify key production traits and constraints, thereby generating efficient schedules. Further building on the EEH, we propose a non-dominated sorting genetic algorithm II enhanced with reinforcement learning (NSGAeRL) to address the multi-objective scheduling optimization problem. The incorporation of reinforcement learning (RL), particularly through a model-based Markov Decision Process (MDP), allows for the dynamic adjustment of genetic algorithm (GA) parameters, including crossover and mutation mechanisms, based on the optimal policy derived from RL. We validate our proposed methodologies through empirical studies and numerical experiments. The results demonstrate that NSGAeRL outperforms standard benchmarking algorithms, offering superior performance in minimizing transition products and scheduling tardiness. • Focus on the scheduling optimization with minimizing transition products and total tardiness. • Five constraints including conversion constraint and precedence constraint are considered. • Propose the engineering experience heuristic (EEH) to generate a feasible scheduling. • Propose the non-dominated sorting genetic algorithm II embedded with reinforcement learning (NSGAeRL). • Both empirical study of the petrochemical industry and benchmark dataset are conducted. [ABSTRACT FROM AUTHOR]

The synchronization of production capacity with demand plays a significant role in maximizing profit. Firms attempt to align their activities, requiring that businesses be occasionally turned away. A trade-off exists between the revenue gained by a specific order and all related processing costs. We solve an important order acceptance and scheduling problem arising from this business model. A firm that receives multiple orders must determine: (i) whether to accept or reject any order, (ii) how to assign accepted orders to identical parallel machines, (iii) the sequence of accepted orders, and (iv) when to schedule order starting times. The objective is to maximize the total profit, measured as the difference between revenue and tardiness penalty costs. We propose an effective hybrid algorithm (GA-RP-TS) that combines a refining process (RP), a genetic algorithm (GA), and a tabu search (TS) step to tackle this NP-hard problem. The hybrid GA-RP-TS employs GA's strong global search ability, uses RP to improve a solution, and performs exploitation with TS. In particular, the proposed GA-RP-TS balances the intensification and diversification ability of metaheuristics and evolutionary algorithms. We evaluate the performance of the proposed GA-RP-TS by testing benchmark instances. Results show that the proposed GA-RP-TS outperforms the state-of-the-art algorithm in solving large-sized instances. GA-RP-TS obtains 22 new best solutions out of the 54 tested benchmark instances, besides helping prove optimality for two instances. In addition, GA-RP-TS consumes less computation time. Sensitivity analysis shows that the excellent performance of GA-RP-TS comes from the hybridization of its three components, i.e., GA, RP, and TS. • We develop an efficient hybrid heuristic for the order acceptance and scheduling problem. • The heuristic combines the strengths of a genetic algorithm, tabu search, and a refining process. • The performance of the heuristic is demonstrated using benchmark instances. • The heuristic successfully identifies 22 new best solutions. • The heuristic's excellence is attributed to the hybridization of its three components. [ABSTRACT FROM AUTHOR]

This study proposes a novel approach combining Machine Learning (ML) techniques and Genetic Algorithms (GA) for predicting High-Performance Computing (HPC) job run times. The objective is to create a prediction method universally applicable to any HPC system, irrespective of workload characteristics, application specific parameters, user behavior, or hardware architecture. Since user-supplied run time estimates are often inaccurate, we aim to categorize job runtimes into several classes, allowing users to select appropriate classes for their jobs. A Genetic Algorithm is developed to optimally define these runtime classes, determining both the number of classes and the time intervals represented by them. Four Machine Learning algorithms (K Nearest Neighbours, Support Vector Regression, Extreme Gradient Boosting and Deep Neural Networks) are implemented for run time prediction. A unique set of features extracted from historical job data serves as input to the Machine Learning models. The generalized nature of our method is demonstrated by validating its performance on data from six clusters with distinct configurations, applications and runtime distributions. Our results illustrate the superior performance of Machine Learning models incorporating GA-defined runtime classes for all datasets. Across all six datasets, our method achieves R2 scores exceeding 0.8, and accuracy greater than 0.7. [Display omitted] • Proposed an innovative method for predicting HPC job run times combining Machine Learning (ML) and Genetic Algorithms (GA). • Replaced run time estimations by users with runtime classes defined by GA, enabling users to select from predefined classes. • Implemented and compared four ML models for run time predictions: KNN, SVR, XGBoost and DNN. • The proposed method demonstrated superior performance across data from six diverse clusters. [ABSTRACT FROM AUTHOR]

In this paper, a single-objective redundancy allocation problem (RAP) is considered for a series system with k -out-of- n subsystems. In each subsystem, the components are binary, homogeneous, and repairable. Component failure and repair processes are independent and exponentially distributed. Three standby modes of systems are considered: cold, warm, and hot. Assuming a constant number of components in the entire system, component allocation aims to maximize system availability. A continuous-time Markov chain (CTMC) model is developed to calculate system availability. A genetic algorithm (GA) is developed to optimize the system configuration. A new way of encoding chromosomes as a sequence of binary numbers that represents the priority of the subsystem in component allocation is proposed. This method achieves feasible solutions through crossover and mutation for the entire population in each generation. First, using a hybrid of GA and CTMC, research is conducted on the solution of RAP in a series system with 10 k -out-of- n subsystems. Optimal system configurations are determined for a total number of components equal to 80, 90, 100, 110, 120, and 130. Then, a sensitivity analysis is performed to determine the impact of the model parameters and the standby mode on system availability. Finally, an analysis of performance and computation time is performed depending on the parameters of the genetic algorithm. The results demonstrate the high efficiency of the GA-CTMC algorithm in solving RAP in complex technical systems. Less than 0.01 % of feasible solutions are estimated to find an optimal solution. Furthermore, GA-CTMC is distinguished by a short computation time compared to other algorithms in the literature. • The Redundancy Allocation Problem (RAP) in a series system was investigated. • Binary, homogeneous, and repairable components were considered. • An analysis of the impact of cold, warm, and hot standby on system availability was performed. • A hybrid genetic algorithm and Markov chains were developed to solve the RAP. [ABSTRACT FROM AUTHOR]

The container loading problem (CLP) holds significant importance in logistics and supply chain management due to its direct influence on transportation costs and times, thereby impacting the competitive advantages of enterprises across industries. Although there are existing studies for CLP, there remains a gap in addressing the optimization of multiple objectives while satisfying practical constraints. Moreover, container loading methods must also meet performance requirements such as high computational speed, explainability, and implementation capability. Although meta-heuristic-based algorithms have shown effective computational capabilities and performance, such algorithms often being trapped in the local optima especially when searching in the vast solution space of the CLP, particularly as the quantity and diversity of cargo sizes and shapes increase. Motivated by realistic needs, this study aims to develop a UNISON-based framework that integrates the merging spaces algorithm, a novel simple but effective hybrid simplified swarm optimization genetic algorithm (SSO-GA), and a multi-populations co-evolution strategy to determine the loading sequence of parcels to maximize space utilization and weight balance while satisfying to practical constraints. Specifically, the merging spaces algorithm merges fragmented small spaces resulting from loaded parcels into larger spaces, thereby facilitating the accommodation of additional parcels. The hybrid SSO-GA splits encoded solutions and updates segment using novel strategies resulting in the rearrangement of a group of parcels and their corresponding layouts for better space utilization. Furthermore, the multi-populations co-evolution strategy enhances the diversity of the search space and stabilizes solution quality by simultaneously exploiting the best solutions and exploring alternative solutions during the solution update process. An empirical study was conducted by using a public benchmark dataset which demonstrated the practical effectiveness of the proposed framework by significantly improving average space utilization while ensuring center balance and satisfying practical constraints. Furthermore, this study can also serve as a digital support system capable of assisting decision-makers in optimizing container loading operations, thereby improving productivity and saving valuable time. [Display omitted] • Digital solution for optimizing container loading is developed. • Multiple objectives and constrains are considered for container loading optimization. • Multi-objectives Multi-populations Simplified swarm optimization (MOMP-SSO) is developed. • This solution integrated data preparing, meta-heuristics, and domain knowledge. • Container loading can be effectively optimized to improve space utilization and safety. [ABSTRACT FROM AUTHOR]

