Basic Genetic Algorithm Research Articles

Experimental Comparison between Genetic Algorithm and Ant Colony Optimization on Traveling Salesman Problem

This paper is based on bio-inspired optimization algorithms. Optimization is the process of selecting the best element by following some rules and criteria from some set of available alternatives. In this paper, we have solved Traveling Salesman Problem (TSP) using Swarm Intelligence algorithms and we have compared them. First we have implemented the basic Genetic Algorithm (GA) on TSP. Then we have implemented Ant Colony Optimization (ACO) Algorithm on TSP. In optimization problem, Genetic Algorithm (GA) and Ant Colony Optimization (ACO) Algorithm have been known as good meta-heuristic techniques. GA is designed by adopting the natural law of evolution, while ACO is inspired by the foraging behavior of ant species. Balancing the exploitation-exploration tradeoff is required in ACO. In contrast with the GA implementation, ACO was much easier to control.

An enhanced genetic algorithm for unmanned aerial vehicle logistics scheduling

This paper examines a scheduling problem with heterogeneous logistics unmanned aerial vehicles (UAVs) in urban environment. Different from traditional vehicle routing problem (VRP), it introduces some new characteristics such as the loading capacity, the maximum flight time and the flight speed. As a variant of VRP, the considered scheduling problem is known to be an non-deterministic Polynomial (NP)-hard problem. The UAV scheduling problem model with the heterogeneous UAV settings is formulated first. Secondly, a genetic-based algorithm framework is presented for solving the scheduling problem, in which the encoding/decoding method, the initial population generation method and genetic operations are delicately designed. In order to reduce the search space and faster the execution of this algorithm, a weight-based loading method is adopted. For the purpose of performance evaluation and statistical analysis, the proposed algorithm is compared with the other two existing algorithms. The experimental results show that the presented algorithm can solve this problem efficiently.

Modeling and solving the last-shift period train scheduling problem in subway networks

Train arrivals and departures should be scheduled over a certain period when talking about train timetabling problems. For the midnight train operations, passengers significantly concern about the network transfer issue. Currently, some existing studies address the last train timetabling problem by only optimizing the timetable for the last train on a single subway line whereas this study takes into consideration the complete last-shift period. We first put forward a last-shift train scheduling model aiming to minimize the transfer waiting time and maximize the network connectivity. Two genetic-based algorithms, an integer-coded genetic algorithm (ICGA) and a binary-coded genetic algorithm (BCGA) are developed. The relevance and applicability of the algorithms have been demonstrated by several testing networks and real-world implementation. The ICGA and the branch-and-bound approaches show high efficiency in obtaining the optimal solutions for a small network, while the BCGA approach that bases on an integer-programming model shows low efficiency in addressing problems of sparse solution spaces. However, the branch-and-bound approach has limited ability in solving medium-sized networks. On the contrary, the ICGA generates satisfactory results in solution quality and computational efficiency when applied to large-sized networks.

Non-dominated sorting genetic-based algorithm for exploiting a large-sized fuzzy outranking relation

Electre III is a well-known multiple criteria decision aiding method based on pairwise comparisons. However, it cannot be applied to ranking problems involving many alternatives, because the number of pairwise comparisons can then be rather large. In this paper, we present an evolution-based approach for exploiting large fuzzy outranking relations and deriving a crisp outranking relation with desirable properties. Therefore, the utilization of a fuzzy outranking relation is modeled as a three-objective optimization problem, which is solved by an evolutionary algorithm. The proposed ranking algorithm is a hybrid of the elitist non-dominated sorting genetic algorithm-II (NSGA-II) and a reference point method with the repeated use of a choice mechanism. In addition, a method that portrays the obtained ranking in a Hasse diagram is used for recommendation purposes. We designate the new method RP2-NSGA-II+H. In our experiments, the proposed ranking procedure demonstrates a better performance in terms of ranking error rates than other ranking procedures based on multi-objective evolutionary algorithms. Our experimental results also demonstrate that, with the new procedure, this method can be scaled for hundreds of alternatives.

