Classical Genetic Operators Research Articles

Interval multi-objective grey wolf optimization algorithm based on fuzzy system

PurposeThe interval multi-objective optimization problems (IMOPs) are universal and vital uncertain optimization problems. In this study, an interval multi-objective grey wolf optimization algorithm (GWO) based on fuzzy system is proposed to solve IMOPs effectively.Design/methodology/approachFirst, the classical genetic operators are embedded into the interval multi-objective GWO as local search strategies, which effectively balanced the global search ability and local development ability. Second, by constructing a fuzzy system, an effective local search activation mechanism is proposed to save computing resources as much as possible while ensuring the performance of the algorithm. The fuzzy system takes hypervolume, imprecision and number of iterations as inputs and outputs the activation index, local population size and maximum number of iterations. Then, the fuzzy inference rules are defined. It uses the activation index to determine whether to activate the local search process and sets the population size and the maximum number of iterations in the process.FindingsThe experimental results show that the proposed algorithm achieves optimal hypervolume results on 9 of the 10 benchmark test problems. The imprecision achieved on 8 test problems is significantly better than other algorithms. This means that the proposed algorithm has better performance than the commonly used interval multi-objective evolutionary algorithms. Moreover, through experiments show that the local search activation mechanism based on fuzzy system proposed in this study can effectively ensure that the local search is activated reasonably in the whole algorithm process, and reasonably allocate computing resources by adaptively setting the population size and maximum number of iterations in the local search process.Originality/valueThis study proposes an Interval multi-objective GWO, which could effectively balance the global search ability and local development ability. Then an effective local search activation mechanism is developed by using fuzzy inference system. It closely combines global optimization with local search, which improves the performance of the algorithm and saves computing resources.

Performance Analysis of Multi-Objective Simulated Annealing Based on Decomposition

Simulated annealing is a metaheuristic that balances exploration and exploitation to solve global optimization problems. However, to deal with multi- and many-objective optimization problems, this balance needs to be improved due to diverse factors such as the number of objectives. To deal with this issue, this work proposes MOSA/D, a hybrid framework for multi-objective simulated annealing based on decomposition and evolutionary perturbation functions. According to the literature, the decomposition strategy allows diversity in a population while evolutionary perturbations add convergence toward the Pareto front; however, a question should be asked: What is the effect of such components when included as part of a multi-objective simulated annealing design? Hence, this work studies the performance of the MOSA/D framework considering in its implementation two widely used perturbation operators: classical genetic operators and differential evolution. The proposed algorithms are MOSA/D-CGO, based on classical genetic operators, and MOSA/D-DE, based on differential evolution operators. The main contribution of this work is the performance analysis of MOSA/D using both perturbation operators and identifying the one most suitable for the framework. The approaches were tested using DTLZ on two and three objectives and CEC2009 benchmarks on two, three, five, and ten objectives; the performance analysis considered diversity and convergence measured through the hypervolume (HV) and inverted generational distance (IGD) indicators. The results pointed out that there is a promising improvement in performance in favor of MOSA/D-DE.

A multi-objective quantum-inspired genetic algorithm for workflow healthcare application scheduling with hard and soft deadline constraints in hybrid clouds

Recently, the use of quantum cloud computing for different applications has been increasing. For instance, weather forecasting, financial modeling, healthcare, and automation are geographically distributed in practice. These applications are workflows and consist of compute-intensive dependent tasks with precedence constraints. However, workflow processing on quantum-based cloud services still faces issues in the literature regarding makespan and energy consumption. This study presents the Multi-objective Quantum-inspired Genetic Algorithm (MQGA) to address the problems of workflow scheduling in the hybrid cloud, attempting to reduce makespan and energy consumption simultaneously. The proposed algorithm relies on the concept and principle of quantum mechanics, which explores the computational power of quantum computing. It adopted a qubit to represent the individual chromosome for better population diversity. It also uses a quantum rotation gate to lead the schedule to better convergence and avoids classical genetic operators. The simulation results show that the proposed algorithm can effectively reduce energy consumption by 23.36% and makespan 20% on average.

Usage of swarm intelligence strategies during projection of parallel neuroevolution methods for neuromodel synthesis