Limited samples and high-dimensional feature spaces often hinder the accuracy of machine learning (ML) models in regional groundwater potential assessment (GPA). This study proposes a novel framework, the GPA with Dimensionality Optimization (GPADO), that optimizes feature dimension reduction to enhance prediction performance. Taking the Jianghan Basin as an example, data on nine continuous variables and five categorical variables influencing the region's GPA were gathered, expanding the feature set to 37 through One-hot encoding for categorical variables. Three scenarios were devised to assess prediction outcomes following various dimensionality reduction approaches. Comparative analysis revealed that a hybrid dimension reduction method, incorporating both continuous and categorical variables, yielded the highest validation set accuracy. Consequently, genetic algorithm and Markov Chain Monte Carlo methods were employed to determine the optimal solution and uncertainties associated with four unknown parameters: the chosen dimension reduction method for continuous and categorical variables, and the number of dimensions retained. Results indicated that utilizing singular value decomposition to reduce categorical variables to three dimensions, coupled with principal component analysis reducing continuous variables to three dimensions, produced the highest model validation accuracy of 0.834 within the GPADO framework. This optimal configuration facilitated automated ML training, resulting in a final validation set accuracy of 0.851 and a test set accuracy of 0.836. The resulting model provided a more precise spatial distribution of groundwater potential and demonstrated the GPADO framework's effectiveness in improving GPA accuracy, particularly in data-scarce regions. The GPADO framework offers a valuable approach for enhancing GPA studies. [Display omitted] • Improved groundwater potential assessment (GPA) by introducing a hybrid method. • Effective for limited samples & high-dimensional data. • Defined four parameters for dimension reduction schemes. • Used genetic algorithm and MCMC for optimization, enhancing GPA accuracy. [ABSTRACT FROM AUTHOR]

This paper studies a delivery problem in a pharmaceutical cold chain. In this problem, each customer orders multiple pharmaceutical products and requires a due date for each order. Each product has sensitive characteristics of storage temperature range and shelf life. A refrigerated vehicle routing and scheduling problem with order acceptance is formulated as a mixed-integer-linear-programming model. In the model, the acceptance of each pharmaceutical order and the route and schedule of a limited number of refrigerated vehicles is simultaneously determined to maximize the total profit. A hybrid genetic algorithm embedding an insertion heuristic with modification operation is proposed. The performance of the proposed algorithm is evaluated and verified by comparing other metaheuristics. Managerial insights about various experimental environments for the order acceptance are suggested by conducting a sensitivity analysis. • The addressed problem is formulated as a mixed integer linear programming model. • An effective genetic algorithm is proposed to efficiently solve the optimization model. • The managerial insights of the order acceptance strategy are suggested. [ABSTRACT FROM AUTHOR]

Planning the public transportation system of a city is a complex process that depends on various factors, including transportation modes, origin–destination demands, service quality and reliability, and operational costs. The vehicle frequency setting (FS) problem is a particularly challenging aspect of this planning process. This work proposes a novel methodology, based on biased random-key genetic algorithms (BRKGA), for optimizing the FS of a bus-based public transport system. The proposed approach considers two optimization models that aim to address the following key metrics: (i) passengers' waiting time, and (ii) the operational cost for the concessionaire company, specifically the distance covered by buses. We apply our BRKGA methodology to a real case study using bus transport data from the city of Maceió (AL, Brazil). Our results demonstrate that, for each metric, the proposed methodology improves the performance of the city's public transport system by over 10%, compared to the current configuration. • A novel approach using BRKGA to optimize bus frequency in public transport systems. • Two models proposed: one minimizes waiting time, the other bus distance. • A comparison between the proposed solutions and the city's current system. • The method enhances the current city's transport configuration by over 10%. [ABSTRACT FROM AUTHOR]

Computed tomography (CT) is a non-destructive evaluation technique to know the internal structure of the objects under scan. It has numerous applications in engineering as well as in the medical field. The prime objective of this manuscript is to reduce the radiation dose to the object under scanning and to reconstruct the low-noise CT images, even in limited-view projections data scenarios. The present manuscript proposes a novel priority-based self-guided serial-parallel hybrid genetic algorithm for low-dose CT reconstruction. The current algorithm combines serial and parallel processing elements to reduce the loss of diversity in the population and increase the convergence rate. The proposed algorithm uses a novel priority-based self-guided competition operator. It consists of two approaches for maintaining exploration and exploitation tradeoffs. Here, the first approach uses serial processing, whereas the second approach uses parallel processing. The algorithm also uses priority-based hybrid crossover (PHCO) and adaptive priority-based mutation operator (PMO) to generate the most suitable offspring in every generation. Experimental results reveal that the presented algorithm produces satisfactory results with limited-view projections. The presented algorithm also outperforms other limited-view CT reconstruction algorithms. • The proposed algorithm reconstructs low-noise CT images with limited view projections. • The algorithm reduces the radiation dose and cost per scan. • The presented algorithm proposes a self-guided competition operator. • The self-guided competetion operator ensure the diversity of the solution space. • The presented algorithm supports a valuable diversity of the solution space. • The algorithm preserves the texture and structure properties of regenerated images. • The algorithm provides a high convergence rate to find an optimal solution. • The proposed algorithm is appropriate for low-dose CT reconstruction. [ABSTRACT FROM AUTHOR]

Schizophrenia (SCZ) is a severe mental and debilitating neuropsychiatric disorder that disrupts a person's thought processes, emotions, and behavior. Due to misdiagnosis, self-denial, and social stigma, many SCZ cases go untreated. Magnetic resonance imaging (MRI) is an excellent noninvasive tool for soft tissue contrast imaging because it provides crucial data on tissue structure size, position, and shape. The Resnet50 network is a deep residual learning framework used for feature extraction. Random-vector functional link network (RVFL) is an example of a single-hidden-layer feedforward network in which input features and hidden layer features are fed to the output layer. In this paper, we introduced a kernel ridge regression-based random vector functional link (KRR-RVFL) classifier which focuses on addressing the linearity issues in RVFL by designating the kernel function in the input layer for the precise diagnosis of SCZ. The genetic algorithm (GA) seeks to minimize the loss function by optimizing the weights and biases of the KRR-RVFL network. The classification performance is investigated on the SCZ and cognitive normal (CN) subjects, collected from the available open neuro platform, including 99 participants. The results of the suggested network show superior performance to the recent state-of-the-art networks in terms of accuracy 93.66%, sensitivity 92.22%, specificity 95.17%, precision 95.33%, F-measure 93.74%, and G-mean 93.68%. The performance metrics demonstrated the applicability of this framework for assisting clinicians in the automatic, precise evaluation of SCZ. • This paper proposes Schizophrenia diagnosis model using KRR-Opt RVFL classifier. • ResNet50 is deployed for the extraction of complex features from pre-processed data. • GA is incorporated with KRR-RVFL to optimize the parameters of the classifier. • The performance of the proposed KRR-Opt RVFL is compared with the state-of-the-art classifiers. [ABSTRACT FROM AUTHOR]

Imbalanced data classification presents a challenge in machine learning, inducing biased model learning. Moreover, data dimensionality poses another challenge as it highly impacts classifier performance. This paper proposes a new deep-learning method that combines feature selection with oversampling to address these challenges. The proposed approach, GA-SMOTE-DCNN, integrates a genetic algorithm (GA) for feature selection, SMOTE for oversampling, and a deep 1D-convolutional neural network (DCNN) for classification. This study reveals that pre-splitting the data into training and testing sets before applying SMOTE results in higher accuracy, showing an improvement in accuracy ranging between 1.94% and 3.98% compared to post-SMOTE splitting for each dataset. This method achieved accuracy rates of 86.81% for the Balance Scale dataset, 86.15% for the Oil Spill dataset, 89.21% for the Yeast dataset, 91.32% for the Mammography dataset, 88.23% for the Australian credit dataset, and 89.53% for the German Credit dataset when compared with benchmark methods, underscoring its significance in tackling high-dimensional and imbalanced data classification problems. This method demonstrates scalability in effectively addressing challenges associated with high-dimensional and imbalanced data classification across various domains. [Display omitted] • Propose a novel GA-SMOTE-DCNN technique to solve biased prediction and overfitting due to data dimensionality and imbalance. • Test the proposed model on six datasets related to different domains and compare it to models proposed in the literature. • Compare feature selection method with other filter and wrapper methods to prove effectiveness in enhancing classification. [ABSTRACT FROM AUTHOR]

Binary Decision Diagrams (BDDs) have become the state-of-art data structures in numerous fields, wherein, small-sized BDDs are required to reduce the companion cost. Since BDD size is sensitive to the order of variables for the Boolean function in use, evolutionary algorithms have been extensively exploited to solve the BDD reordering problem which is provably an NP-hard problem. However, getting trapped in a local minima is more likely as the population gets homogeneous during the evolution of the algorithm. This paper compares various diversity measures correlated to BDDs and studies the impact of different crossover and mutations used in the Genetic Algorithm (GA) on those diversity measures. Thereafter, we propose an enhanced GA-based reordering algorithm for BDD, wherein, the chosen diversity measure is calculated per generation to steer the application of proper variation operators, in such a way that an acceptable level of equilibrium between exploration and exploitation is preserved. Finally, we utilize a new heuristic Cyclic Crossover (HCX) that further improves the performance of the proposed algorithm following the fitness value. Experimental results on public benchmarks show that our proposed algorithm outperforms other algorithms from the literature when switching between HCX and swap mutation operators is made following the measured diversity metric. • We analyze various diversity measures in terms of their correlation with BDD size. • We analyze the impact of variation operators on the chosen diversity measures. • We empirically find the proper threshold value of the chosen diversity measures. • We propose an enhanced GA based BDD reordering driven by the chosen diversity metric. • We demonstrate the performance of the proposed algorithm compared with the basic GA. [ABSTRACT FROM AUTHOR]

