Hybrid Metaheuristic Research Articles

Optimized Anfis Model with Hybrid Metaheuristic Algorithms for Facial Emotion Recognition

Optimized Anfis Model with Hybrid Metaheuristic Algorithms for Facial Emotion Recognition

A novel hybrid meta-heuristic contrast stretching technique for improved skin lesion segmentation

The high precedence of epidemiological examination of skin lesions necessitated the well-performing efficient classification and segmentation models. In the past two decades, various algorithms, especially machine/deep learning-based methods, replicated the classical visual examination to accomplish the above-mentioned tasks. These automated streams of models demand evident lesions with less background and noise affecting the region of interest. However, even after the proposal of these advanced techniques, there are gaps in achieving the efficacy of matter. Recently, many preprocessors proposed to enhance the contrast of lesions, which further aided the skin lesion segmentation and classification tasks. Metaheuristics are the methods used to support the search space optimisation problems. We propose a novel Hybrid Metaheuristic Differential Evolution-Bat Algorithm (DE-BA), which estimates parameters used in the brightness preserving contrast stretching transformation function. For extensive experimentation we tested our proposed algorithm on various publicly available databases like ISIC 2016, 2017, 2018 and PH2, and validated the proposed model with some state-of-the-art already existing segmentation models. The tabular and visual comparison of the results concluded that DE-BA as a preprocessor positively enhances the segmentation results.

Combining hybrid genetic search with ruin-and-recreate for solving the capacitated vehicle routing problem

The Capacitated Vehicle Routing Problem (CVRP) has been subject to intense research efforts for more than sixty years. Yet, significant algorithmic improvements are still being made. The most competitive heuristic solution algorithms of today utilize, and often combine, strategies and elements from evolutionary algorithms, local search, and ruin-and-recreate based large neighborhood search. In this paper we propose a new hybrid metaheuristic for the CVRP, where the education phase of the hybrid genetic search (HGS) algorithm proposed by (Vidal Hybrid Genetic Search for the CVRP: Open-Source Implementation and SWAP* Neighborhood 2020) is extended by applying large neighborhood search (LNS). By performing a series of computational experiments, we attempt to answer the following research questions: 1) Is it possible to gain performance by adding LNS as a component in the education phase of HGS? 2) How does the addition of LNS change the relative importance of the local search neighborhoods of HGS? 3) What is the effect of devoting computational efforts to the creation of an elite solution in the initial population of HGS? Through a set of computational experiments we answer these research questions, while at the same time obtaining a good configuration of global parameter settings for the proposed heuristic. Testing the heuristic on benchmark instances from the literature with limited computing time, it outperforms existing algorithms, both in terms of the final gap and the primal integral.

Uma nova abordagem para o problema de roteamento de veículos green bi-objetivo: otimização na distribuição de jornais

O objetivo deste trabalho é apresentar uma metodologia para fornecer uma solução para o Problema de Roteamento de Veículos Green Bi-objetivo (Bi-objective Green Vehicle Routing Problem, BGVRP). A metodologia, ilustrada por meio de um estudo de caso (problema de distribuição de jornais) e instâncias da literatura, foi dividida em três etapas: Etapa 1, tratamento dos dados; Etapa 2, abordagens meta-heurísticas (híbridas ou não híbridas), utilizadas comparativamente; Etapa 3, análise dos resultados, com comparação dos algoritmos. Uma otimização de 19,9% foi alcançada para a Função Objetivo 1 (FO1; minimização das emissões de CO2) e, consequentemente, o mesmo percentual para a minimização da distância total, e 87,5% para a Função Objetivo 2 (FO2; minimização da diferença na demanda). Abordagens meta-heurísticas híbridas alcançaram resultados superiores para o estudo de caso e as instâncias. Desta forma, o procedimento aqui apresentado poderá trazer benefícios para a sociedade já que considera questões ambientais, além do balanceamento do trabalho entre os roteiros, garantindo economia e satisfação para possíveis usuários.

A Hybrid Metaheuristic and Deep Learning Approach for Change Detection in Remote Sensing Data

This study aimed to adapt Convolutional Neural Networks (CNN) to solve the problem of change detection using remote sensing imagery. Specifically, the goal was to investigate the impact of each CNN layer to detect changes between two satellite images acquired on two different dates. As low-level CNN layers detect fine details (small changes) and higher-level layers detect coarse details (large changes), the idea was to assign a weight to each layer and use a genetic algorithm based on a training dataset to generalize the detection process on the test dataset. The results showed the effectiveness of the proposed approach based on two real-life datasets.

