Search

Improving object recognition in computer vision by combining the Whale Optimization Algorithm (WOA) with Seagull Optimization Algorithm (SOA), two nature-inspired metaheuristic optimization techniques. Object detection plays a critical role in automated object recognition and localization in images and videos. The ResNet architecture has established its effectiveness in learning deep representations, while convolutional neural networks (CNNs) have demonstrated exceptional object detection performance. Both the WOA and the Seagull Algorithm are well-known for their unique exploration and exploitation skills, inspired by natural actions. By leveraging the complementary capabilities of these algorithms, the proposed approach aims to enhance the accuracy and efficiency of object identification in deep learning models. The WOA specializes in global exploration and convergence, mimicking the hunting thoughts of humpback whales. Meanwhile, the Seagull Algorithm simulates the swarming behavior of seagulls, focusing on local search and exploitation. Extensive studies on popular object detection are carried out to validate the efficacy of the suggested technique benchmarks, including KITTI, COCO, CUSTOM and PASCAL VOC datasets. The results demonstrate that the fusion of the Seagull and Whale Optimization Algorithm with the ResNet model consistently outperforms traditional optimization techniques and baseline object detection methods. The algorithm exhibits improved accuracy, robustness, and generalization on various challenging scenarios, showcasing its potential for real-world applications. [ABSTRACT FROM AUTHOR]

Wind energy (WE) is a clean and sustainable alternative to traditional forms of energy production. Wind-power generating systems are perhaps the most viable near-term sources of energy to address the worldwide energy dilemma. In this study, Meta-heuristic optimization techniques (MOTs) are used for enhancing the dynamic behavior of the blade pitch control (BPC) system of wind turbine (WT) and output power at desired level by controlling the WT blade pitch angle, rotational speed, and mechanical torque. Thus, the proportional-integral-differential (PID) controller is used to regulate the multivariable BPC of wind turbine system linked large power system. In order to do this, the BPC system parameters are optimally tuned using the whale optimization algorithm (WOA) and the grey wolf optimizer (GWO). To confirm the superiority of WOA-PID and GWO-PID over traditional techniques like the simplex algorithm (SA) and the Ziegler-Nichols (ZN), a thorough comparison analysis is provided. The time response parameters, such as maximum overshoot, settling time, and steady-state error, are used for this comparison analysis. The simulation results show that the MOTs-PID are more efficient than the conventional methods. It can be inferred from the simulation findings that under various operating situations, the WOA-PID controller offers sufficient tracking of the reference signals for the electrical torque. The robustness of the BPC system is studied under various operating conditions of torque, wind speed variation, and system parameter change. The MOTs-PID provides more robustness than the conventional optimization algorithms. [ABSTRACT FROM AUTHOR]

The energy consumption in the building operation stage accounts for about 20% of the total energy consumption of the building's whole life cycle, among the energy consumption, the air-conditioning system accounts for the largest proportion, significantly contributing to 50% of the energy consumption. In the conventional method in the air conditioning control system, a constant target temperature is usually set. However, due to the real-time changes in environmental parameters and personnel activities, the energy consumption of buildings and the comfort of users will also change accordingly. This paper studies the optimization of air conditioning energy consumption under dynamic conditions, comprehensively adopts technologies such as the Internet of Things (IoT), computer simulation, and meta-heuristic algorithms to adjust the control parameters of the air conditioner in real-time according to the on-site situation, to achieve the goal of reducing energy consumption and improving comfort. The building air-conditioning energy-saving optimization solution proposed in this paper takes into account real-time, adaptability and accuracy. According to the actual comparison, the energy consumption is reduced by about 10%, and the comfort satisfaction degree is increased by about 8%. [ABSTRACT FROM AUTHOR]

Location-routing problem is a problem that integrates both facility location problem and vehicle routing problem. The problem is aimed to solve the allocation of customers to the established depots and optimize the distance between the open depot and the assigned customers. This study considers multi-objective where the first objective is minimizing the cost of operating depot and cost of travel distance, meanwhile in the second objective is maximizing the profit from customer's returns. As the problem is known as NP-hard problem and have the conflict objectives, metaheuristic method which is Multi-objective Harmony Search Algorithm (MOHSA) was proposed. The multi-local search neighborhood technique is applied in this algorithm. A numerical analysis is conducted in simulated dataset to find a set of Pareto front by non-dominated sorting genetic algorithm II (NSGA-II) in MOHSA using both fuzzy membership approach and the trade-off ranking method. The best method is determined by the evaluation of performance metric. The results show that the fuzzy membership approach can gives the better compromise solution as compared to the trade-off ranking method. [ABSTRACT FROM AUTHOR]

The utilization of electrical energy in the Radial Distribution System (RDS) is increasing day by day due to various industrial, commercial, and residential loads. The increasing load demand shows negative impact in terms of degrades the voltage profile, voltage stability and increase the power loss of RDS. Negative impacts are counteracted by placement of solar and wind DGs connected to DS in an optimal way. In this article an efficient technique is used for proper allocation of DGs by using Particle swarm optimization and differential evolution (PSODE). Different cases are considered for the analysis. Further, the effectiveness of the proposed algorithm was tested by IEEE 85 bus system, and the simulated results are compared with other Metaheuristic algorithms. [ABSTRACT FROM AUTHOR]

Development of online social networks (OSN_models) significantly increasing over the years. Data from social networks has been shared publicly for the purpose of deeper understanding user behaviour and data mining tasks. However, securing individual identification from the publicly published data is the primary issue in sharing social network databases. K-anonymization is a well-known technique for preventing individual identification of publically available datasets. According to this method, each record in the anonymized dataset must be identical to k attributes assigned within the search. The trained data is initially utilized to make an accurate model that meets the K-anomaly detection criteria. Recently, clustering methods have been used to apply K-anonymization to the social dataset successfully. This paper applies a swarm-based clustering technique, called Whale Optimization Algorithm (WOA) for clustering social data. The WOA algorithm replicates humpback whale behavior and employs fewer internal parameters to support a better balance of exploitation and exploration. The method is tested on three publicly available data sets: Yelp, Facebook, and the adult dataset. The proposed model using clustering scheme improves the detection accuracy in terms of metric score is evaluated here. The error rate is significantly reduced that satisfy the similarity issues produced by the existing frameworks. [ABSTRACT FROM AUTHOR]