This study presents a new imaging paradigm for overcoming the challenges limiting the improvement of the imaging quality in the electrical capacitance technique. The new imaging model enables the integration of measurement physics, sparsity-induced prior and physics-informed multi-fidelity learning prior (PIMFLP), as well as the synergy between data-driven and measurement physics modeling paradigms. The transformed L 1 norm is used to model the PIMFLP, and the maximum correntropy criterion is used as a data misfit term to restrict the adverse impact of noises or outliers. The PIMFLP is learned from data and characterizes the structural details of imaging targets. The half-quadratic optimization method is developed to overcome computational challenges in solving the novel imaging model. A new physics-informed multi-fidelity learning method is developed to predict PIMFLP by synergizing deep convolutional neural network with measurement physics, and a new bilevel optimization model solved by a new nested algorithm that merges the genetic algorithm and the split Bregman method is proposed for training. Comparisons with other popular reconstruction algorithms confirm that the new imaging method leads to the improved reconstruction accuracy and noise immunity, and opens up new possibilities for unlocking the potential of the measurement technology. • A data and measurement mechanism integrated imaging method is proposed for ECT. • A physics-informed multi-fidelity method is built for inferring image priors. • Measurement physics and data-driven modeling paradigms are unified. • The performance advantages of the proposed imaging technique are confirmed. [ABSTRACT FROM AUTHOR]

The emergence of smart cities is an example of how new technologies, such as the Internet of Things (IoT), have facilitated the creation of extensive interconnected and intelligent ecosystems. The widespread deployment of IoT devices has enabled the provision of constant environmental feedback, thereby facilitating the automated adaptation of associated systems. This has brought about a fundamental transformation in the way contemporary society functions. The security of emerging technologies such as IoT has become a significant challenge due to the added complexities, misconfigurations, and conflicts between modern and legacy systems. This challenge has a notable impact on the reliability and accessibility of existing infrastructure. Edge computing (EC) is a collaborative computing system that brings data processing and analysis closer to the edge of the network, where the data is generated, rather than in a centralized cloud environment. The utilization of the IoT has become more prevalent in both everyday life and the manufacturing sector, with a particular emphasis on critical infrastructure. The IoT is presently being utilized across diverse domains, including but not limited to industrial, agricultural, healthcare, and logistical sectors. The security of IoT networks has implications for the safety of individuals, the security of the nation, and economic development. Notwithstanding, conventional intrusion detection techniques that rely on centralized cloud-based systems that have been suggested in previous studies for IoT network security are insufficient to meet the requirements for data confidentiality, network capacity, and prompt responsiveness. In addition, the integration of IoT applications into smart devices has been shown to augment their functionalities. However, it is important to note that this integration also brings about potential security vulnerabilities. Furthermore, a significant number of contemporary IoT devices exhibit restricted security capabilities, rendering them vulnerable to intricate attacks and impeding the extensive integration of IoT technologies. Also, a lot of IoT network devices have been put in place that don't have hardware security measures. This means that traditional intrusion detection systems (IDS) aren't enough to protect the IoT network ecosystem. To address these issues, this research suggests the IoT-Defender framework, which combines a Modified Genetic Algorithm (MGA) model with a deep Long Short-Term Memory (LSTM) network to find cyberattacks in IoT networks. This research represents a pioneering attempt to employ the MGA for feature selection and the GA for fine-tuning the LSTM parameters within an EC framework. The parameters of the LSTM model were fine-tuned through the manipulation of the number of hidden layers, utilizing the GA fitness function. The customization of the MGA aimed to enhance its performance in selecting relevant features, optimizing the use of limited resources on IoT devices and edge nodes. The fine-tuning process involved optimizing hyperparameters, architecture, and training strategies to maximize the LSTM network's effectiveness in learning and detecting patterns in IoT network traffic. The synergy between the MGA and LSTM aimed at creating a comprehensive and efficient IDS. The feature selection by the MGA contributes to improving the LSTM's performance by providing it with more relevant and discriminating features. In order to solve the issue of class imbalance, we utilize the focal loss function, which provides greater weights to minority classes, hence improving the model's capacity to learn from those particular classes. The performance of the IoT-Defender model was assessed on the BoT-IoT, UNSW-NB15, and N-BaIoT datasets utilizing a Raspberry Pi IoT device. The results of our study show that the IoT-Defender model works better than other methods. This is shown by its accuracy score of 99.41%, detection rate of 99.78%, precision score of 98.50%, false alarm rate of 2.56%, mean intersection over union (mIoU) of 0.68, and training time of 81.3 seconds on BoT-IoT. The proposed IoT model is designed to be lightweight and can be installed on edge servers to detect cyber-attacks in real-time, specifically in the context of IoT security. • This study proposed a new framework named IoT-Defender based on edge computing for intrusion detection in IoT networks. • We employed a Modified Genetic Algorithm for choosing the best subset of features on the BoT-IoT, UNSW-NB15, and N-BaIoT datasets. • We design a fine-tuned LSTM parameter using the genetic algorithm to adjust the number of hidden layers. • The proposed model IoT-Defender is a lightweight IDS that can be deployed on edge servers. • The effectiveness of IoT-Defender in comparison to existing shallow models outperformed the state-of-the-art shallow models. [ABSTRACT FROM AUTHOR]

Fog-integrated Cloud has emerged as a novel computing paradigm that brings Cloud computing services to the network's edge in real-time, though with limited capabilities. Despite its advantages, there are several challenges including workload distribution, energy consumption, computational time, and network latency, that require attention. The workload of IoT applications can be distributed over the Fog or Cloud devices based on their priority, deadline, and latency restrictions. In this work, we introduce a novel population-based metaheuristic called MAYA, a modified variant of the JAYA algorithm, to address the Energy-Efficient Workload Distribution of Sensors (EEWDS) in the Fog-Cloud ecosystem. The workload distribution of IoT applications depends on several factors such as request deadlines, the energy consumed during transmission, and needed computation. The performance of the proposed model for the energy consumption, computation time, C O 2 emission, fairness index, and the convergence rate, is evaluated through simulation experiments. The results are compared in two scenarios: one concerning to methodology, where the performance is compared with JAYA, Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Ant Colony Optimization (ACO) techniques. The other scenario is based on the environment, where we examine Cloud-only, Fog-only, and Fog-Cloud integrated environments. Compared to JAYA, GA, PSO and ACO, the proposed MAYA technique demonstrates significant improvements, including reduction in energy consumption by 34.76%, 88.92%, 85.36% and 93.84%; decrease in computation time by 37.64%, 85.07%, 87.22%, and 91.08%; decrease in C O 2 emissions by 23.46%, 76.24%, 97.17%, and 99.02%; and increase in fairness index by 9.62%, 3.72%, 16.90%, and 15.26% respectively. • An IoT workload distribution method has been developed, which aims to minimize energy consumption in uploading, downloading, and computation. • A new technique MAYA, which is a modified version of JAYA metaheuristic, has been introduced. • In addition, C O 2 emission is also measured associated with the proposed MAYA technique in the Fog-integrated Cloud environment. • Furthermore, computation time, running time, convergence rate, and fairness index have also been studied. • A comparative study has been performed in two classic scenarios: Methodology-based and Environment-based. [ABSTRACT FROM AUTHOR]