Retracted] Computer Mathematical Modeling Based on the Improved Genetic Algorithm and Mobile Computing

In order to change the problem of data redundancy in a genetic algorithm, this paper proposes a computer mathematical model based on the combination of an improved genetic algorithm and mobile computing. Combined with the least square method, MATLAB software is used to solve the equations, determine the range of parameters, and solve the estimation parameter range and identification problems. The improved genetic algorithm combined with mobile computing and least square method to establish a mathematical model greatly increased the individual search space and increased the operation rate of 90% compared to the basic genetic algorithm or mobile computing. The results show that the improved genetic algorithm and mobile computing have a certain ability to identify the optimal solution and greatly improve the work efficiency.

The Application of an Adaptive Genetic Algorithm Based on Collision Detection in Path Planning of Mobile Robots

An adaptive genetic algorithm based on collision detection (AGACD) is proposed to solve the problems of the basic genetic algorithm in the field of path planning, such as low convergence path quality, many iterations required for convergence, and easily falling into the local optimal solution. First, this paper introduces the Delphi weight method to evaluate the weight of path length, path smoothness, and path safety in the fitness function, and a collision detection method is proposed to detect whether the planned path collides with obstacles. Then, the population initialization process is improved to reduce the program running time. After comprehensively considering the population diversity and the number of algorithm iterations, the traditional crossover operator and mutation operator are improved, and the adaptive crossover operator and adaptive mutation operator are proposed to avoid the local optimal solution. Finally, an optimization operator is proposed to improve the quality of convergent individuals through the second optimization of convergent individuals. The simulation results show that the adaptive genetic algorithm based on collision detection is not only suitable for simulation maps with various sizes and obstacle distributions but also has excellent performance, such as greatly reducing the running time of the algorithm program, and the adaptive genetic algorithm based on collision detection can effectively solve the problems of the basic genetic algorithm.

Epigenetic Modification of Genetic Algorithm for Outlier Detection

The article presents a new operator in the genetic algorithm. The proposed operator mimics the epigenetic process of prion inheritance. For living organisms, epigenetic processes have a large impact on the differentiation of the population, hence the idea to imitate these processes in genetic algorithms. For the purposes of the experiments, a genetic algorithm to detect outliers was used. The proposed operator mimics the epigenetic process was added to the basic genetic algorithm and the impact of the operator on the effectiveness of the algorithm was assessed. The impact of the proposed modification on the efficiency of the genetic algorithm was tested on six data sets. The article also presents an experimental analysis of the proposed operator.

An Efficient Deployment Model for Maximizing Coverage of Heterogeneous Wireless Sensor Network Based on Harmony Search Algorithm

Due to the increase of Wireless Sensor Network (WSN) technologies demand, the optimal sensor node deployment is considered as one of the most important factors that directly affect the network coverage. Most researches in WSNs that solved the problem of coverage in homogeneous and heterogeneous cases are suffering from many drawbacks such as consumed energy and high cost. In this paper, we propose an efficient deployment model based on probabilistic sensing model (PSM) and harmony search algorithm (HSA) to achieve the balance between the network coverage performance and the network cost in a heterogeneous wireless sensor network (HEWSN). The HSA is used for deployment optimization of HEWSN nodes which makes a balance between the coverage and financial cost. The PSM is used to solve the overlapping problem among the sensors. The performance of the proposed model is analyzed in terms of coverage ratio and cost evaluations. The simulation results showed the capability of the proposed heterogeneous deployment model to achieve maximum coverage and a minimum number of sensors compared to homogeneous deployment. Furthermore, a comparative study with a meta-heuristic genetic-based algorithm in HEWSN has also been conducted, and its results confirm the superiority of the proposed model.

Energy-Saving Production Scheduling in a Single-Machine Manufacturing System by Improved Particle Swarm Optimization

A single-machine scheduling problem that minimizes the total weighted tardiness with energy consumption constraints in the actual production environment is studied in this paper. Based on the properties of the problem, an improved particle swarm optimization (PSO) algorithm embedded with a local search strategy (PSO-LS) is designed to solve this problem. To evaluate the algorithm, some computational experiments are carried out using PSO-LS, basic PSO, and a genetic algorithm (GA). Before the comparison experiment, the Taguchi method is used to select appropriate parameter values for these three algorithms since heuristic algorithms rely heavily on their parameters. The experimental results show that the improved PSO-LS algorithm has considerable advantages over the basic PSO and GA, especially for large-scale problems.