This research suggested a deep convolution neural network (DCNN) based technique towards image denoising that has been strengthened by a Whale-Penguin Optimization Algorithm (WPOA) which drew inspiration from nature. Deep convolution neural networks are first fed noisy images, and to circumvent training difficulties and promote network convergence, they use batch normalization and residual learning approaches. The developed network also leverages dilated convolutions to broaden the relevant information and minimize the computational expense. Additionally, the Whale - Penguin Optimization Algorithm (WPOA), an amalgamation of the Whale Optimization Algorithm (WOA) as well as the Penguin Search Optimization Algorithm (PeSOA), has been the envisaged algorithm that facilitates in the neural network in training to recognize the overall optimal value. During both the exploration and development stages, WPOA has a rapid convergence rate. Finally, the optimized Convolution Neural Network (CNN) surpasses the far more advanced noise removal approaches. In comparison to various image denoising processes that have been previously suggested, the proposed method is tested employing benchmarking tool notably peak signal to noise ratio (PSNR). [ABSTRACT FROM AUTHOR]

The main motive of this research is to design an efficient image encryption method. To achieve this goal, the chaotic tent map algorithm is taken into consideration for secret key generation. In the tent map algorithm, the control parameter μ plays an important role for key generation. Therefore, the optimal value of μ parameter is determined using the JAYA algorithm. JAYA is a metaheuristic algorithm that determines the optimal value by putting the random value of μ parameter in the tent map algorithm based on the objective function. The beauty of the JAYA algorithm is that it does not require any parameter tuning, is simpler, and provides a better convergence rate to find the optimal solution. After determining the optimal parameter value, a secret key is generated. Further, exclusive-OR and circular shifting are performed for image encryption. The performance analysis is performed on different images. The subjective analysis provides a better noisy image, and the objective analysis shows a low PSNR, a high correlation coefficient, a high value of entropy, a maximum deviation, and a mean square error. [ABSTRACT FROM AUTHOR]

Energy consumption and network lifetime are significant issues for a WSN. Numerous optimization techniques can be employed to select a path between nodes at source and destination. The challenge is in establishing a routing algorithm which takes into account the crucial factors of reducing the energy use and enhancing network longevity. This study article introduces a broad categorization of WSN optimization methods for energy-aware routing. This review's objective is to present several decentralized energy-aware routing strategies that are appropriate for most WSN applications that demand reducing energy use and improving network lifetime. The study is then concluded with a comparative analysis that highlights the key problems and possible future paths. [ABSTRACT FROM AUTHOR]

Fibre Wireless (Fi-Wi) access technology integrates the existing broadband access technology and wireless access technology to fulfil the users' demand for better Internet speed in "anytime anywhere" approach and cost-efficient manner. It is designed to make the best use of their advantages in terms of vast bandwidth, mobility, and cost effectiveness. However, there remain there remain several open and challenging issues that require concentrated research efforts to build such an access technology. ONU placement problem is one among these issues as ONU placement in FiWi plays a critical role in overall network deployment cost and resource optimization. In this work, we present performance analysis of modified wale optimization algorithms by considering different of scenarios for ONUs and wireless routers in a FiWi network. We also compare present throughput gain for the algorithms under varied ONU and router deployment. [ABSTRACT FROM AUTHOR]

In Handwritten Character Recognition (HCR), interest in feature extraction has been on the increase with the abundance of algorithms derived to increase the accuracy of classification. In this paper, a metaheuristic approach for feature extraction technique in HCR based on Crow Search Algorithm (CSA) was proposed. Freeman Chain Code (FCC) was used as data representation. The main problem in representing a character using FCC is that the results of the extractions depend on the starting points that affected the route length of chain code. To solve this problem, the metaheuristic approach via CSA was proposed to find the shortest route length and minimum computational time for HCR. The performance measurements of the proposed CS-FCC extraction algorithm are the route lengths and computation times. The experiments on the algorithms are performed based on the chain code representation derived from established previous works of Center of Excellence for Document Analysis and Recognition (CEDAR) dataset which consists of 126 upper-case letter characters. Based on the result, the proposed CS-FCC obtained 1880.28 in term of route length and only needs 1.10 second to solve the whole character images. [ABSTRACT FROM AUTHOR]

Infants crying patterns are a means to communicate with the external world. Infants problems can be explored through their cry within first year. Significant changes in crying patterns give an idea about the neurological disorders, genetic problems and many more. The infant cry signal analysis is carried out to identify the cry causes like hunger, pain, discomfort, anxiety, etc. Different causes of infant crying are characterized by signals in the cry segments. In the proposed work a whale optimization algorithm is implemented to detect the emotion of the crying infant. The previous works include various algorithms for classification, however the novelty in this work can be attributed to processing only voiced part of the cry signal. Feature extraction techniques that include Melcepstral frequency coefficients (MFCC) are used to study and analyze the reason. Infant cry signal is analyzed for knowing the reason as well as for diagnosing pathological conditions because the signals contain more information about physical and physiological conditions. The database has been obtained from the neonatal ward of Harshita Hospital, Madurai. The algorithm has been tested on the test dataset consisting of 125 samples. The performance of the proposed method has been evaluated and the efficiency is found to be 94%as compared to 90% efficiency. [ABSTRACT FROM AUTHOR]