A variety of domination concepts have been defined to provide better routing and defense strategies under different constraints. A double Roman dominating function (DROMDF) on a simple, undirected graph G is a function g : V → { 0 , 1 , 2 , 3 } such that every vertex x ∈ V with g (x) = 0 is adjacent to at least two vertices y 1 , y 2 with g (y 1) = g (y 2) = 2 or a vertex z 1 with g (z 1) = 3. Also, a vertex p with g (p) = 1 is adjacent to at least one vertex q 1 with g (q 1) ≥ 2. γ d R (G) , the double Roman domination number of G , is the smallest possible weight of all possible DROMDFs of G. Determining double Roman domination number of a graph is known to be NP-hard. Hence in this paper, we propose a genetic algorithm based approach for solving double Roman domination problem in which three heuristic algorithms have been proposed and problem specific crossover operator and a feasibility function has been developed. Further, we propose an ant colony optimization algorithm to solve double Roman domination problem. This paper provides an in-depth illustration of two algorithms for solving double Roman domination problem. Effectiveness of the proposed meta-heuristic algorithms is tested on the random graphs generated using NetworkX Erdős-Rényi model, a popular model for graph generation and Harwell–Boeing dataset, a well-known dataset for graph related problems. Further, we compare the results of both the meta-heuristic algorithms and the experimental results show that the proposed meta-heuristic algorithms for solving double Roman domination problem give a near optimal solution in reasonable time. Experimental results also show that the proposed ant colony optimization algorithm for solving double Roman domination problem outperforms genetic algorithm based procedure. • Double Roman domination problem (MDR) is NP-hard. • Proposed first ever meta-heuristic algorithms GA and ACO based solutions for MDR. • Have constructed three heuristics to generate initial population. • GA and ACO based solutions for MDR is tested on different classes of graphs. • Sign test suggests that ACO based solution for MDR outperforms GA based method. [ABSTRACT FROM AUTHOR]

Rule extraction from artificial neural networks (ANNs) refers to the process of identifying the underlying decision-making logic used by the network in its predictions. This involves converting the complex, non-linear relationships between inputs and outputs in the network into a set of interpretable rules. The main objective of rule extraction is to enhance the interpretability of decision-making in artificial neural networks (ANNs) and make it more understandable for human experts, especially in critical domains like healthcare, where clear decision-making is essential. Despite the ability of machine learning methods to interpret ANNs, interpreting complex neural networks remains limited, as most studies tend to sacrifice one criterion, related to rule quality, in favor of another. Rule extraction methods applied directly to complex neural networks may produce overly specific rules that are tailored to the training data. Such rules can suffer from overfitting and may not generalize well to unseen data. Also, The use of multiobjective genetic algorithms for rule extraction can result in a large number of rules, making it challenging for decision-makers to interpret them. This study proposes an innovative approach to address challenges in rule extraction from complex artificial neural networks (ANNs) by integrating a multiobjective genetic algorithm with the PROMETHEE method. By integrating these techniques, the study aims to identify meaningful relationships between inputs and outputs of ANNs, generating concise and reliable rules that are ranked based on support, confidence, and lift values. The use of PROMETHEE's rule ranking mechanism enabled the consideration of rule priority during candidate selection, guiding the genetic algorithm towards discovering high-quality rule sets and efficiently exploring the search space. This approach not only enhances the quality of extracted rules but also facilitates efficient decision-making by striking a balance between rule accuracy, fidelity, and comprehensibility, thus contributing to advancing the understanding of complex neural networks. By equally weighting the criteria (support, confidence, and lift) during the rule ranking mechanism, our approach achieved high coverage rates ranging from 94.88 % to 100 % and generated a manageable number of rules that could be easily interpreted by human experts, ranging from 5 to 10.5. These findings demonstrate the potential of our approach to significantly improve decision-making accuracy and interpretability across various real-world applications, making it a promising tool for applications in healthcare, finance, and other fields. • Extract precise rules from artificial neural networks (ANN) for improved predictions. • Promethee II ranks rules based on support, confidence, and Lift, forming a comprehensive rule set. • Eliminate invalid rules with low support, confidence, and lift to ensure a set of accurate rules. • PROMETHEE guides the genetic algorithm, reducing computational effort for Pareto-optimal solutions. • PROMETHEE's guidance enhances search efficiency, minimizing the risk of local optimal solutions. [ABSTRACT FROM AUTHOR]

Financial portfolio formation is usually a multi-objective decision-making problem concerning return and risk on the investment. In this study, we make use of an extension of regular triangular fuzzy numbers, known as coherent triangular fuzzy numbers, to describe portfolio returns in a credibility-based framework. The paper's novelty lies in proposing and deriving a crisp equivalent for computing the coherent triangular fuzzy number-based credibilistic semivariance, credibilistic skewness, and credibilistic semikurtosis. Utilizing these analytical expressions, along with additional formulations found in existing literature, we present three multi-objective portfolio optimization models involving the practical constraints related to investment decisions. All the proposed analytical expressions, when used with the returns of the portfolio as a whole, help to overcome the computationally expensive process of simulating results using the returns of individual assets. The proposed models differ with respect to different risk measures, viz. semivariance, Mean-Absolute-Semi-Deviation (MASD), and Conditional Value-at-Risk (CVaR). These models are solved using an adaptation of an efficient Multi-Objective Genetic Algorithm (MOGA) specifically designed to solve portfolio optimization problems with practical constraints. Data from the National Stock Exchange (NSE) in Mumbai, India and the New York Stock Exchange (NYSE) in New York, USA, are used to demonstrate the effectiveness of the proposed portfolio optimization models and solution methodology. All the proposed models are compared with respect to each other and the benchmarks considered in this study to bring out the performance stability. • Coherent triangular fuzzy number is used to model the portfolio problems. • Three multiobjective credibilistic portfolio optimization models are proposed. • Semivariance, MASD, and CVaR are risk measures in the proposed models. • Skewness and semikurtosis are used as higher-order moments in the model. • An adaptation of MOGA is used to solve the proposed models. [ABSTRACT FROM AUTHOR]

Electric power system applications are intricate optimization problems. Most literature reviews focus on studying an electrical paradigm through different optimization techniques. Since no review targets Metaheuristics (MHs) in electric power system applications, our work provides a general panorama of the paradigms that underlay such applications: Renewable Energies, Load Forecasting, Power Flow, Microgrids and Smart grids, and Power Quality. Our analysis revealed that the most employed MHs are Particle Swarm Optimization, Gray Wolf Optimizer, Genetic Algorithms, Cuckoo Search, and Differential Evolution. Historically, MHs have been classified into metaphor-based and non-metaphor-based. However, in some cases, this categorization does not correspond to pure MH procedures. Therefore, we also analyze MHs from a more formal perspective: their search operators. Moreover, we detected that the Renewable Energies paradigm presents a strong synergy with the remaining ones. Plus, there is a significant interest in topics related to Load-Forecasting optimization problems. Based on these data, we provide helpful recommendations for the current challenges and potential research paths. In doing so, our insights can support researchers and practitioners interested in furthering this field. • Systematic review of metaheuristics in electric power systems optimization. • Detailed taxonomy of optimization problems in electric power systems. • Identification of dominant metaheuristics for electric power system paradigms. • In-depth analysis of predominant metaheuristics from their search operators. • Discussion on challenges and opportunities in electric power systems optimization. [ABSTRACT FROM AUTHOR]

Compressing big data and model parameters via tensor decomposition such as the tensor train (TT) format has gained great success in recent years. The application of tensor compression methods requires the data be high dimensional. However, not all the real-world data primarily are high-dimensional, and sometimes reshaping is necessary before the application of tensor compression methods. Meantime, reordering and reshaping data may affect the efficiency of the compression. This work utilizes tensor reshaping to improve the efficiency of tensor compression using the TT format. An optimization model is proposed that maximizes the space-saving of tensor compression with respect to the shape of a given tensor while the compression error is bounded. The study is narrowed down to the TT decomposition and the TT-SVD algorithm is linked with a genetic algorithm (GA) to find an optimal tensor shape. The proposed method is applied to compress RGB images and a neural network to exemplify its capability. The results of the proposed tensor shape search using the GA are also compared with a purely random search. The results demonstrate that the proposed tensor shape search method significantly improves the space-saving and compression ratio of the data compression and enhances the efficiency of tensorized neural networks using the TT decomposition. • The effect of tensor reshaping is studied for the tensor train (TT) compression. • The space-saving of tensor compression is optimized with respect to the tensor shape. • A genetic algorithm (GA) is applied to solve the tensor shape optimization problem. • The proposed model is applied to compress data and tensorized neural networks. [ABSTRACT FROM AUTHOR]

Planning a personalized itinerary that satisfies the preferences and limitations of a tourist can be a complex task. The challenge is further compounded by the lack of information about tourists' tastes, which makes it difficult to ensure their satisfaction, and the impracticality of extracting this information through extensive questioning. To overcome these challenges, this paper presents a novel integrated recommender system for predicting tourists' interests and planning multi-day itineraries. The system utilizes a hybrid prediction method based on interests of similar tourists to attractions, attraction categorization, and the similarity between touristic attractions, to predict the tourist's tastes. The optimization and planning problem is formulated as a multi-day categorized orienteering problem with count-dependent profits, with the goal of finding multiple paths between points of interest to maximize the cumulative profit. As the problem is NP-hard, a hybrid meta-heuristic algorithm combining Genetic Algorithm and Variable Neighborhood Descent is proposed to solve it. The performance of the hybrid prediction method and the hybrid meta-heuristic algorithm is evaluated on the MovieLens and Tisligirides datasets. Finally, the proposed integrated recommender system is thoroughly evaluated on a real-world dataset from Tehran, demonstrating its effectiveness in recommending personalized multi-day itineraries. • Proposes integrated recommender system for multi-day tourist itineraries. • Uses hybrid prediction method based on interests, categorization, and similarity. • Formulates problem as multi-day orienteering with count-dependent profits. • Proposes hybrid meta-heuristic algorithm to solve problem. • Evaluates system on MovieLens, Tisligirides, and Tehran datasets. [ABSTRACT FROM AUTHOR]