Machine learning of tax avoidance detection based on hybrid metaheuristics algorithms

This paper addresses the performances of machine learning classification models for the detection of tax avoidance problems. The machine learning models employed automated features selection with hybrid two metaheuristics algorithms namely particle swarm optimization (PSO) and genetic algorithm (GA). Dealing with a real dataset on the tax avoidance cases among companies in Malaysia, has created a stumbling block for the conventional machine learning models to achieve higher accuracy in the detection process as the associations among all of the features in the datasets are extremely low. This paper presents a hybrid meta-heuristic between PSO and adaptive GA operators for the optimization of features selection in the machine learning models. The hybrid PSO-GA has been designed to employ three adaptive GA operators hence three groups of features selection will be generated. The three groups of features selection were used in random forest (RF), k-nearest neighbor (k-NN), and support vector machine (SVM). The results showed that most models that used PSO-GA hybrids have achieved better accuracy than the conventional approach (using all features from the dataset). The most accurate machine learning model was SVM, which used a PSO-GA hybrid with adaptive GA mutation.

Boosting the training of neural networks through hybrid metaheuristics

Boosting the training of neural networks through hybrid metaheuristics

An improved hybrid metaheuristics and rule-based approach for flexible job-shop scheduling subject to machine breakdowns

This article proposes an improved hybrid heuristic algorithm to handle the flexible job-shop scheduling problem subject to machine breakdowns. An initial prescheduling sequence is generated by a genetic algorithm. In the rescheduling process, the generated sequence with an uncertain repair time is modified by three typical adjustments of the route-change and the right-shift: (1) the route-change and the right-shift are combined to gain the best improvement; (2) one operation can be inserted after the last operation on any machine that has no impact on the related operations to improve the final solution; and (3) the lowest cost modification option is chosen to generate a new schedule to adapt to breakdown. The objective of these adaptive methods is to reduce the makespan and idle time of each machine when a random failure happens. Experimental results are evaluated to confirm the advantages of the developed algorithm.

Pallet location and job scheduling in a Twin-Robot system

This paper introduces the Twin-Robot Pallet Assignment and Scheduling Problem (TRPASP) in which two robots operating on a rail must be scheduled to pick up and deliver a set of products. The objective is to minimise the makespan, defined as the time taken for the robots to transfer all products from their pickup locations to their delivery locations, and return to their starting positions. Pickup locations are known a priori, while delivery locations must be assigned to minimise the makespan. The robots must respect a safety distance to avoid collisions. The paper presents a mathematical model for the TRPASP before introducing four heuristic algorithms for solving the problem. Computational experiments demonstrate that the best results are returned by a parallel hybrid metaheuristic.

PSO-GAScript: A Domain-specific Scripting Language for Meta-heuristics Algorithm

PSO-GAScript is a domain-specific scripting language designed to support easy and rapid implementation of meta-heuristics algorithms focused on Particle Swarm Optimization (PSO) and Genetic Algorithm (GA). The programming language has been developed to allow the hybridization of the two meta-heuristics algorithms. Hybridizations between PSO and GA are proven to be a comprehensive tool for solving different kinds of optimization problems. Moreover, the two algorithms have achieved a remarkable improvement from the adaptation of dynamic parameterization. Nevertheless, implementing the suitable hybrid algorithms is a considerably difficult, which in most cases is time consuming. To the best of our knowledge, the existing tools are not adequately designed to enable users to easily develop the meta-heuristics hybridization of PSO-GA with dynamic parameterizations. This paper presents the fundamental research methodology of domain-specific scripting language from the language design, constructs and evaluations focused on the case of PSO-GA hybridization and dynamic parameterizations. The PSO-GAScript are shown to easily use with minimal number of codes lines and concisely describe the meta-heuristics algorithms in a directly publishable form.