Support Vector Machine (SVM) modelling approach focuses on dimensional reduction with a powerful generalization capability to solve problems such as non-linearity, and local extremes. However, the ability of SVM to learn and generalize depends on the collection of acceptable parameters that directly affect the output of the model. In addition, deploying relevant feature sets is also an issue in obtaining optimal classification. Thus, this work deploys eight swarm intelligence algorithms to improve the SVM classifier's accuracy by conducting two experiments: feature selection and tuning SVM parameters using the obtained feature set. The deployed swarm intelligence algorithms include Particle swarm optimization (PSO), Genetic Algorithm (GA), Firefly Algorithm (FA), Salp Swarm Algorithm (SSA), Bat Algorithm (BA), Sine Cosine Algorithm (SCA), Whale Optimization Algorithm (WOA), and Multi Verse Optimization (MVO). The undertaken experiments were on six medical benchmark datasets that are of various dimensions. Evaluation was based on two metrics: classification accuracy and size of feature set. This work's significance is to present the comparison and determine which swarm algorithm is the best to be used in a large size medical dataset. The experimental results support existing literature that noted the swarm intelligence are useful feature selection as well as optimizing SVM parameter while maintaining acceptable accuracy. [ABSTRACT FROM AUTHOR]

Resource allocation in the IoT-Fog environment is a challenging and critical problem with profound implications for application performance and service provider profitability. Efficiently distributing tasks across fog nodes enhances Quality of Service (QoS) metrics like latency for application users and reduces network resource utilization for service providers. This paper proposes an elite version of the meta-heuristic algorithm multi-objective grey wolf optimization (E MOGWO) for better resource allocation in the IoT-Fog environment. The proposed algorithm was validated with benchmarking functions and simulated with the YAFS simulator. The results show superior results in both evaluations. [ABSTRACT FROM AUTHOR]

Nowadays, because of the high demand for car transportation, not everyone has the purchasing power of cash, and people sometimes pay for it on credit. ACC is a finance company that acts as a facilitator for car loan financing. ACC collects car loans through collectors who are assigned an area in Medan City and also the address of the customer who will be billed in installments. To maximize the installment collection process, the collector's journey needs to be optimized. A metaheuristic algorithm can solve the collector-collection trip optimization problem. The metaheuristic algorithm used to solve this problem is the Steepest Ascent Hill Climbing algorithm and the genetic algorithm to find the optimum route from the collector billing path. After the two algorithms are implemented and tested, the final result of this research is the complexity of the Steepest Ascent Hill Climbing algorithm is θ(n) with a running time of 429.1936 ms on a 5-point test, while the complexity of the genetic algorithm is θ(n3) with a running time of 8921.8021 ms on a 5-point test. [ABSTRACT FROM AUTHOR]

The aim of this study is to present the effectiveness of two algorithms, namely Simulated Annealing (SA) and Grey Wolf Optimization (GWO), in the optimization of the water-abrasive cutting process. The aim was to maximize the thickness of the cut by accurately predicting the optimum parameters of the water-abrasive cutting process, including the nozzle diameter, the concentration of the abrasive, and the feed rate. The objective function for the optimization has been derived using the Response Surface methodology. The results highlight the potential utility of the SA and GWO algorithms in the effective solution of optimization problems in the context of waterjet cutting processes. [ABSTRACT FROM AUTHOR]

The paper presents the application of an ant lion meta-heuristic algorithm to optimize the parameters of the water-abrasive cutting process of white marble. The properties of target and abrasive materials were described. Details of the ant lion behavior that the optimization algorithm was inspired by are also provided. Both the high-pressure abrasive jet material cutting technology and the details of the ant lion algorithm are also described. Based on the control parameters of the process of marble cutting by AWJ, the cost function was determined using the response surface method (RSM). For the ant lion algorithm, the process of optimization of control parameters was conducted for the objective function to reach the minimum value. Based on the optimization the by ant lion method, a set of control parameters and the achieved extremes were determined and compared with laboratory results. Details of the quality of the surface cut under optimal conditions along with roughness parameters are also presented. [ABSTRACT FROM AUTHOR]

The multi-unit production planning problem is one of the most challenging optimization tasks with 270 decision variables and multiple constraints. An enhanced social group optimization (ESGO) algorithm to enhance the population diversity of the improving phase and extend of the global scope of exploration through a novel emulation phase in addition to the acquiring phase is developed. ESGO is compared with the original SGO, three of its recent variants and ten modern swarm-intelligent meta-heuristics and delivered better optimality with greater robustness and lower computational times for all the eight cases tested against. [ABSTRACT FROM AUTHOR]

This state-of-the-art review paper provides a comprehensive overview of meta-learning techniques for learning to learn in machine learning. Meta-learning called "learning to learn," is an approach that enables models to learn from new tasks learning quickly and efficiently, based on prior knowledge and experience. The paper discusses various meta-learning techniques, such as model-based meta-learning, metric-based meta-learning, optimization-based meta-learning, and memory-based meta-learning. The review also covers the applications of meta-learning in different domains, including natural language processing, computer vision, and reinforcement learning. Additionally, the paper discusses the benchmark datasets, challenges along with the future directions of meta-learning research, highlighting the need for developing more efficient and scalable meta-learning algorithms. Moreover, the proposed Meta-Learn framework incorporates a hierarchical structure to facilitate knowledge transfer across multiple levels of abstraction, and it also includes a memory component that enables the system to store and reuse previous experiences. Overall, this review aims to provide a comprehensive overview of Meta-Learning techniques for learning to learn in Machine Learning and proposes a novel Meta-Learn framework that could potentially enhance the performance and scalability of Meta-Learning algorithms. [ABSTRACT FROM AUTHOR]