Joint Service Caching and Task Offloading in Multi-Access Edge Computing: A QoE-Based Utility Optimization Approach

In multi-access edge computing (MEC), computation tasks offloaded from users are usually associated with specific services that need to be cached in MEC nodes to enable task execution. The decisions as to which services to cache and which tasks to execute on each resource-limited MEC node are critical to maximizing the offloading efficiency. Moreover, quality of experience (QoE) is a key factor driving offloading decisions, so that limited computing resources can be effectively utilized to keep users satisfied. Therefore, in this letter, we introduce a new QoE-based utility optimization approach to address the problem of joint service caching and task offloading in MEC systems. Our utility model reflects the trade-off between the user’s perception of service latency and the cost the user pays for the allocated computing resources. We formulate total utility maximization as an integer nonlinear programming problem and propose a genetic-based algorithm to solve it efficiently. Finally, evaluation results show that our proposal can significantly improve total user utility over traditional baselines.

Multi-layer collaborative optimization method for solving fuzzy multi-objective integrated process planning and scheduling

Research on integrated process planning and scheduling (IPPS) is of great significance to the improvement of the overall quality of machinery manufacturing system. In the actual manufacturing process, the manufacturing system is often accompanied by some unpredictable uncertain disturbance factors, for instance uncertain processing time of jobs and changes of due date, etc. These uncertain disturbance events will ultimately affect production efficiency and customer satisfaction. Consequently, this paper considers the multi-objective IPPS problem with uncertain processing time and uncertain due date. A multi-layer collaborative optimization (MLCO) method is designed for the fuzzy multi-objective IPPS (FMOIPPS) problem, including three layers. For the process planning layer, the basic genetic algorithm is used to provide various near optimal process plans for the process selection system. For the process selection layer, a multi-objective genetic algorithm (MOGA) is designed to optimize the process selection population. A sharing function method is introduced to maintain population diversity. An individual comprehensive evaluation method is introduced to evaluate non-dominated solutions. The crowded distance, fast non-dominated sorting and elite strategy based on NSGAII is adopted in the proposed MOGA. The external archive method is employed to preserve the non-dominated solutions generated during population evolution. For the scheduling layer, a MOGA with a boundary search strategy is proposed. The boundary search strategy is designed to improve the search ability of boundary solutions. Three optimization objectives are minimizing the spread of fuzzy makespan, minimizing fuzzy makespan and maximizing average customer satisfaction simultaneously. The target of scheduling layer is to make scheduling arrangements for the process information obtained by process selection layer. Through mutual cooperation among each layer, guide the overall optimization process, and finally get satisfactory solutions. Different problem examples of various scales are employed to verify feasibility and effectiveness of the MLCO method. The experimental results indicate that the MLCO method can effectively address FMOIPPS problem.

Artificial Evolution and Design for Multi-Material Additive Manufacturing.

Limitations of the traditional manufacturing methods often force engineered components to be made of single material systems. However, this is going through changes due to the advent of additive manufacturing (AM) methods, as the point-by-point consolidation allows for a possible change of the material constitution within a given part domain. This will give rise to a plethora of new material and property options for the designers, where just human perception may fail to realize the full benefits. Automated design tools integrating material choice, dispersion, analysis, and optimization algorithms need to be developed to assist in finding the optimal multi-material dispersion solutions achieving given performance criteria sets. Considering the fact that the multi-material manufacturing systems are only recently coming into use, design solutions targeting optimal placement of multiple materials are not common. This article addresses this gap, evaluating a numerical model integrated with different optimization schemes to find the optimal material solutions achieving certain preset performance criteria such as combinations of natural frequencies in different degrees of freedom. A case study of three different metaheuristic optimization schemes based on genetic algorithms indicates, first, that it is possible to create a beam with six uniformly spaced natural frequencies and to change these frequencies without modifying the structural geometry; and second that the basic genetic algorithm generally outperforms neural net-based alternatives for this problem. This tailoring of the structural resonance spectrum demonstrates that evolutionary computing combined with multi-material AM can be used to unlock previously unavailable structural functionality.