Towards Intelligent Road Traffic Management over Weighted Large Graphs Hybrid Meta-heuristic-Based Approach

This paper introduces a new approach of hybrid meta-heuristics based optimization technique for decreasing the computation time of the shortest paths algorithm. The problem of finding the shortest paths is a combinatorial optimization problem which has been well studied from various fields. The number of vehicles on the road has increased incredibly. Therefore, traffic management has become a major problem. We study the traffic network in large scale routing problems as a field of application. The meta-heuristic we propose introduces new hybrid genetic algorithm named IOGA. The problem consists of finding the k optimal paths that minimizes a metric such as distance, time, etc. Testing was performed using an exact algorithm and meta-heuristic algorithm on random generated network instances. Experimental analyses demonstrate the efficiency of our proposed approach in terms of runtime and quality of the result. Empirical results obtained show that the proposed algorithm outperforms some of the existing technique in term of the optimal solution in every generation.

Variable neighborhood-based Cuckoo Search for production routing with time window and setup times

Major corporations compete over the strengths of their supply chains. Integrating production and distribution operations helps improve supply chain connectedness and responsiveness beyond the standalone optimization norms. This study proposes an original Mixed-Integer Linear Programming (MILP) formulation for the Production scheduling-based Routing Problem with Time Window and Setup Times (PRP-TWST). For this purpose, the identical parallel machine scheduling is integrated with the vehicle routing problem. Considering the highly intractable solution spaces of the integrated problem, hybrid metaheuristics based on the Variable Neighborhood Search (VNS), Particle Swarm Optimization (PSO), and Cuckoo Search (CS) algorithms are developed to solve the PRP-TWST problem. Extensive numerical experiments are conducted to evaluate the effectiveness of the developed algorithms considering the total delay time as the objective function. The results are supportive of the VNS-based CS algorithm’s effectiveness; the developed metaheuristics can be considered strong benchmarks for further developments in the field. This study is concluded by suggesting directions for modeling and managing integrated operations in the supply chain context.

A cost-effective solution to security constrained dispatch problem of Islanded microgrid using chaotic spotted Hyena optimization

ABSTRACT Power system network is growing day by day as per the increased demand due to technological development around the globe. The increased power demand needs optimized planning for all the sections of power system. Economic dispatch is one of the important areas in which new optimization techniques can add to the overall improvement in the power system planning. Recent technological improvement in microgrids results in economic and efficient operation considering both conventional and non-conventional methods of generation. This paper highlights the optimal scheduling requirement in the advanced power system control by introducing a new proposed metaheuristic hybrid approach, i.e., Chaotic Spotted Hyena Optimization (CSHO). The proposed approach improves and intensifies the conventional approach and the results obtained are compared with some well-known published approaches using standard data. The CSHO is implemented on an islanded microgrid having five distributed generators out of which three are conventional generators and two are renewable sources. Three objectives are tested on the microgrid-(1) Economic scheduling, (2) Environmental emission and (3) CEED for Microgrid and the results of proposed approach are tested in four scenarios and compared with the PSO, DE, SOS, GWO and WOA approaches. The result achieved from the proposed approach is remarkable over previous approaches and showing enhanced economic operation, which are verified and presented in different sections of paper.

The Optimal Integration of Multiple DGs under Different Load Models using Artificial Bee Colony-Hill Climbing Algorithm

With the continual advances in technologies, there is a significantly increased demand for electricity leads to a drastic increase in usage and penetration of DGs into the distribution network. The incorrect addition of DGs into the distribution system resulting in declining stability, power losses, voltage profile, reliability, and power quality of the system. Therefore, the optimal planning of DGs (OPDG) has always been under constant investigation and development for the maximum beneficiary. The proposed framework presents a novel meta-heuristic hybrid optimization technique called Artificial Bee Colony-Hill Climbing (ABC-HC) to find the best siting and sizing of multiple DGs using a weighted multi-objective function which includes real and reactive power loss index and voltage deviation index. The results clearly indicate the significant reduction in losses and voltage deviation achieved by integration of single and multiple DGs. The research work is also extended to assess the voltage harmonics of the system before and after the placement of DGs. The effectiveness of the proposed ABC-HC algorithm has been demonstrated on the standard IEEE-33 bus RDS under different voltage dependent load models using MATLAB 2017a.
 HIGHLIGHTS
 
 As demand grows, technologies advance, and power system infrastructure changes, there are more opportunities to integrate distributed generators into the distribution system
 The maximum benefit from the DG integration is achieved by selecting the optimal size and optimal site in the distribution system
 A new hybrid optimization technique has been developed for optimal planning of DGs under different voltage dependent loads. The work was also extended to carry out harmonic analysis
 