Air pollution is a serious problem encountered by most Indonesian people. The world air quality report published in 2018 by IQAir listed Indonesia to be in the 11th rank as one of the countries with the worst air quality in the world as indicated by the average pollution concentration of 42 (µg/m3). A straightforward solution to overcome this problem is to reduce the harmful gas emission content released into the air by an internal combustion engine. This research then aimed to optimise the performance of CAT 3401 internal combustion diesel engine using two sequential stages. The first stage was Backpropagation Neural Network (BPNN) method. Then, the effect of various input parameters (compression ratio, start of injection angle, fuel injection pressure, and exhaust gas recirculation) on the diesel engine performance was examined in the first stage. The resulting analytical modelling produced by the Backpropagation Neural Network (BPNN) method was further used to optimise the diesel engine's performance. Next, in the second stage, two well-known metaheuristic methods, namely Genetic Algorithm (GA) and Particle Swarm Optimization (PSO), were used to achieve the highest peak pressure and minimal NOx gas emission and soot content. [ABSTRACT FROM AUTHOR]

The best switching angles are chosen in this study with Artificial Bee Colony (ABC), Genetic algorithms (GA), and Gray wolf algorithm (GWO), slime moiled algorithm (SMA), Whale optimization algorithm (WOA). These angles are selected according to the lowest total harmonic distortion of output load voltage from reduced switches multilevel inverter. These algorithms are used in a hybrid method to solve the transcendental equations for determining switching angles. A 27-level inverter fed by isolated unequal PV panel as DC sources with reduced switches and sources is chosen for this study. Theoretical analysis and Simulation are accomplished using Matlab/Simulink for 27 level reduced switches multilevel inverter. The simulated results validated the practical outcomes. [ABSTRACT FROM AUTHOR]

Harmonic Search is one of the well-known metaheuristic algorithms and has been used in various fields as a tool to obtain optimization solutions. But sometimes, researchers need to modify and improve the original algorithm once it is used in their respective fields to get the best results. Modified Harmonic Search introduced a phase in generating new solution vectors to improve the Harmony Search algorithm in terms of the rate of convergence. Modified Harmony Search is still a derivative-free real parameter optimization algorithm. This paper presents the scalability study of three types of Harmony Search namely original Harmony Search, Global Harmony Search and Modified Harmony Search against ten benchmark optimization problems. Studies on the effect of parameters in the algorithm are also carried out to obtain the most accurate values so that the efficiency of the algorithm in terms of convergence speed and final accuracy is achieved. [ABSTRACT FROM AUTHOR]

Scheduling is the assignment of different resources with the intention of organising and managing to discover an optimisation in the work process. The emergence of job shop scheduling problems has a significant impact on many industries, particularly manufacturing because these problems have an impact on the sector's productivity and profitability. It is crucial to improve the approach taken to solve the job shop scheduling problem in order to maximise manufacturing productivity and reduce costs. The tabu search algorithm, one of the metaheuristic techniques, has been chosen to improve the disjunctive programming strategy used to solve the job shop scheduling problem. Disjunctive programming is solved using the Lingo software provides the initial solution. Then, by swapping the jobs on critical paths, the neighborhood structure's makespan value is improved. The makespan value has been improved by 24% over the initial solution. [ABSTRACT FROM AUTHOR]

Due to its outstanding performance despite its sensitivity to input parameters, the bat algorithm (BA) is a popularly used metaheuristic algorithm in numerous applications. The algorithm's convergence behavior is affected by the algorithm's frequency, the number of bats, pulse emission rate, and loudness parameters. These parameters must be set to their optimal values to increase the outcome's quality. This research aims to find the optimal number of bats of the binary bat algorithm (BBA). In this situation, the fitness criterion for the fitness function will be a classifier error rate. K-nearest neighbors (KNN) are used as the classifier for error rate calculation and validation methods. This study is conducted using 13 datasets with various dimensional sizes. The analysis results indicate that the number of bats substantially impacts the classification accuracy. Based on the results, the optimal number of bats must be considered when tuning the algorithm's parameters to achieve adequate performance. [ABSTRACT FROM AUTHOR]

Metaheuristics and Hopfield neural networks (HNN) are frequently employed to address complex optimization problems. In particular, the iterative and robust metaheuristics technique such as the Election algorithm (EA) is frequently used to dynamically enhance the training phase and converging of the neural network. In this research, we suggest using a recently developed Hybrid Election Algorithm (HEA) in conjunction with Discrete Hopfield Neural Network (DHNN) to solve problems of varying degrees of complexity in Boolean satisfiability programming. Because of its robust operator set, the HEA can be utilised to ease the computational load of DHNN. The primary objective is to enhance the training phase of DHNN such that it optimally represents higher-order logic in terms of Random k Satisfiability. To win over voters, political parties have advocated for a HEA that would allow them to have greater control over areas outside their borders. This strategy is crucial for accelerating the learning of DHNN. Different numbers of neurons (NN) invalidated the robustness and efficiency of HEA in DHNN for RANkSAT logical expressions. Different statistical error accumulations, global minimum solutions, and similarity analysis were used to compare the proposed model against a preexisting EA model during the training phase. The results indicated that the proposed DHNNRANkSAT-HEA model was superior to the previously used DHNNRANkSAT-EA model. [ABSTRACT FROM AUTHOR]

KM clustering is the most popular technique that finds each cluster's hidden intrinsic pattern in a given real-life application dataset. But it has some limitations, like its run time complexity is very high for high dimensional data sets, and this clustering is very sensitive to its initial centroids. This traditional approach requires the number of clusters, k, in advance for KM, which will not give good quality of clusters, and at the time, this algorithm converges to local optimal solution. So, for solving the problem of local optimal solutions, this paper presented some of the updated NI meta-heuristic algorithms, which has been hybridized with the KM clustering. This hybridized algorithm converges to global optimal solutions rather than local optima in a large search space. Further, we discussed some python API packages from scikit-learn.org for hybrid KM clustering that gives good quality of clusters, and at last, we describe some comparative analyses of KM clustering with integrated KM (KM+NIC). Most of the integrated KM show better accuracy results compared to traditional KM clustering. [ABSTRACT FROM AUTHOR]