The paper proposes the ways to apply swarm intelligence strategies to parallelize neuroevolution methods for synthesizing artificial neural networks. The proposed approaches will solve a number of problems that usually arise during designing high-performance computing related to the synthesis of neural networks. The object of research is the process of developing a parallel approach for the neuroevolution synthesis of artificial neural networks, namely, the use of swarm intelligence strategies to solve a number of problems in designing a method that would use the resources of a parallel computer system.One of the most problematic areas is the highly adaptive nature and significant operating time of neuroevolution methods. One way to solve these problems is to use parallel computer systems and distributed computing. However, a number of questions arise when designing a parallel neuroevolution method.During research a number of tasks were solved, which included the analysis and study of neuroevolution methods for synthesizing artificial neural networks and problems of their parallelization. Attention is also paid to swarm intelligence methods, which have gained popularity recently and show good results.The new method developed during the work was based on strategies for organizing work with swarm particles. Thus, sub-populations distributed between threads and individuals were analyzed as individual particles that interact with each other and depend on the local environment. Classical genetic operators were modified by criterion mechanisms to improve adaptability.During the experiments, the developed method was compared with classical methods. During the work, special attention was paid not only to the characteristics of the resulting neuromodels, but also to the load on the processor during Operation. The developed method showed acceptable results for all comparisons. The new approach has significantly improved the quality level of the parallel neuroevolution synthesis method, allowing to evenly use the capabilities of computing nodes in a parallel system

Revealing dynamic communities in networks using genetic algorithm with merge and split operators

Community structures are pervasive in real-world networks, portraying the strong local clustering of nodes. Unveiling the community structure of a network is deemed to be a crucial step towards understanding its dynamics. Actually, most real-world networks are dynamic, and their community structures are evolving over time accordingly. How to reveal these dynamic communities has recently become a pressing issue. This paper presents an evolutionary method termed MSGA for accurately identifying dynamic communities in networks. First, we propose temporal asymptotic surprise (TAS), an effective measure to evaluate the quality of a partition on the snapshot of the dynamic network. Then we develop ad-hoc merge and split operators to perform an information-directed large-scale search at a low cost. Finally, large-scale search, coupled with classic genetic operators, are used to reveal a better solution for each snapshot of the network. MSGA does not require specifying the proposed number of communities. It can break the resolution limit and satisfies temporal smoothness constraints. Experimental results show that MSGA outperforms other state-of-the-art approaches on both synthetic networks and real-world networks.

An enhanced Genetic Algorithm with an innovative encoding strategy for flexible job-shop scheduling with operation and processing flexibility

This paper considers the Flexible Job-shop Scheduling Problem with Operation and Processing flexibility (FJSP-OP) with the objective of minimizing the makespan. A Genetic Algorithm based approach is presented to solve the FJSP-OP. For the performance improvement, a new and concise Four-Tuple Scheme (FTS) is proposed for modeling a job with operation and processing flexibility. Then, with the FTS, an enhanced Genetic Algorithm employing a more efficient encoding strategy is developed. The use of this encoding strategy ensures that the classic genetic operators can be adopted to the utmost extent without generating infeasible offspring. Experiments have validated the proposed approach, and the results have shown the effectiveness and high performance of the proposed approach.

��� ����'�������� �����I ��������� �������� ������, �������������� ���������� ��������

The timetable problem of classes is an urgent problem in higher education institutions. There are a number of classical methods for solving it: dynamic programming methods, integer programming, nonlinear programming, the branch and bound method, simulation methods, the graph coloring method, assignment problem, and others. The peculiarity of these methods is the mathematical rigor of both the formulation of the timetable problem of classes and algorithms for its solution. They have predetermined convergence time and accuracy of the solution and allow to estimate the inux of dierent factors for time of nding the solution. The disadvantage of all classic methods is that they basically use an iterative search procedure or renement of some initial approximation, whereby the result is searched around that approximation. This means that the result directly depends on some initial approximation and naturally there is a problem of its choice, which leads to the need for a multiple experiment with dierent values of the initial approximation, which signicantly increases the time to nd the nal solution. Also, the classical methods are characterized by the complexity of the mathematical model obtained and the sharp (exponential) increase in time spent nding an acceptable solution as the volume of source information increases. To avoid the above disadvantages of the classical methods, the timetable problem of classes can be solved by applying a genetic algorithm. The paper proposes one variant of setting a objective function for timetable optimization and dening a tness function based on it. The article describes the implementation of classical genetic operators: crossover, mutations, and selection for a population of timetables, and also proposes a correction operator that improves the variants of timetables obtained by calls of classical genetic operators. For the implementation of crossover and correction operators, the concept of a local target function is introduced. The general scheme of the genetic algorithm for solving the timetable problem is presented in the paper, and the results of the numerical experiment are given.

Weighted Hierarchical Grammatical Evolution.

Grammatical evolution (GE) is one of the most widespread techniques in evolutionary computation. Genotypes in GE are bit strings while phenotypes are strings, of a language defined by a user-provided context-free grammar. In this paper, we propose a novel procedure for mapping genotypes to phenotypes that we call weighted hierarchical GE (WHGE). WHGE imposes a form of hierarchy on the genotype and encodes grammar symbols with a varying number of bits based on the relative expressive power of those symbols. WHGE does not impose any constraint on the overall GE framework, in particular, WHGE may handle recursive grammars, uses the classical genetic operators, and does not need to define any bound in advance on the size of phenotypes. We assessed experimentally our proposal in depth on a set of challenging and carefully selected benchmarks, comparing the results of the standard GE framework as well as two of the most significant enhancements proposed in the literature: 1) position-independent GE and 2) structured GE. Our results show that WHGE delivers very good results in terms of fitness as well as in terms of the properties of the genotype-phenotype mapping procedure.