 GRAPHICAL ABSTRACT

APPLICATION OF HYBRID SFLA AND ACO ALGORITHM TO OMEGA PLATE FOR DRILLING PROCESS PLANNING AND COST MANAGEMENT

<p>Tool traveling and tool switching time constitute almost seventy percent (70%) of the total time<br />consumed in drilling process. This fact that 70% of the total time is nonproductive and does not add any value to the job, grabs attention of the researchers and the industrialist for optimization. A literature on drilling process revealed that very few studies have been done on hybridization of metaheuristics for optimization of tool travel time. This research gap is the motivation of the present study. In this study, two metaheuristic approaches – the shuffled frog leaping algorithm (SFLA) and ant colony optimization (ACO) were hybridized. With respect to hybridization of SFLA and ACO, this study signifies its originality and novelty in which main objective is to minimize the tool travel time. The literature review also revealed that the shortest path generated through commercially available software is not optimal all the time. This aspect emphasizes the application of metaheuristic algorithms on the real-world industrial problems. In this study, the proposed hybrid algorithm was applied to drilling of omega plate which is used in automobile manufacturing industry. The results of the proposed hybrid algorithm were compared with those of manual drilling path and software generated path. The results obtained through proposed hybrid algorithm were improved by 11.1% when compared to results of manual drilling path. The results of proposed algorithm were also better than results of commercial software Creo 6.0 and Siemens NX by 5.9% each. This showed that hybrid algorithm outperformed the commercially available software. This not only validates the efficacy of proposed hybrid algorithm, but also indicates the significance of the metaheuristic algorithm applications in industrial optimization problems.</p>

Hybrid metaheuristics for the integrated and detailed scheduling of production and delivery operations in no-wait flow shop systems

In today's complex business environment, the optimisation of operational processes is an essential factor for the survival of organisations. Many companies struggling to maintain competitiveness search for better operational methods and models to minimise costs and improve market service levels. In these companies, the coordination between manufacturing and logistics activities is critical. The integration and synchronisation of these activities are alternatives that companies require to improve their operational performance and, consequently, increase customer service levels. This article contributes to this approach by proposing and evaluating new heuristic methods to solve the problem of scheduling production and distribution operations in no-wait flow shop systems integrated with transportation systems which operate with different distribution strategies. Based on the assumption that the improvement in synchronisation and coordination of production and transportation operations minimises delivery times and increases service levels, we propose and evaluate heuristic methods for the problem. These heuristic methods implement scheduling algorithmic structures based on the iterated greedy (IG), iterated local search, simulated annealing (SA), and variable neighbourhood search metaheuristics, which jointly schedule manufacturing and transportation operations. In this context, the decision level is operational. The minimum total flow time is the main optimisation criterion. The problem studied herein is inspired on the real-life case of an electronic goods company. The analysis of the experimental results demonstrates the potential of these methods to solve the problem of different models of transportation. Specifically, the IGSA2 method obtained the best results, improving the quality of the initial solutions by up to 17% in small, medium, and large instances depending on the adopted initialisation heuristic. The best proposed algorithm IGSA2 was compared with the two state-of-the-art algorithms to obtain superior results. The study also verified a proposed hypothesis to improve the quality of the solutions generated using heuristic methods, which was proved to be true for most of the experiments.

Botanical Leaf Disease Detection and Classification Using Convolutional Neural Network: A Hybrid Metaheuristic Enabled Approach

Botanical plants suffer from several types of diseases that must be identified early to improve the production of fruits and vegetables. Mango fruit is one of the most popular and desirable fruits worldwide due to its taste and richness in vitamins. However, plant diseases also affect these plants’ production and quality. This study proposes a convolutional neural network (CNN)-based metaheuristic approach for disease diagnosis and detection. The proposed approach involves preprocessing, image segmentation, feature extraction, and disease classification. First, the image of mango leaves is enhanced using histogram equalization and contrast enhancement. Then, a geometric mean-based neutrosophic with a fuzzy c-means method is used for segmentation. Next, the essential features are retrieved from the segmented images, including the Upgraded Local Binary Pattern (ULBP), color, and pixel features. Finally, these features are given into the disease detection phase, which is modeled using a Convolutional Neural Network (CNN) (deep learning model). Furthermore, to enhance the classification accuracy of CNN, the weights are fine-tuned using a new hybrid optimization model referred to as Cat Swarm Updated Black Widow Model (CSUBW). The new hybrid optimization model is developed by hybridizing the standard Cat Swarm Optimization Algorithm (CSO) and Black Widow Optimization Algorithm (BWO). Finally, a performance evaluation is undergone to validate the efficiency of the projected model.