Teaching-Learning Based Optimization (TLBO) approach is widely accepted in numerous engineering applications. Despite the similarity of this algorithm to other Meta-heuristic techniques in terms of employing a set of solution and stochastic nature, the inspiration of algorithm is unique. This paper aims at discussing the position of TLBO within the do-main of Nature-Inspired Algorithms (NIA). The effectiveness of this simple and robust paradigm in the fields of engineering and sciences is gaining popularity rapidly due to its less complex approach in tuning with the algorithm and other common parameters like population size, number of iterations and stopping criteria, etc. The paper also highlights the development, mechanism, mathematical formulation and algorithm of TLBO. Furthermore, it is concluded that TLBO is a promising variant in the class of stochastic methods that can be extended to find solutions of challenging real-world problems. The review is expected to be useful for the freshers in research area of Optimization techniques. [ABSTRACT FROM AUTHOR]

Over the past three decades, the travelling salesman problem (TSP) has been the most common discrete optimization issue. In TSP, a salesperson leaves the warehouse to distribute the goods to a number of clients before returning to the point of origin, or the warehouse. A constrained global optimization challenge is the identification of the shortest path connected to n consumers. TSP has changed throughout time from laborious manual computations to cutting-edge computational solutions. By offering the best outcomes along the shortest path, the development of meta-heuristic optimization strategies has been successfully shown in TSP. By analyzing thirty different benchmark issues, the effectiveness of metaheuristic algorithms like Teaching Learning (TLBO) is examined in the current study. The individual problem has been analyzed for swarm size 150 and maximum 20 separate runs totaling 1000 iterations. The findings show that Teaching Learning Based optimization is closer to the best outcomes and to other well-known strategies. Comparing the findings from previous publications also shows how well various algorithms perform in locating the ideal path. When the results are compared to the published literature, it is discovered that they are consistent. Furthermore, using particular benchmark issues, deviations from the optimal path indicated by the various methods are demonstrated. [ABSTRACT FROM AUTHOR]

Iterated Local Search (ILS) is one of the classical metaheuristics that is frequently used for solving the Quadratic Assignment Problem (QAP), a commonly studied NP-hard combinatorial optimization problem. Despite the effectiveness shown by the ILS when solving the QAP, many possible enhancements can still be implemented, based on studies conducted on other combinatorial problems, for example vertex coloring problem, vehicle routing problem, network design optimization, and many others. Among possible approaches for improving the ILS are by implementing adaptive ILS and multi-start strategy. Adaptive ILS is implemented by dynamically changing the perturbation strength of ILS during the run, which focus on developing a robust algorithm that can solve a wide range of instances. Multi-start strategy on the other hand, focus on applying restart or diversification during ILS run to prevent early stagnation. Adaptive ILS with frequent and large perturbation strength change has shown the best overall result when it has obtained the lowest deviation to the best-known solution compared to others. Implementation of Multi-start strategy showed bad performance, except on Taillard75e instance, in which it frequently found the best-known solution compared to others. Both adaptive and multi-start strategy have shown interesting results that may be further studied for ILS improvement. [ABSTRACT FROM AUTHOR]

This study proposes an improved Genetic Algorithm (GA) so-called Adaptive Genetic Algorithm (AGA) and introduces a new variant of Vehicle Routing Problem (VRP) inspired from liquid fertilizer distribution in Indramayu, Indonesia. This case is basically the combination of some VRP variants where the company has a heterogeneous vehicle with asymmetric distance and fuzzy demand (HVRPADFD). Given that VRP is NP-Hard, this study utilizes AGA as a metaheuristics algorithm. AGA has been modified in terms of self-tuning parameter feature and reproduction modification to remove the exhaustive tuning parameter process and enhance the effectivity of the algorithm simultaneously. AGA is then integrated with the fuzzy set theory based on chance-constrained programming to handle the fuzziness. The verification is also done by comparing AGA with original GA and Ant Colony Optimization (ACO) in solving HVRPADFD. The result shows that AGA is statistically better than others in terms of effectivity. AGA also has faster computational time and is more consistent. The sensitivity analysis is also done to understand the risk of fuzzy demand against traveling distance. [ABSTRACT FROM AUTHOR]

The switching angles are chosen in this study with Artificial Bee Colony (ABC), Genetic algorithms (GA), and Gray wolf algorithm (GWO), slime moiled algorithm (SMA), Whale optimization algorithm (WOA). These angles are selected according to the lowest total harmonic distortion of output load voltage from reduced switches multilevel inverter. These algorithms are used in a hybrid method to solve the transcendental nonlinear equation for determining switching angles. A three-phase 27-level inverter fed by isolated unequal PV panels as DC sources with reduced switches and sources is chosen for this study and to reduce THD equal to 0.85%. Maximum power point tracking (MPPT) works with perturb and observe algorithm to match the resistive of PV and load to harvest the maximum power of the PV panel. Theoretical analysis and Simulation are accomplished using Matlab/Simulink for a three-phase 27-level reduced switch multilevel inverter. [ABSTRACT FROM AUTHOR]

The inspiration that has been drawn from the newly discovered natural phenomena, finds substantial application in computer science. Genetic algorithms are a metaheuristic gear for finding solutions of problems that the explorer does not even understand and the results are beneficial to many industrial spheres. However, can that inspiration be applied in a computer game trainer for personal qualities development and this way closing the cycle: from nature to machine and from it to the inner nature of a person? This would require research on the inner occurrences or more precisely the semi-conscious reactions of the player, derived as a response from their insight. Although the fact that the various characters of people are a challenge for the development, maintaining more "individuals" - behaviour of bots generated with genetic algorithms would help in the searching of a useful opponent for the improving player. As much as challenging the goal of "teaching a machine to be a teacher for a person" may seem, it should definitely be set because once accomplished it would bring, not only discovery of the benefits of video games, but would also lead us to the inspiring question: "How much more can a modern software be useful to people and nature?" [ABSTRACT FROM AUTHOR]