The capacitated vehicle routing problem (CVRP) aims to find the routes that minimize transportation costs. However, in real-world applications, the equitable workload is essential for the drivers and logistics companies. For assigning an appropriate load for each vehicle, this paper formulates the CVRP with workload balancing as a multi-objective optimization problem (MO-CVRP). The goal is to simultaneously minimize the total length of routes and the variance of loads assigned to each vehicle. Although several multi-objective evolutionary algorithms (MOEAs) have been proposed for the MO-CVRP, they often suffer from slow convergence speeds due to the complexity of the search space, which hinders the identification of clear search directions. To address this issue, the knowledge transfer method is introduced that can leverage search experiences from related problems to expedite the solution process of the current problem. Specifically, a co-evolutionary genetic algorithm with knowledge transfer is proposed, where a simplified version of the traveling salesman problem (TSP) by disregarding the vehicle capacity constraint is utilized to facilitate the quick solution of the original MO-CVRP. To enable information interaction between these two problems, the widely used denoising autoencoder is adopted to construct a transformation relation between solutions from the TSP to the MO-CVRP. Additionally, an adaptive migration strategy is proposed to adaptively select the migration methods and an improved split operator strategy is designed to ensure the feasibility of the solutions. The proposed algorithm is compared with four state-of-the-art MOEAs on MO-CVRP test sets with different scales. Experimental results show that the proposed algorithm exhibits faster convergence speed and achieves solutions with superior convergence. • A co-evolutionary genetic algorithm with knowledge transfer is proposed for the VRP. • An information interaction strategy based on the denoising autoencoder is designed. • An adaptive migration strategy is proposed to adaptively select the migration method. [ABSTRACT FROM AUTHOR]

Activity-based scheduling optimization is a combinatorial problem built on the traveling salesman problem intending to optimize people schedules considering their trips and the available transportation network. Due to the difficulty of scheduling, traditional and exact methods are unable to provide appropriate solutions. Hence, new approaches have been introduced in the literature to settle these complex problems. One group of new techniques is known as metaheuristic algorithms, which provides a robust family of problem-solving methods created by mimicking natural phenomena. Although these new techniques might not find an optimal solution, they can find a near-optimal one in a moderate period. Furthermore, a myriad of novel algorithms has been introduced making it tedious for academics to select the appropriate technique. Thus, this paper investigates the contribution of metaheuristics to solve transportation-related optimization problems. To achieve this aim, we conducted a bibliometric analysis, and defined the descriptive and assessment features for 120 metaheuristics. The findings of the study reveal the usage tendencies of the algorithms, identify the most prevalent ones, and highlight those metaheuristics that have a potential use in upcoming research. The results demonstrate that the most applied metaheuristic algorithm is the genetic algorithm, but the ant colony optimization algorithm is the most popular one based on the number of citations. Lastly, we open a discussion on a few unexplored research gaps and expectations. • Review of 120 metaheuristics solving TSP-based scheduling optimization problems. • Tabular summary of descriptive and assessment features of the metaheuristics. • Automation of large amount of search terms using a programming language and an API. • GA is the most applied algorithm in publications, but ACO is the most cited one. [ABSTRACT FROM AUTHOR]

The classical autoregressive moving average (ARMA) and generalized autoregressive conditional heteroskedastic (GARCH) models have been widely adopted to forecast Value-at-Risk (VaR), the most popular risk measure in quantitative risk management. Their parameters are commonly estimated using the maximum likelihood estimation (MLE) method, which may show biasedness. This reason motivated us to require optimization through a genetic algorithm (GA) or implement artificial intelligence (AI) to improve the VaR forecast. In this paper, we aimed to compare the performances of the GA-based ARMA-GARCH model and the AI models (linear regression, support vector regression, multilayer perceptron, and long short-term memory) in forecasting the VaR. More specifically, we examined which model performs best. Using Bitcoin, crude oil, and stock index returns, we showed that the GA-ARMA-GARCH and AI models outperformed the benchmark MLE-ARMA-GARCH model. We also revealed that the AI models (particularly multilayer perceptron) integrated with kernel density estimation (KDE) performed relatively better than the GA-ARMA-GARCH model because the former resulted in a higher p -value of the VaR backtesting with a smaller error value. In addition, the cross combination between our models was found to have a higher potential to produce an improvement in the VaR forecast. • We forecasted VaR using a genetic algorithm (GA) combined with ARMA-GARCH. • We also integrated artificial intelligence (AI) with kernel density estimation (KDE). • AI models (particularly MLP) tended to outperform GA-ARMA-GARCH in forecasting VaR. • A cross combination between AI-KDE models could improve the VaR forecast. [ABSTRACT FROM AUTHOR]

Parallel island models (PIMs) are used to enhance the performance of evolutionary algorithms (EAs). In PIMs, each island executes an EA for evolving its local population, and periodically individuals migrate to neighborhoods synchronously or asynchronously. Neighborhoods are organized through a topology of communication, and migration policies guide individual exchange. This work explores migration policies over different communication topologies in synchronous and asynchronous PIMs to improve the speed-up and accuracy of genetic algorithms (GAs). The aim is to explain such models' adequacy from a general perspective, attempting to answer questions such as the best way to implement GAs in PIMs. To reach this goal, the quality of the solutions and the running time provided by proposed PIMs are evaluated over four N P -hard problems of different combinatorial nature: reversal and translocation evolutionary distance, task mapping and scheduling, and N -Queens. The experiments show that tuning the parameters of the breeding cycle and migration policies is vital to guarantee that all proposed PIMs reach good speed-ups and more accurate solutions than the sequential GA. In addition, experiments ratify that synchronous models provide the best solutions, while asynchronous models deliver the best speed-ups. Finally, the results show that no model provides either better speed-up or accuracy in general since, as for sequential models, the nature of the problem define which would be the best-adapted PIM. [ABSTRACT FROM AUTHOR]

In this paper, a new hybrid of lightning search algorithm (LSA) and non-dominated sorting genetic algorithm II (NSGAII) is proposed in order to increase the accuracy and efficiency of the conventional multi-objective NSGAII. In the proposed method, the important parameters of the NSGAII are optimized using LSA. To verify the effectiveness of the proposed algorithm, it is applied to 19 well-known standard test functions, including unconstrained functions, UF1–UF7, constrained functions, CF1–CF7, Schaffer1, Schaffer2, ZDT1, ZDT2, and ZDT3. The inverted generational distance (IGD) is used to check the efficiency of the proposed algorithm. Accordingly, the proposed algorithm is compared with the most used multi-objective algorithms. Besides, the proposed algorithm is applied to a practical case study of optimal switch, including reclosers and disconnectors, placement in the presence of distributed generation (DG) sources. The distribution networks are the Roy-Billinton test system (RBTS) and a part of the real network in Mazandaran province, in Iran. The simulation results confirm the superiority of the proposed method. For example, in practical case studies, the proposed algorithm for optimal recloser placement in the real network shows 47.09% and 4.691% cost improvement compared with multi-objective particle swarm optimization (MOPSO) and NSGAII, respectively. These improvements for RBTS-BUS2 are 3.725% and 3.502%, respectively. Also, for the RBTS-BUS2, the proposed algorithm result shows 1% and 0.732% reliability improvement compared with MOPSO and NSGAII, respectively. • A new hybrid meta-heuristic multi-objective algorithm is proposed. • Lightning search algorithm, optimizes the important parameters of NSGAII. • The proposed algorithm is validated by using of 19 test problems. • The proposed method is applied on two practical optimization problem. • Optimal recloser placement is done by using the proposed HLSA-NSGAII. [ABSTRACT FROM AUTHOR]