Developing brief versions of the Moral Foundations Vignettes using a genetic algorithm-based approach

The time-efficient assessment of moral values using systematically validated measures is a high priority in moral psychology research. However, few such options exist for researchers working with Moral Foundations Theory, one of the most popular theories in moral psychology. Across two samples totaling 1336 participants (756 Australian undergraduates and 580 American Mechanical Turk workers), we used a genetic algorithm-based (GA) approach to construct and validate abbreviated versions of the Moral Foundations Vignettes (MFV), a 90-item scale comprising vignettes of concrete violations of each of the six moral foundations. We constructed 36- and 18-item versions of the MFV, demonstrating close correspondence with the complete MFV, and adequate reliability, predictive validity, and factor-analytic goodness of fit for both abbreviated versions. Overall, the abbreviated scales achieve substantially reduced length with minimal loss of information, providing a useful resource for moral psychology researchers.

An improved memetic genetic algorithm based on a complex network as a solution to the traveling salesman problem

A genetic algorithm (GA) is not a good option for finding solutions around in neighborhoods. The current study applies a memetic algorithm (MA) with a proposed local search to the mutation operator of a genetic algorithm in order to solve the traveling salesman problem (TSP). The proposed memetic algorithm uses swap, reversion and insertion operations to make changes in the solution. In the basic GA, unlike in the real world, the relationship between generations has not been considered. This gap is resolved using the proposed complex network to allow selection among possible solutions. The degree measure has been used for analysis the network. Different scenarios have been evaluated to solve seven TSPLib problems. For example, the results indicated that the memetic algorithm with a complex network, the memetic algorithm with the proposed local search and basic GA have 0.31%, 1.15% and 38% errors, respectively, when solving the TSP for 70 cities compared to the best solution in the TSPLib database. These results offered better performance of the memetic algorithm with a complex network compared to the memetic algorithm with the proposed local search and the basic GA. Also, the average run time of the algorithms showed their scalability.

Hyper-Parameter Optimization by Using the Genetic Algorithm for Upper Limb Activities Recognition Based on Neural Networks

A method of Upper Limb Activities Recognition (UPLA) based on Neural Networks is presented. The accuracy of activity recognition will be influenced by the size of sliding window, the overlapping of adjacent sequences and the number of neurons for neural networks. Whereas, there is less work in hyper parameters optimization of neural networks automatically. It is very time-consuming to optimize hyper parameters by experts through an experience and error approach. In the paper, Genetic algorithm is used to find the best hyper parameters automatically: the size of sliding window, the overlapping of adjacent sequences and the number of neurons for neural networks. The basic genetic algorithm has a slow convergence problem and it is very easy to fall into a local optimum. To solve the problem, the population selection mechanism is improved. A comparison is made for the improving method with seven traditional classification algorithms and convolutional neural network, an accuracy of 97.9% is reached by using the new method. Finally, an App is developed that can collect and recognize upper limb activity in real time.

ConEx: Efficient Exploration of Big-Data System Configurations for Better Performance

Configuration space complexity makes the big-data software systems hard to configure well. Consider Hadoop, with over nine hundred parameters, developers often just use the <i>default</i> configurations provided with Hadoop distributions. The opportunity costs in lost performance are significant. Popular learning-based approaches to auto-tune software does not scale well for big-data systems because of the high cost of collecting training data. We present a new method based on a combination of <i>Evolutionary Markov Chain Monte Carlo (EMCMC)</i> sampling and cost reduction techniques to find better-performing configurations for big data systems. For cost reduction, we developed and experimentally tested and validated two approaches: using scaled-up big data jobs as proxies for the objective function for larger jobs and using a dynamic job similarity measure to infer that results obtained for one kind of big data problem will work well for similar problems. Our experimental results suggest that our approach promises to improve the performance of big data systems significantly and that it outperforms competing approaches based on random sampling, basic genetic algorithms (GA), and predictive model learning. Our experimental results support the conclusion that our approach strongly demonstrates the potential to improve the performance of big data systems significantly and frugally.