There has been a lot of research in recent years into employing optimization approaches to improve artificial intelligence (AI). In this research review paper, we have compared and contrast some of the most usual optimization algorithms, such as Backtracking searching method (BSA), the genetic algorithm (GA), particle swarm optimization (PSO), an artificial bee colony (ABC), and the genetic algorithm, which are all artificial neural networks (ANNs)-based algorithms. The number of recently developed optimization techniques, such as the lightning search algorithm (LSA) and the whale optimization algorithm (WOA) are also compared. All the techniques are categorized based on randomly generated populations. To produce the best possible results, the processing parameters are set within a certain range according to their knowledge. This review paper emphasis on applying optimization techniques to improve the accuracy of simply adjusting the parameters of the neural network. This review paper also presents some results for enhancing neural network performance using various optimization techniques like PSO, GA, and ABC optimization methods to get optimal processing parameters, such as the number of hidden layers, neurons and learning rate etc. The findings of this research review paper will aid in the improvement in the quality of plastic injection molded parts. [ABSTRACT FROM AUTHOR]

The current industrial development is increasing and it has a considerable impact on the consumption of resources to serve as the main and supporting facilities. Garbage management operations generally collect garbage from the source of the garbage generation and then transport it to a landfill. Starting from garbage collection on the streets, garbage collection at public facilities, at each point of the temporary disposal site and then transported by truck or garbage collection vehicle to the final disposal site. The problem of route for transporting garbage is a problem with the Vehicle Routing Problem (VRP), namely the problem of finding a route with minimal costs from one disposal site to another site which is scattered with different quantities of garbage. NN-SOS is believed to be able to make a good contribution to the metaheuristic algorithm By integrating NN and SOS, it is hoped that it will be able to improve the performance of local search which is more effective and the results are stable (steady state) in achieving values that resemble global optimal. It is proven that the total distance generated by the NN-SOS algorithm is 61,34 kilometers, while the total distance traveled in real conditions is 68 kilometers. Then this algorithm is able to provide a solution with a very fast average computation time of 11,9 seconds. [ABSTRACT FROM AUTHOR]

Metaheuristic Algorithms have gained remarkable attention in the past years due to their capability of optimization. These algorithms tend to provide near-optimal solutions to complex computational problems. Nature-inspired metaheuristic algorithms take inspiration of their approach from natural processes. In this paper, we have studied 3 nature-inspired metaheuristic algorithms: Genetic Algorithm, Particle Swarm Optimization and Ant Colony Optimization. Artificial Neural Networks are adaptive models that have the capability to learn from the past mistakes and keep improving on a continuous basis. These mimic the biological neuron network in a human brain. However, there is always a need to optimize these neural networks. Many tasks such as parameter selection, data training etc. are major challenges in ANNs. In this study, we see how numerous researchers have applied metaheuristic techniques to ANNs in order to tackle these problems. [ABSTRACT FROM AUTHOR]

In this paper, two new hybrid algorithms are proposed to solve global optimization problems. The first algorithm combines the sooty tern optimization algorithm with the arithmetic optimization algorithm, and the second algorithm combines the sooty tern optimization algorithm with the conjugate gradient method. The sooty tern optimization algorithm is a metaheuristic algorithm proposed by Dhiman and Kaur in 2019 Sooty migration and attack behaviors seabirds in nature were the primary inspiration for this algorithm. In this paper, this algorithm was hybridized using the arithmetic optimization algori thm and the conjugate gradient method, and the two algorithms were measured by applying them to 10 functions, and the results of the hybrid algorithms were very good compared to the original algorithm. [ABSTRACT FROM AUTHOR]

One of the most important purposes of image processing is finding the information about objects in the image file especially detecting the boundary for an edge. Image edge detection is the critical step in image processing. This paper presents the Ant colony optimization (ACO) as a metaheuristic optimization algorithm that depends on the ant's behavior in for searching food by detecting the edge in the image gray file by proposing a new formula for computing the count of ants by decreasing it with 10%, 30% and 50% of the square root of the multiplication between some rows with columns of the image file. it is performed on different images sizes (128x128, 256x256) with different values of parameters such as the number of ants, the number of iterations then examine the edge detection in the image file by computing the important parameters for solving any problem, especially the time complexity (elapsed time) and mean square error (MSE) with PSNR. When analyzing the results of ACO to evaluate the best value of parameters that influenced the success of detecting the edge points, we deduced that when decreasing the number of iterations, the number of ants gives the best result for edge detection. [ABSTRACT FROM AUTHOR]

Naked mole-rat algorithm (NMRA) is a recently developed swarm intelligence-based algorithm that is inspired by mole rats' (workers and breeders) mating behaviour. This algorithm has been used to solve optimization problems and is based on breeding capability of breeders to breed with the queen. However, the algorithm has issues with inadequate exploration and less speed of convergence to optimal solution. The improved NMRA (INMRA) is proposed in this research to boost its exploration's ability. The new version is based on trigonometric mutation and used for generation of new worker solution This attribute has been introduced to increase NMRA's exploration properties by keeping the search agents diverse. On CEC2005 benchmark problems, the proposed technique INMRA was tested and demonstrate its superiority to other algorithms, including well-known algorithms such as whale optimization based on opposition learning (OEWOA), differential evolution with self-adaptive properties (SaDE), external archive based differential evolution (JADE), extended grey wolf optimization (GWO-E), hybrid sine cosine crow search algorithm and NMRA. Further, performance of proposed version is also better than the original NMRA for CEC 2019 test problems according to the experimental data. The results are further validated by statistical testing and convergence profiles. [ABSTRACT FROM AUTHOR]