As the tourism industry switches from mass to personalized solutions, tourists' preferences should be considered. This study aims to optimize tourists' activity chains based on their preferences and 10 realistic parameters. A method using activity-based modeling and genetic algorithm is developed. The method handles flexible touristic activities with time window restrictions. Then scenarios including private vehicles (PV) and public transport (PT) are created. The results show high priority for security and provide significant reduction in journey times, particularly in case of flexible activities. The outcomes confirm that the proposed method provides an optimized solution to generate personalized travel itineraries that enhance tourists' travel experience. To evaluate the performance of the algorithm, Wilcoxon Signed-Ranks Test is applied. The statistical results confirm that the performance of the GA significantly differs in the various scenarios. Travelers can benefit from the proposed approach since it offers a solution to complex real-world scheduling problems. In addition, travel patterns can be derived thus supporting the work of planners and decision-makers in the tourism industry. • The aim is to optimize tourist activity chains based on preferences and parameters. • The method handles flexible touristic activities with time window restrictions. • Activity modeling and genetic algorithm is used to provide benefits for travelers. • The simulations are conducted using Matlab based on realistic touristic activities. • The flexible activities reduce travel time and enhance the utility for all scenarios. [ABSTRACT FROM AUTHOR]

The classical genetic algorithm utilizes random population initialization, an unguided crossover operator, and an unguided mutation operator for feature selection. However, this approach may be too stochastic and result in slow convergence. This paper proposes a hybrid feature selection algorithm named the Feature-Thresholds Guided Genetic Algorithm (FTGGA) to overcome this deficiency. FTGGA first employs ReliefF to filter out redundant features and retains crucial ones. Then, it generates a feature-thresholds set that contains all the feature thresholds. Each feature threshold represents the probability that the corresponding feature will be selected. The feature-thresholds set continuously updates to guide the iteration process of the genetic algorithm, accelerating its convergence. The experimental data demonstrates that FTGGA has a smaller feature subset and better classification accuracy compared to other algorithms. • A feature-thresholds guided genetic algorithm (FTGGA) is proposed to solve high-dimensional feature selection problems. • A multi-objective feature scoring mechanism is proposed to update the feature thresholds. • New crossover and mutation operators are proposed based on the guidance of feature thresholds. [ABSTRACT FROM AUTHOR]

In viticulture, training a deep learning model to assess the number of berries in a cluster image requires labeling by manually counting the berries on each grape cluster. This repeating, demanding task directly opposes the need for large amounts of data of deep learning methods. The objective of this work is the development of evolutionary conditional generative adversarial networks (GANs) for supervised data augmentation in the task of assessing berry number per cluster in grapevine. Ninety-seven grape cluster images were collected and labeled by manually counting the berries, making up the original dataset, and it was used for the training of a conditional GAN using evolutionary schemes involving a population of generators competing within a common and dynamic environment: the discriminator. After generative networks training, the best performing generator was used for supervised data augmentation: generation of labeled images conditioned to a berry number, making up the augmented dataset. Two models were trained with one dataset each, original and augmented, both having approximately the same number of samples, around 400. The original dataset was enriched with traditional image transforms, while the augmented dataset had incorporated images from the trained GAN generator. The two models were trained using the same convolutional regression network architecture, and then tested on an external image dataset, with more than 1300 images not used in any training, to compare the performance of GAN data augmentation. Results showed that the model augmented with GANs yielded lower error values, with a validation error of 43 berries, than those from the original model, with a validation error of 65 berries. In the test dataset, the augmented model obtained an error of 39 berries (R2 = 0.75), surpassing again the original model, that obtained an error of 57 berries (R2 = 0.65). These results evidence that evolutionary conditional GANs generate synthetic labeled images that lead to higher performance deep learning regression models for assessing berry number from cluster images. • Evolutionary cGANs produced synthetic images that improved CNN regression models. • Evolutionary training resulted in a stable behavior that reduced the common problems with GANs. • The augmented model successfully assessed an external dataset with more than 1300 images. • Generative models are useful tools for effective deep learning in viticulture. [ABSTRACT FROM AUTHOR]

We propose a novel neural architecture search (NAS) approach for the challenge of designing convolutional neural networks (CNNs) that achieve a good tradeoff between complexity and accuracy. We rely on Cartesian genetic programming (CGP) and integrated real-based and block-chained CNN representation, for optimization using multi-objective evolutionary algorithms (MOEAs) in the continuous domain. We introduce two variants, CGP-NASV1 and CGP-NASV2, which differ in the granularity of their respective search spaces. To evaluate the proposed algorithms, we utilized the non-dominated sorting genetic algorithm II (NSGA-II) on the CIFAR-10, CIFAR-100,and SVHN datasets. Additionally, we extended the empirical analysis while maintaining the same solution representation to assess other searching techniques such as differential evolution (DE), the multi-objective evolutionary algorithm based on decomposition (MOEA/D), and the S metric selection evolutionary multi-objective algorithm (SMS-EMOA). The experimental results demonstrate that our approach exhibits competitive classification performance and model complexity compared to state-of-the-art methods. • Evolutionary methods for NAS based on a CGP-based representation are proposed. • The CGP-encoding enables the use of off-the-shelf MOEAs to approach the problem. • Two versions, CGP-NASV1 and CGP-NASV2, differ in the granularity of the search space. • The results demonstrate our methods' effectiveness, while maintaining low complexity. [ABSTRACT FROM AUTHOR]

Capacity expansion strategy consists of fulfilling the demand with capacity portfolio via alternative configurations with long lead time for procurement and installation. Sequential dependent decisions including demand planning and cost structure of capital expenditure shall be taken into consideration for achieving enterprise profitability. Though a number of studies have been done to address related issues, limitations of existing approaches can be traced in part to the lack of a systematic framework in which multiple objectives of related total resource management problems can be considered and integrated. Little research has been done to address the present problem for capacity expansion for matured fabs from the perspective of total resource management. To fill the gaps, this study applies the concept of total resource management to integrate operational strategies and the overall usage of resources. This study develops a capacity expansion model with multiple objectives including minimizing resource costs, maximizing overall return, maximizing revenue, and minimizing capacity risk. Since the formulation of the multi-objectives can be non-linear and the size of the problem is increasing, it is difficult to solve the problem in reasonable time for practical use. A subpopulation preference adjective non-dominated sorting genetic algorithm (SPANS-GA) is developed to solve the decision problem. An empirical study in a leading semiconductor company in Taiwan is conducted for validation. The results have shown practical viability of the developed solution for multi-objective capacity planning for matured wafer fabs for total resource management. [Display omitted] • Novel capacity expansion model for total resource management is developed. • Subpopulation preference adjective non-dominated sorting GA is developed. • Multi-objective capacity expansion decision is structured for "More than Moore". • The developed solution is validated by empirical study in Taiwan. • This solution is implemented in a leading semiconductor company. [ABSTRACT FROM AUTHOR]

Evolutionary Computation algorithms have been used to solve optimization problems in relation with architectural, hyper-parameter or training configuration, forging the field known today as Neural Architecture Search. These algorithms have been combined with other techniques such as the pruning of Neural Networks, which reduces the complexity of the network, and the Transfer Learning, which lets the import of knowledge from another problem related to the one at hand. The usage of several criteria to evaluate the quality of the evolutionary proposals is also a common case, in which the performance and complexity of the network are the most used criteria. This work proposes MO-EvoPruneDeepTL, a multi-objective evolutionary pruning algorithm. MO-EvoPruneDeepTL uses Transfer Learning to adapt the last layers of Deep Neural Networks, by replacing them with sparse layers evolved by a genetic algorithm, which guides the evolution based in the performance, complexity and robustness of the network, being the robustness a great quality indicator for the evolved models. We carry out different experiments with several datasets to assess the benefits of our proposal. Results show that our proposal achieves promising results in all the objectives, and direct relation are presented among them. The experiments also show that the most influential neurons help us explain which parts of the input images are the most relevant for the prediction of the pruned neural network. Lastly, by virtue of the diversity within the Pareto front of pruning patterns produced by the proposal, it is shown that an ensemble of differently pruned models improves the overall performance and robustness of the trained networks. • We prune a Deep Learning Network using a multiobjective evolutionary algorithm, MOEA. • The MOEA guides the search considering the performance, complexity and robustness. • Most influential neurons identify parts of the image more useful for the prediction. • An ensemble of several pruned models achieves the best performance and robustness. [ABSTRACT FROM AUTHOR]

The main objective of this study is to formulate and solve the model of non-linear fixed-charge transportation problem (NFCTP), which is one of the NP-hard problems. This type of problem is challenging to solve using conventional methods because these methods are usually inefficient, costly to compute, and may stuck at the local optimum solution. To overcome this issue, a non-linear particle swarm optimization (NPSO) algorithm with new non-linear acceleration parameters is proposed in this study. Also, to preserve the feasibility condition (non-negative integer solution) of the transportation problem, two novel negative repair and fraction repair strategies have been incorporated into the proposed NPSO. The efficiency of the NPSO algorithm is tested on both the small-scale and large-scale NFCTPs. The dataset are considered from the existing studies. The obtained results are compared with those obtained by the spanning tree-based genetic algorithm (st-GA), priority-based genetic algorithm (pb-GA), and minimum cost flow-based genetic algorithm. The comparative study reveals that for all the considered problems, the proposed NPSO provides better feasible solutions in lesser computational time. Furthermore, the effectiveness of the NPSO algorithm is shown by comparing it to other seven existing variants of PSO and is found to outperformed these variants for the small scale as well as large scale NFCTPs. • A non-linear programming model for fixed-charge transportation problem is developed. • A new algorithm known as a non-linear particle swarm optimization is proposed. • The performance of the proposed algorithm is evaluated on some benchmark problems. • The obtained results are compared with the existing evolutionary approaches. [ABSTRACT FROM AUTHOR]