A bi-level model for a closed-loop agricultural supply chain considering biogas and compost

In today’s competitive marketplace, to increase customer satisfaction and profitability, supply chain management has become more prominent. Therefore, thorough planning and designing the supply chain by seeing all levels and units are essential to growing the efficiency of the entire supply chain. In the present study, an eight-echelon network is designed for a closed-loop agricultural supply chain. These eight echelons are consisting of suppliers, farms, distribution centers (DCs), customers, recycling depots, biogas centers, compost production centers, and biogas applicants. To design the agricultural logistics network, a bi-level programming mathematical model is presented. The first objective seeks to minimize total costs of the upper-level which consist of shipping costs, construction costs, production costs, inventory holding costs, and buying costs. Besides, the second objective attempts to maximize total profits of the lower-level using subtraction of incomes from the costs which the total income is calculated by selling manufactured biogas and compost to its applicants. Since the bi-level programming problems are part of the NP-hard class and due to the computational complexity of the problems, the meta-heuristic algorithms are utilized to solve the formulated problem. To this end, two meta-heuristics consisted of Genetic Algorithm (GA) and Stochastic Fractal Search (SFS) are employed. Moreover, two hybrid metaheuristics created from these algorithms which include GA-SFS and SFS-GA are suggested to search for more appropriate solutions. Finally, various comparisons and analyses are performed to evaluate the model’s performance and the capabilities of the solution methods and the results showed the superiority of SFS-GA over other methods. Also, the results imply that considering biogas and compost can not only prevent environmental pollution, but also lead to profitability and the production of new products. Therefore, using this plan in countries that have agricultural products and also do not produce fossil fuels can be more attractive and practical.

A Hybrid Feature Selection Framework Using Improved Sine Cosine Algorithm with Metaheuristic Techniques

Feature selection is the procedure of extracting the optimal subset of features from an elementary feature set, to reduce the dimensionality of the data. It is an important part of improving the classification accuracy of classification algorithms for big data. Hybrid metaheuristics is one of the most popular methods for dealing with optimization issues. This article proposes a novel feature selection technique called MetaSCA, derived from the standard sine cosine algorithm (SCA). Founded on the SCA, the golden sine section coefficient is added, to diminish the search area for feature selection. In addition, a multi-level adjustment factor strategy is adopted to obtain an equilibrium between exploration and exploitation. The performance of MetaSCA was assessed using the following evaluation indicators: average fitness, worst fitness, optimal fitness, classification accuracy, average proportion of optimal feature subsets, feature selection time, and standard deviation. The performance was measured on the UCI data set and then compared with three algorithms: the sine cosine algorithm (SCA), particle swarm optimization (PSO), and whale optimization algorithm (WOA). It was demonstrated by the simulation data results that the MetaSCA technique had the best accuracy and optimal feature subset in feature selection on the UCI data sets, in most of the cases.

Local search based methods for scheduling in the unrelated parallel machines environment

In many real-world situations it is necessary to make timely scheduling decisions. In most cases, metaheuristic algorithms are used to solve various scheduling problems because of their flexibility and their ability to produce satisfactory results in a short time. In recent years, several novel or hybrid metaheuristics have been proposed for scheduling problems. Although such research leads to new insights, it inevitably causes certain problems. First, it becomes unclear which methods perform best, especially if they are not properly compared with existing ones. Second, the proposed methods become increasingly complex, making them more difficult to understand and apply. The goal of this study is to investigate the possibility of defining efficient but simple iterative local search (ILS) methods for the parallel unrelated machines environment with minimisation of the total weighted tardiness. To improve the efficiency of ILS methods, several design decisions, such as the construction of the initial solution and choice of local search operators. The proposed methods have been compared with several metaheuristics, of which they achieve significantly better results. Thus, we conclude that it is not necessary to increase the complexity of metaheuristics to achieve better results. Rather, better results can be obtained with simple but well-designed local search methods.