For many years, multilevel inverter (MLI) is a system well-known for converting DC voltage to AC voltage which is suitable for energy resources conversion for industrial power sources. With the help of Selective Harmonic Elimination Pulse Width Modulation (SHEPWM) switching technique, the output power from the MLI system is better efficiency with low Total Harmonics Distortion (THD). The SHEPWM switching technique is used to eliminate the lower order of the harmonics and reduces the Total Harmonics Distortion (THD) of the MLI system. Along with the optimization algorithms to solve difficult non-linear equations involving with SHEPWM, applying SHEPWM into MLI system helps to further improve the output power efficiency. Whale Optimization Algorithm (WOA) is one of the algorithms developed which capable of solving the non-linear equation and obtained suitable results for MLI. Additionally, many types of improvement done on WOA, called as Improved Whale Optimization Algorithms (IWOA) also developed to obtain better results compared with WOA. Developed in a MATLAB environment, a proposed IWOA is applied to solve the equation and compared with WOA. The results show that the proposed IWOA capable of achieving higher probability with fast convergence speed reaching global optimal compared to WOA. With the proposed approach, the IWOA efficiently computed required switching angles, to eliminate the selective lower-order harmonics for different modulation indices (Ma). In this paper, the proposed IWOA is performed on a three-phase nine-level cascaded H-bridge multilevel inverter (CHBMLI) for a wide range of modulation indexes between 0.1 until 1. The results show the eliminations of 5th, 7th, and 11th harmonics from the output of the three-phase nine-level MLI system thus reducing THD from the system up to 5.82%. [ABSTRACT FROM AUTHOR]

This study explores a dynamic checkpoint strategy for an on-demand flexible transit service called a Mobility Allowance Shuttle Transit with Dynamic Checkpoint (MAST-DC) that clusters customers by spatial and temporal proximities. By introducing a dynamic checkpoint option to customers in their booking process, we allow customers to walk less distance with a slight increment on the fare price, which is expected to improve the system's level of service and increase the service provider's profit. Hence, we propose an optimal operational framework to cluster customers to dynamic checkpoints by minimizing their walking distance and create the shortest vehicle route between checkpoints without violating the promised scheduled running time at two consecutive fixed checkpoints. This paper proposes a sequential two-phase heuristic method that cluster first and routing second to find a sequence that finds: 1) route that originates and terminates at the fixed checkpoints by minimizing vehicle distance traveled, 2) each cluster centroid visited only once by the same vehicle, and 3) customers' walking distance minimized towards their assigned dynamic checkpoint. The sequential heuristic method clusters customers using the memetic differential evolution (MDE)-based clustering algorithm, one of the state-of-the-art metaheuristic algorithms to solve the minimum sum-of-squares clustering (MSSC) problem. The clustering model is later fused with the second phase, which uses the branch-and-cut method for the traveling salesman problem (TSP) model to find the shortest possible path between identified checkpoints. The parametric simulation experiment is conducted to explore the impact of system parameters on the number of dynamic checkpoints in the analysis region and average walking distance. The results show that the average vehicle travel time uses approximately 90 percent of the scheduled running time. As the demand rate increase, the average number of dynamic checkpoints decrease and customers' average walking distance increases significantly. Therefore, considering the average passengers' walking distance of a fixed transit system, the proposed MAST-DC system is expected to benefit from the demand rate of less than four per square miles and the slack time of 150 percent or 200 percent of the direct travel time. [ABSTRACT FROM AUTHOR]

The Bat algorithm, inspired by microbat foraging behavior, has been utilized to solve optimization issues in continuous and discrete spaces. This algorithm can stagnate after some initial stage. Many methods and tactics have been tried to improve performance by increasing the diversity of the solution and therefore increasing the performance. A variant of the bat algorithm is the binary bat algorithm, in which the new bat position is limited to binary numbers only. The binary bat technique is now widely used to address issues in practically every area of optimization, feature selection, classification, and engineering procedures. Because there have been few reviews of the binary bat algorithm, the proposed study conducted a review of recent applications and modifications. [ABSTRACT FROM AUTHOR]

Now-a-days mobile robot is used to replace man power due to its ability to mimic activities of human beings. In this context, a trajectory exploration analysis with Khepera-III robot is carried out using modified wind driven optimization (WDO) controller. WDO is a metaheuristic optimization method that draws inspiration from nature. The controller is designed to control the robotic agents during steering in cluttered environments. Here, V-REP software is used to check the navigational capability of proposed controller. With simulation, it is found that the robot successfully achieved the goal post in an optimized way with less time consumption without any collision. [ABSTRACT FROM AUTHOR]

A new swarm optimization algorithm known as the sparrow search algorithm (SSA) consider in this work. It fascinated a lot of consideration due to its extremely good attributes. However, it has some pitfalls like minimum global search ability, falling into local optima etc which were found by many researchers. In this paper,first time sparrow search algorithm (SSA) is tested on welded beam design (WBD) optimization problem. Welded beam design (WBD) optimization problem is an engineering optimization problem that has been solved by various metaheuristic algorithms such as SiC-PSO, neuromorphic optimization (NSO) technique etc. Obtained results of SSA were compared with othertechniques such as GWO, PSO, TLBO and it is found that although it works very well on WBD problem with regards to convergence speed, in relation to optimality, convergence accuracy and stability SSA gave the results which were obtained from other algorithms. So, SSA needs to be modified for future work. [ABSTRACT FROM AUTHOR]

The hunting strategy of Dingo (Australian Dog) inspired a newly developed metaheuristic optimization technique called Dingo Optimization algorithm. Using the basic concept of this current approach, a hybrid form known as the Levy dingo Optimization algorithm is created. This new levy function-based technique is devised to boost local search capacity during the exploration phase. This algorithm is effectively validated on nonlinear and constrained multidisciplinary engineering applications and standard benchmark functions. The recommended solution LDOA has been verified and providing optimal solutions as compare to other search algorithms including hHHO-SCA, SCA, HS, GA, GSA etc. and there is a neck-to-neck competition with some of the techniques like GWO, PSO etc. This method is put to the test on four engineering problems: speed reducers, welded beams, pressure vessels, and tension/compression spring design. The resulting results are compared to various well-known approaches. [ABSTRACT FROM AUTHOR]