Energy efficiency is a critical issue in the management and operation of data centers, which form the backbone of cloud computing. Virtual machine placement has a significant impact on the energy efficiency of virtualized data centers. Among various methods to solve the virtual-machine placement problem, the genetic algorithm has been well accepted for its quality of solution. However, it is computationally demanding largely due to the complex form of fitness function, limiting its applications in data centers. To enhance the computational efficiency of the genetic algorithm while maintaining its quality of solution, a progressive-fidelity approach is developed in this paper for genetic-algorithm computation. It starts with a low-fidelity genetic algorithm with a simple fitness function. Then, for solution refinement, it switches to a medium-fidelity genetic algorithm with a more complicated fitness function. Finally, it progresses to the fine tuning of solution through a high-fidelity genetic algorithm with the energy consumption of data centers as fitness function. Heuristics are presented for the adaptive switching of genetic-algorithm computation from low fidelity to medium fidelity and finally to high fidelity. Experiments show that compared with the standard genetic algorithm of high fidelity, our progressive-fidelity approach of genetic algorithm computation is 50% faster for large-scale data centers while maintaining similar quality of solution in terms of the energy consumption of data centers. • A progressive-fidelity approach developed for speeding up Genetic Algorithm. • 3 different fitness functions included for a quick and fine solution. • Heuristics are presented for the adaptive switching in our approach. [ABSTRACT FROM AUTHOR]

Researchers have been studying genetic algorithms (GAs) extensively in recent decades and employing them to address extremely challenging combinatorial optimization problems (COPs). Although GAs achieve superior performance, they are less efficient because most GAs are designed manually without intelligent parameter configuration to support scalable problem-solving strategies and learnable evolutionary operators. To address this issue, machine learning (ML) techniques have been integrated with GAs for operator and parameter selection, however, few studies have focused on intelligent genetic locus selection for influential operators in GAs. To fill this gap, this paper proposes an intelligent genetic locus selection algorithm that serves as the foundation of parameter configuration for critical operators. With the established framework, the Cross Information Synergistic Attention (CISA) model and the n-step proximal policy optimization (PPO) have been utilized to intelligently select the appropriate genetic locus for the most influential phase, i.e., crossover, during the evolutionary process. The proposed NeuroCrossover algorithm is validated on extensive COPs, including the traveling salesman problem, capacitated vehicle routing problem, and bin packing problem. The results demonstrate the efficiency and effectiveness of our algorithm, which outperforms other methods in terms of solution quality, convergence speed, and generalization. For instance, with the CISA, the average percentage gaps of our algorithm are 3.64% and 6.38% for instances in TSPLIB and CVRPLIB, respectively, obtaining gains of about 0.47% and 1.20% compared to those of GA. The proposed algorithm provides a novel solution to lead GAs to an efficient search and improve their scalability and learnability. • The intelligent genetic locus selection scheme for the crossover phase in GAs is formulated as an MDP. • Cross Information Synergistic Attention for better perceiving the consistency and avoiding incompatible correlations. • A n-step PPO-based genetic locus selection algorithm (NeuroCrossover) is proposed, which significantly enhances the GA. • Superior performances on solution quality, convergence speed and generalization with the new algorithm. [ABSTRACT FROM AUTHOR]

Physics-Informed Neural Networks (PINNs) are artificial neural networks that encode Partial Differential Equations (PDEs) as an integral component of the ML model. PINNs are successfully used nowadays to solve PDEs, fractional equations, and integral–differential equations, including direct and inverse problems. Just as in the case of other kinds of artificial neural networks, the architecture, including the number and sizes of layers, activation functions, and other hyperparameters can significantly influence the network performance. Despite the serious work in this field, there are still no clear directions on how to choose an optimal network architecture in a consistent manner. In practice, expertise is required, with a significant number of manual trial and error cycles. In this paper, we propose PINN/GA (PINN/Genetic Algorithm), a fully automatic design of a PINN by an evolutionary strategy with specially tailored operators of selection, crossover, and mutation, adapted for deep neural network architecture and hyperparameter search. The PINN/GA strategy starts from the population of simple PINNs, adding new layers only if it brings clear accuracy benefits, keeping PINNs in the population as simple as possible. Since the examination of dozens of neural networks through the evolutionary process implies enormous computational costs, it employs a scalable computational design based on containers and Kubernetes batching orchestration. To demonstrate the potential of the proposed approach, we chose two non-trivial direct problems. The first is 1D Stefan transient model with time-dependent Dirichlet boundary conditions, describing the melting process, and the second is the Helmholtz wave equation over a 2D square domain. The authors found that PINNs accuracy gradually improves throughout the evolutionary process, exhibiting better performance and stability than parallel random search and Hyperopt Tree of Parzen Estimators, while keeping the network design reasonably simple. • PINN/GA evolutionary strategy for optimizing architecture and hyperparameters. • GA operators specially tailored for PINNs. • Search strategy designed to keep PINN architecture as simple as possible. • Distributed PINN/GA engine reduces search time for almost two orders of magnitude. • Decreases PINN error for 1D Stefan and Helmholtz problems by an order of magnitude. [ABSTRACT FROM AUTHOR]

Research on the assembly line balancing problem has been continuously incorporated the characteristics of actual production lines without considering the mixed-model two-sided assembly line with unpaced synchronous transfer, which is common in body shops of cars or trucks. This study attempts to model a mixed-model two-sided assembly line balancing and sequencing problem with unpaced synchronous transfer, and designs a genetic algorithm to solve this problem. First, to describe the process whereby model sequence and its execution, station and its number, and the process time caused by task allocation lead to the variable output of an unpaced synchronous line, a model without an observation point restriction is proposed. Based on this, a mathematical model of a balancing and sequencing problem of a mixed-model two-sided assembly line with an unpaced synchronous transfer is established, in which the balancing problem is aimed at Type-II. Then, an improved genetic algorithm adapted to the complexity of the problem is proposed. The algorithm designs a combination and evaluation mechanism to ensure that the solution populations of the balancing and sequencing problems are sufficiently combined and evaluated in each iteration to improve efficiency. The proposed algorithm is illustrated using a small numerical example. Finally, the algorithm is experimentally investigated using a set of test cases and several comparison algorithms. The experimental results show that the combination and evaluation mechanism is suitable for these problems, especially for large-sized cases, and that the proposed algorithm is effective and has better performance than the comparison algorithms. • To the best of our knowledge, this is the first study of mixed-model two-sided assembly line balancing problem with unpaced synchronous transfer. • A model without observation point restriction is proposed for the variable output of unpaced synchronous line. • A new genetic algorithm adapted to the complexity of the problem is proposed. • Extensive experiments are carried out to illustrate the effectiveness and good performance of the proposed algorithm. [ABSTRACT FROM AUTHOR]

This article investigates the transportation task of Health-care waste (HCW). The particularity of HCW, compared to ordinary waste, is that it may contain infectious and/or toxic substances. The disposal of this waste involves transporting it from healthcare facilities (HCF) to treatment centers. This step can be threatening to humans and the environment if not managed with the utmost care. To ensure the safe management of HCW, the separation policy between dangerous and normal waste should be honored. Hence, a vehicle routing problem with multi-compartment vehicles is adopted with the underlying objective of minimizing the total travel distance. Given the NP-hardness of the problem, an adaptive evolutionary algorithm is proposed based on the results of pre–post statistical tests of the population. The proposed algorithm is compared to the best so far solutions on medium and large instances using benchmarks from the literature. The experimental study confirms the efficiency of the proposed approach against the competing algorithms. Finally, a case study is provided to illustrate the managerial insights of the studied problem. • A MILP model based on the multi-compartment vehicle routing problem is adapted to the HCW. • An Adaptive Genetic Algorithm is proposed to solve the problem. • Simulation results are compared to optimal solutions. • The proposed algorithm is compared to the best so far solutions on medium and large instances using benchmarks from the literature. • A case study is provided to illustrate the managerial insights of the studied problem. [ABSTRACT FROM AUTHOR]