Multi-objective sizing and topology optimization of truss structures using genetic programming based on a new adaptive mutant operator

Most real-world engineering problems deal with multiple conflicting objectives simultaneously. In order to address this issue in truss optimization, this paper presents a multi-objective genetic programming approach for sizing and topology optimization of trusses. It aims to find the optimal cross-sectional areas and connectivities between the nodes to achieve a set of trade-off solutions to satisfy all the optimization objective functions subjected to some constraints such as kinematic stability, maximum allowable stress in members and nodal deflections. It also uses the variable-length representation of potential solutions in the shape of computer programs and evolves to the potential final set of solutions. This approach also employs an adaptive mutant factor besides the classical genetic operators to improve the exploring capabilities of Genetic Programming in structural optimization. The intrinsic features of genetic programming help to identify redundant truss members and nodes in the design space, while no violation of constraints occurs. Our approach applied to some numerical examples and found a better non-dominated solution set in the most cases in comparison with the competent methods available in the literature.

A Multi-Objective Quantum-Inspired Genetic Algorithm (Mo-QIGA) for Real-Time Tasks Scheduling in Multiprocessor Environment

This paper aims at providing a multi-objective real-time scheduling algorithm suitable for scheduling of real-time tasks in multiprocessor system with multi-objective criteria. In this proposed Mo-QIGA, scheduling of multi-objective realtime tasks have been targeted for minimizing completion time and total tardiness of each real-time task simultaneously. The exploration of quantum computation in the suggested Mo-QIGA mimics the principles and basic concepts of quantum mechanics. Variable length chromosomes are used in Mo-QIGA and in order to exhibit full Hilbert hyperspace, qubits representation is preferred. In addition, Quantum rotation gate is employed to update the schedules and it obviates classical genetic operators. In order to obtain Pareto-optimal solutions, a random key distribution has been adopted to convert the qubits particles to valid schedule solutions. Moreover, permutation based trimming technique has been introduced for population diversification which leads to good quality schedules. The experiments have been carried out using number of processors varying with real-time tasks priority and processing time. It has been found that the suggested multi-objective quantum inspired genetic algorithm outperforms its classical counterpart in terms of scheduling accuracy and time in multi-objective sense.

An adaptive genetic algorithm for large-size open stack problems

The problem of minimising the maximum number of open stacks arises in many contexts (production planning, cutting environments, very-large-scale-integration circuit design, etc.) and consists of finding a sequence of tasks (products, cutting patterns, circuit gates, etc.) that determines an efficient utilisation of resources (stacks). We propose a genetic approach that combines classical genetic operators (selection, order crossover and pairwise interchange mutation) with an adaptive search strategy, where intensification and diversification phases are obtained by neighbourhood search and by a composite and dynamic fitness function that suitably modifies the search landscape. Computational tests on random and real-world benchmarks show that the proposed approach is competitive with the state of the art for large-size problems, providing better results for some classes of instances.

Genetic algorithm with adaptive and dynamic penalty functions for the selection of cleaner production measures: a constrained optimization problem

This paper presents a new approach of genetic algorithm (GA) to solve the constrained optimization problem. In a constrained optimization problem, feasible and infeasible regions occupy the search space. The infeasible regions consist of the solutions that violate the constraint. Oftentimes classical genetic operators generate infeasible or invalid chromosomes. This situation takes a turn for the worse when infeasible chromosomes alone occupy the whole population. To address this problem, dynamic and adaptive penalty functions are proposed for the GA search process. This is a novel strategy because it will attempt to transform the constrained problem into an unconstrained problem by penalizing the GA fitness function dynamically and adaptively. New equations describing these functions are presented and tested. The effects of the proposed functions developed have been investigated and tested using different GA parameters such as mutation and crossover. Comparisons of the performance of the proposed adaptive and dynamic penalty functions with traditional static penalty functions are presented. The result from the experiments show that the proposed functions developed are more accurate, efficient, robust and easy to implement. The algorithms developed in this research can be applied to evaluate environmental impacts from process operations.

Neural network simulation metamodels and genetic algorithms in analysis and design of manufacturing cells

A manufacturing cell can be modelled using discrete event simulation in order to predict its performance under various combinations of input parameters, related to design issues as well as operation issues. Such models suffer from inherent lack of generality and are useful on a case-by-case basis. Therefore, determination of the best values of design and operation parameters in combination cannot be performed rigorously, but only on an experimentation basis. This work proposes a systematic procedure for optimisation. The final stage is an optimisation procedure using a genetic algorithm which uses the classic genetic operators tuned with due care to avoid local maxima of the fitness function. The actual value of the fitness function corresponding to each breeding case could be obtained by running the discrete event simulation application, but that would make for lengthy response. Therefore, a first stage of the optimising procedure is proposed that calculates the essential part of the fitness function by a neural network metamodel generalising on simulation results. Conditions for successful application of the above procedures are discussed.