This investigation focuses on the photovoltaic (PV) system's maximum power point tracking (MPPT). The main downsides of conventional MPPT are its slow response due to the fixed-step change, excessive ripple, and increased oscillation caused by swings near the maximum power point. Hybrid MPPT control is proposed to address these concerns employing an unique metaheuristic method known as the Archimedes optimization algorithm(AOA). The derivatives controller Grey Wolf Optimizer (GWO) has very little overshoot or oscillation in the proposed MPPT compared to regular MPPT and other ingenious tactics described in researcher articles. The proposed system parameters are examined by simulation using MATLAB/Simulink According to the simulation results, the proposed MPPT has a maximum efficiency of 99.62%. [ABSTRACT FROM AUTHOR]

In today's world, the notion of virtualization is the essential technology employed in cloud computing. This virtualization technology may run many operating systems with complete resources on a single piece of hardware. Furthermore, the virtualization innovation has restricted security capabilities for verifying total territorial cloud circumstances. This paper presents another security examination way for distinguishing the attack in the virtualized framework of distributed computing. The paper focuses on extracting accurate data from the organization qualities to choose the presence of assault in the virtualized framework. In this perspective, the projected work has 2 significant stages Feature extraction (I) and Classification (II). In lieu of the fact that the proposed technique is a "Bigdata" based problem on qualities like volume, velocity, variety, value, and veracity, it is wanted to utilize Python's multi-process framework to isolate the add up to Information proper and facilitate the process. The feature extraction happens for each isolated data, and features are extracted from Mutual information-based features. Finally, the characteristics obtained from the multi-functions, are passed through the classification process to detect an attack on the cloud. The presence of an attack in the network is detected using optimized DBN. To improve the detection accuracy, the Deep Belief Network(DBN) weights will be optimally tuned using a whale optimization algorithm[2] that's helps in DBN learning. Then the proposed model performance is compared with the traditional method like SVM, naïve Bayes, k nearest neighbors on various parameters like accuracy, precision, sensitivity recall and f1-score. [ABSTRACT FROM AUTHOR]

In the optimum design of reinforced concrete (RC) members, the design requirements and composite behavior of concrete and steel make the problem a nonlinear one that can be only solved via numerical iterations. In that case, the best option for these problems is to use a metaheuristic algorithm. In this study, a design variable that is the height of the RC members is investigated for the different maximum value cases, because the optimum design tends to maximize this value to increase the moment capacity of the member by directly increasing the length of the moment arm distance of the force acting from steel rebar having more cost than concrete. [ABSTRACT FROM AUTHOR]

Metaheuristic optimization algorithms are increasingly applied to the solution of engineering problems. The current study focuses on the minimization of the CO2 emission associated with the production of a plate girder. Considering the tendency to lateral-torsional buckling in these structures, the critical buckling moment has been used as one of the optimization constraints. Furthermore, the flanges and the web of the plate girder are kept within the non-compact range throughout the optimization process. As the method of optimization, the metaheuristic Jaya algorithm has been modified using Lévy probability distributions for each design variable. The flange width and thickness, the web thickness and the clear distance between the flanges are chosen to be the variables of optimization. [ABSTRACT FROM AUTHOR]

Distribution is one of the activities included in the main component of the downstream supply chain segment in supply chain management. An effective and efficient distribution process is one of the factors whose position is starting to align with other indicators in its efforts to achieve customer satisfaction. The higher the level of competition in the industrial world will require companies to make better distribution strategies. On the other hand, distribution can generate the highest costs. Therefore, it is essential to keep distribution costs to a minimum by optimizing vehicle routes for product distribution. Determining the optimal route for distribution is known as the Vehicle Routing Problem (VRP), a distribution problem to find a series of routes for several vehicles with a specific capacity from one or more depots to serve customers. There are several methods to solve CVRP problems, among others, using two approaches, namely the exact approach and the heuristic approach. This study applies several approaches, namely, using a heuristic and an exact approach to obtain the most optimal route implemented in the distribution of three kilograms of LPG cylinders. Based on the data processing results in solving this CVRP problem, the total vehicle mileage obtained is 161 km (using a heuristic approach) and 159.6 km (using an exact approach). In comparison, the whole vehicle mileage of this company is 200.9 km. Thus, the heuristic approach can provide mileage savings with a percentage of 19.86% and an exact approach of 20.56%. [ABSTRACT FROM AUTHOR]

The current paper introduces a simple hybrid approach to predict the photovoltaic modules' behaviors based on the single-diode model. The proposed technic consists of calculating the five electrical parameters (Iph, Is, Rsh, a, and Rs) of the PV generator's equivalent circuit. The extraction is based only on the values of the three typical points of the current-voltage (I-V) curve; the short-circuit point (0, Isc), the maximum power point (Vmp, Imp), and the open-circuit point (Voc, 0). The method in this work avoids the problem facing the meta-heuristic optimization algorithms, which is the requirement of the integral experimental I-V curve's availability. The proposed technic follows two extraction steps. First, the three parameters (Iph, Is, and Rsh) are calculated analytically, based on a matrix denotation used for simplification and to avoid tedious calculations. Second, the fourth and fifth parameters (a and Rs) are determined using a fast numerical technic. The introduced method's performance is proved using the measured data accessible on the datasheet of the thin-film PV module ST40. The accuracy is evaluated against some of the most reviewed hybrid methods of the literature, where the current method provides the best results by offering the least root mean square error. Hence, it ensures the most accurate estimation, which is more than four times better than the best performance given by the selected methods of the literature. [ABSTRACT FROM AUTHOR]