This paper presents a general method for assessment of the performance of a genetic algorithm (GA) in cases where the global optimum of the objective function is unknown. The method involves discretization of the search space, making it possible to apply a brute force calculation to find the global optimum for the discretized case. Then, this method is used to study the performance of a GA applied to the problem of speed profile optimization for heavy-duty vehicles, in which the optimization must be carried out within a rather short time. In this performance analysis, the discretization involves generating speed profiles as piecewise linear functions. It is demonstrated that the GA is able to find near-optimal solutions for the cases considered here: The speed profiles generated by the GA have objective function values that are typically within 2% of the global optimum. • Search space discretization can make benchmarking of a genetic algorithms possible. • Finding the exact optimum is very time-consuming but possible for comparison purposes. • Fuel-efficient driving can be formulated as a speed profile optimization problem. • With piecewise linear profiles, the exact optimal fuel consumption can be computed. • Near-optimal speed profiles can rapidly be found using a standard genetic algorithm. [ABSTRACT FROM AUTHOR]

In the present study, strength of meta-heuristic computing techniques is exploited for estimation problem of Hammerstein controlled auto regressive auto regressive moving average (HCARARMA) system using differential evolution (DE), genetic algorithms (GAs), pattern search (PS) and simulated annealing (SA) algorithms. The approximation theory in mean squared error sense is utilized for construction of cost function for HCARARMA model and highly uncorrelated adjustable parameter of the system is optimized with global search exploration of DE, GAs, PS and SA algorithms. Comparative study is carried out from desired known parameters of the HCARARMA model for different degree of freedom and noise variation scenarios. Performance analysis of the DE, GAs, PS and SA algorithms is conducted through results of statistics based on sufficient large independent executions in terms of measure of central tendency and variation for both precision and complexity indices. The exhaustive simulations established that the population-based heuristics are more accurate than single solution-based methodologies for HCARARMA identification. • Application of computational paradigm for parameter estimation problem of HCARARMA systems. • The population and single solution based algorithms are exploited for HCARARMA identification. • Validation on the basis of statistical performance indices for different noise scenarios and model dynamics. • Stability, reliability, extendibility, and wider applicability are other worthy perks of the scheme. [ABSTRACT FROM AUTHOR]

Ground Penetrating Radar (GPR) is an electromagnetic sensing technology employed for localization of underground utilities, pipes, and other types of objects. The radargrams typically obtained have a high dimensionality, containing a number of signatures with hyperbolic pattern shapes, and can be processed to retrieve information about the target's locations, depths and material type of underground soil. The classical Hough Transform approach used to reconstruct these hyperbola shapes is computationally expensive, given the large dimensionality of the radargrams. In literature, several approaches propose to first approximate the location of hyperbolas to small segments through a classification stage, before applying the Hough transform over these segments. However, the published classifiers designed for this task present a relatively complex architecture. Aiming at an improved target localization, we propose an alternative classification methodology. The goal is to classify windows of GPR radargrams into two classes (with or without target) using a neural network radial basis function (RBF), designed via a multi-objective genetic algorithm (MOGA). To capture samples' fine details, high order statistic cumulant features (HOS) were used. Feature selection was performed by MOGA, with an optional prior reduction using a mutual information (MIFS) approach. The obtained results demonstrate improvement of the classification performance when compared with other models designed with the same data and are among the best results available in the literature, albeit the large reduction in classifier complexity. [ABSTRACT FROM AUTHOR]

In this work, a novel application of bio-inspired computational heuristic paradigm is presented for micropolar fluid flow and heat transfer system in a channel with permeable walls by modeling competency of neural networks, global search of genetic algorithms, and rapid local convergence of sequential quadratic programming. Approximation theory in the mean squared error sense is exploited for the formulation of an objective function to solve the governing nonlinear fluidics system. The designed scheme is employed to study the dynamics of the model in terms of stream function, microrotation, concentration, and temperature profiles for prominent factors based on Reynolds number, Peclet number for diffusion of heat and mass, coupling, spin-gradient viscosity, micro inertia density parameters. The consistency and robustness of the solver are validated through statistical performance indices based on comparison with state of art Adams numerical method for accuracy and complexity measures. • A novel integrated heuristic is designed for a system of micropolar fluidic model. • Consistent accuracy is achieved in concurrence with the Adams numerical method. • Validation of the accuracy and convergence is certified through results of statistics. • Applicability and smooth implementation are other creditable attractions of the scheme. [ABSTRACT FROM AUTHOR]

Abstract In smart cities, when the real-time control of traffic lights is not possible, the global optimization of traffic-light programs (TLPs) requires the simulation of a traffic scenario (traffic flows across the whole city) that is estimated after collecting data from sensors at the street level. However, the highly dynamic traffic of a city means that no single traffic scenario is a precise representation of the real system, and the fitness of any candidate solution (traffic-light program) will vary when deployed on the city. Thus, ideal TLPs should not only have an optimized fitness, but also a high reliability, i.e., low fitness variance, against the uncertainties of the real-world. Earlier traffic-light optimization methods, e.g., based on genetic algorithms, often simulate a single traffic scenario, which neglects variance in the real-world, leading to TLPs not optimized for reliability. Our main contributions in this work are the following: (a) the analysis of the importance of reliable solutions for TLP optimization, even when all traffic scenarios are consistent with the real-world data and highly correlated; (b) the adaptation of irace , an iterated racing algorithm that is able to dynamically adjust the number of traffic scenarios required to evaluate the fitness of TLPs and their reliability; (c) the use of a large real-world case study for which real-time control is not possible and where data was obtained from sensors at the street level; and (d) a thorough analysis of solutions generated by means of irace , a Genetic Algorithm, a Differential Evolution, a Particle Swarm Optimization and a Random Search. This analysis shows that simple strategies that simulate multiple traffic scenarios are able to obtain optimized solutions with improved reliability; however, the best results are obtained by irace , among the algorithms evaluated. Highlights • Adaptation of an elitist iterated racing with the goal of obtaining reliable traffic light programs. • Thorough analysis of solutions generated by means of 5 algorithms: IRACE, a GA, a DE, a PSO and a RS. • Use of a real-world case study which comes from sensors at the street level. • Future algorithms should consider the reliability of solutions over multiple traffic scenarios. • The hybridization of iterated racing strategies with evolutionary algorithms is one of the most promising directions. [ABSTRACT FROM AUTHOR]

Abstract Current evolutionary many-objective optimization algorithms face two challenges: one is to ensure population diversity for searching the entire solution space. The other is to ensure quick convergence to the optimal solution set. In this paper, we propose a novel two-archive strategy for evolutionary many-objective optimization algorithm. The uniform archive strategy, based on reference points, is used to keep population diversity in the evolutionary process, and to ensure that an evolutionary algorithm is able to search the entire solution space. The single elite archive strategy is used to ensure that individuals with the best single objective value are able to evolve into the next generation and have more opportunities to generate offspring. This strategy aims to improve the convergence rate. Then this novel two-archive strategy is applied to improving the Non-dominated Sorting Genetic Algorithm (NSGA-III). Simulation experiments are conducted on benchmark test sets and experimental results show that our proposed algorithm with the two-archive strategy has a better performance than other state-of-art algorithms. Highlights • A uniform archive is proposed to ensure algorithms searching entire solution space. • A single elite archive is used to keep the best solution participating in evolution. • NSGA-III-UE is proposed which combines the two-archive strategy. [ABSTRACT FROM AUTHOR]

Abstract Regression problems try estimating a continuous variable from a number of characteristics or predictors. Several proposals have been made for regression models based on the use of fuzzy rules; however, all these proposals make use of rule models in which the irrelevance of the input variables in relation to the variable to be approximated is not taken into account. Regression problems share with the ordinal classification the existence of an explicit relationship of order between the values of the variable to be predicted. In a recent paper, the authors have proposed an ordinal classification algorithm that takes into account the detection of the irrelevance of input variables. This algorithm extracts a set of fuzzy rules from an example set, using as the basic model a sequential covering strategy along with a genetic algorithm. In this paper, a proposal for a regression algorithm based on this ordinal classification algorithm is presented. The proposed model can be interpreted as a multiclassifier and multilevel system that learns at each stage using the knowledge gained in previous stages. Due to similarities between regression and ordinal problems as well as the use of a set of ordinal algorithms, an error interval can be returned with the regression output value. Experimental results show the good behavior of the proposal as well as the results of the error interval. Highlights • We present a new fuzzy rule learning algorithm for regression problems. • The algorithm is based on an ordinal one which uses the sequential covering strategy with a genetic algorithm as the search mechanism. • The final regression algorithm is a multiclassifier and multilevel system which learns in an incremental way by using the learnt knowledge in previous stages. • An error interval can be returned with the regression output value. • The results show a good behavior compared with the same category algorithms, so as, other category algorithms, for regression problems. [ABSTRACT FROM AUTHOR]