Cooperative Metaheuristics for Exploring Proteomic Data

Most combinatorial optimization problems cannot be solved exactly. A class of methods, called metaheuristics, has proved its efficiency to give good approximated solutions in a reasonable time. Cooperative metaheuristics are a sub-set of metaheuristics, which implies a parallel exploration of the search space by several entities with information exchange between them. The importance of information exchange in the optimization process is related to the building block hypothesis of evolutionary algorithms, which is based on these two questions: what is the pertinent information of a given potential solution and how this information can be shared? A classification of cooperative metaheuristics methods depending on the nature of cooperation involved is presented and the specific properties of each class, as well as a way to combine them, is discussed. Several improvements in the field of metaheuristics are also given. In particular, a method to regulate the use of classical genetic operators and to define new more pertinent ones is proposed, taking advantage of a building block structured representation of the explored space. A hierarchical approach resting on multiple levels of cooperative metaheuristics is finally presented, leading to the definition of a complete concerted cooperation strategy. Some applications of these concepts to difficult proteomics problems, including automatic protein identification, biological motif inference and multiple sequence alignment are presented. For each application, an innovative method based on the cooperation concept is given and compared with classical approaches. In the protein identification problem, a first level of cooperation using swarm intelligence is applied to the comparison of mass spectrometric data with biological sequence database, followed by a genetic programming method to discover an optimal scoring function. The multiple sequence alignment problem is decomposed in three steps involving several evolutionary processes to infer different kind of biological motifs and a concerted cooperation strategy to build the sequence alignment according to their motif content.

Partial abductive inference in Bayesian belief networks - an evolutionary computation approach by using problem-specific genetic operators

Abductive inference in Bayesian belief networks (BBNs) is intended as the process of generating the K most probable configurations given observed evidence. When we are interested only in a subset of the network's variables, this problem is called partial abductive inference. Both problems are NP-hard, and so exact computation is not always possible. In this paper, a genetic algorithm is used to perform partial abductive inference in BBNs. The main contribution is the introduction of new genetic operators designed specifically for this problem. By using these genetic operators, we try to take advantage of the calculations previously carried out, when a new individual is evaluated. The algorithm is tested using a widely-used Bayesian network and a randomly generated one, and then compared with a previous genetic algorithm based on classical genetic operators. From the experimental results, we conclude that the new genetic operators preserve the accuracy of the previous algorithm and also reduce the number of operations performed during the evaluation of individuals. The performance of the genetic algorithm is thus improved.

An Evolutionary Approach for the Design of Nonredundant Sensor Networks

In this work, solution strategies for the optimal design of nonredundant observable linear sensor networks are discussed. The Greedy algorithm allows the problem only to be tackled for a subset of optimization criteria. Particular deterministic techniques or general evolutionary strategies are necessary to solve the problem for more complex objective functions. In this context, a procedure based on the application of genetic algorithms (GAs) and linear algebra is presented. Ad hoc operators are designed for the crossover and mutation operations because the classic genetic operators perform poorly. In contrast to ad hoc deterministic codes, which find the design solution for each specific criteria, this strategy allows the problem to be solved with different objective functions using the same implementation. Furthermore, this code is extended to handle multiobjective problems through a modification of only the selection operator. An industrial example is provided to show the efficiency of the algorithm.

A STRUCTURAL OPTIMIZATION APPROACH AND SOFTWARE BASED ON GENETIC ALGORITHMS AND FINITE ELEMENTS

The use of new evolutionary techniques such as genetic algorithms in optimization has increased lately due to their efficiency in solving non-conventional optimization problems. This fact has led to this work in which genetic algorithms and structural finite-clement analysis methods are combined to develop a useful approach to optimize the geometry, size and shape of a structure. A brief overview of genetic algorithms is presented and discussed as well as some classical and advanced genetic operators, which have demonstrated to be useful in structural optimization. Some guidelines showing how the developed software works are presented as well as flowcharts. Finally, two illustrative engineering examples are discussed in order to show the versatility of both the approach and the computational tool.

Optimization of finite element bidimensional models: an approach based on genetic algorithms

This paper deals with the optimization of 2D finite element shapes using the very promising methods based on genetic algorithms. The codification of the design variables is carried out by generating series of strings in binary code. Classical genetic operators such as crossover, mutation and reproduction are used for the optimization process. A more refined operators needed to improve the performance of the process are used as well. Some illustrative examples are presented and discussed