QUANTUMIZED GENETIC ALGORITHM FOR SEGMENTATION AND OPTIMIZATION TASKS

Specialists mostly assess the skeletal maturity of short-height children by observing their left hand X-Ray image (radiograph), whereas precise separation of areas capturing the bones and growing plates is always not possible by visual inspection. Although a few attempts are made to estimate a suitable threshold for segmenting digitized radiograph images, their results are not still promising. To finely estimate segmentation thresholds, this paper presents the quantumized genetic algorithm (QGA) that is the integration of quantum representation scheme in the basic genetic algorithm (GA). This hybridization between quantum inspired computing and GA has led to an efficient hybrid framework that achieves better balance between the exploration and the exploitation capabilities. To assess the performance of the proposed quantitative bone maturity assessment framework, we have collected an exclusive dataset including 65 left-hand digitized images, aged from 3 to 13 years. Thresholds are estimated by the proposed method and the results are compared to harmony search algorithm (HSA), particle swarm optimization (PSO), quantumized PSO and standard GA. In addition, for more comparison of the proposed method and the other mentioned evolutionary algorithms, ten known benchmarks of complex functions are considered for optimization task. Our results in both segmentation and optimization tasks show that QGA and GA provide the best optimization results in comparison with the other mentioned algorithms. Moreover, the empirical results demonstrate that QGA is able to provide better diversity than that of GA.

Neural network control system of cooperative robot based on genetic algorithms

Attitude detection of cooperative robots can help robots recognize environment, understand tasks, and improve the safety, accuracy and efficiency of robot work. For robots with unknown configuration parameters, their configuration can be estimated; it is convenient for further kinematics and dynamics analysis. For an economical cooperative robot, its kinematic parameters can be corrected by attitude detection. An independent robot attitude detection system can be used as a secondary auxiliary system, and the fault diagnosis system of the robot is composed of the high-precision encoder system of the robot body. This paper studies the neural network control system of cooperative robot based on genetic algorithm. In this paper, the robot is taken as the research object. Aiming at its strong conjunction, non-linearity and multivariable characteristics, the problem of robot motion control based on neural network is mainly discussed. On this basis, the basic genetic algorithm and an improved genetic algorithm called messy are used to optimize the structure of the neural network, so as to better realize the motion control problem of the robot. The results of this study show that: After optimization with messy genetic algorithm, not only the tracking effect is better, but also the number of hidden layer nodes of the network is reduced from 12 to 7. This greatly simplifies the structure of the network and makes the design and training of the network relatively simple.

Numerical weld pool simulation for the accuracy improvement of inline weld depth measurement based on optical coherence tomography

The aluminum alloy AA6082 is often used in high-voltage storage systems due to its favorable electrical and mechanical properties. Laser beam welding is a flexible process for producing the required welds. To protect the sensitive components of the battery cells, the required weld depth must be maintained in order to avoid destruction of the components and to reduce the number of defective parts. Optical coherence tomography (OCT) is a promising method for the inline monitoring of the deep welding process, since the depth of the vapor capillary can be measured without the influence of process emissions such as process light or welding fumes. Due to the robustness of the measurement method, the measured depth of the keyhole can be used as an input variable for a process control. However, there is a geometric difference between the optically detectable keyhole depth and the melt pool depth. This depth is additionally determined by fluid dynamic processes in the melt pool. In order to consider the difference between the melt pool depth and the keyhole depth when evaluating OCT measurement signals, a thermal simulation model was developed. By iteratively adapting the keyhole geometry in the simulation on the basis of genetic algorithms, an optimization of the correlation was achieved. The model was calibrated based on the comparison of metallographic cross sections with the calculated melt pool geometries. The results were used to provide a database for improving the performance of the signal processing algorithms. Based on the calculation results, it was possible to determine and quantify a dependence of the melt film thickness below the keyhole on the welding process parameters. By including the melt thickness below the keyhole, it was possible to increase the accuracy in the interpretation of OCT data with regard to the welding depth. The quantity can thus be used as an input quantity for signal processing algorithms for OCT data evaluation.

Solving the Robust Design Problem for a Two-Commodity Flow Network with Node Failure

The robust design in a flow network is one of the most important problems. It is defined as searching the optimal capacity that can be assigned to the nodes such that the network still survived even under the node’s failure. This problem is considered NP-hard. So, this study presents a genetic-based algorithm to determine the maximum node capacity for a two-commodity flow network with node failure. I.e., searching the minimum sum of the assigned capacities and the maximum network reliability. The obtained results show that The proposed GA-based algorithm succeeded to solve the robust problem for the two-commodity flow network considering the node’s failure.