• Journal of Korean Society of Transportation
• /
• v.22 no.2 s.73
• /
• pp.43-54
• /
• 2004

This study analyzes characteristics and applicability of genetic algorithms and genetic operators to optimize highway alignments. Genetic algorithms, one of artificial intelligence techniques, are fast and efficient search algorithms for generating, evaluation and finding optimal highway alignment alternatives. The performance of genetic algorithms as an optimal search tool highly depends on genetic operators that are designed as a problem-specific. This study adopts low mutation operators(uniform mutation operator, straight mutation operator, non-uniform mutation operator whole non-uniform mutation operator) to explore whole search spaces, and four crossover operators(simple crossover operator, two-point crossover operator, arithmetic crossover operator, heuristic crossover operator) to exploit food characteristics of the best chromosome in previous generations. A case study and a sensitivity analysis have shown that the eight problem-specific operators developed for optimizing highway alignments enhance the search performance of genetic algorithms, and find good solutions(highway alignment alternatives). It has been also found that a mixed and well-combined use of mutation and crossover operators is very important to balance between pre-matured solutions when employing more crossover operators and more computation time when adopting more mutation operators.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.4
• /
• pp.25-32
• /
• 2015

The Genetic algorithm is a search algorithm using evaluation, genetic operator, natural selection to populational solution iteratively. The convergence and divergence characteristic of genetic algorithm are affected by selection strategy, generation replacement method, genetic operator when genetic algorithm is designed. This paper proposes fast convergence genetic algorithm for time-limited robot path planning. In urgent situation, genetic algorithm for robot path planning does not have enough time for computation, resulting in quality degradation of found path. Proposed genetic algorithm uses fast converging selection strategy and generation replacement method. Proposed genetic algorithm also uses not only traditional crossover and mutation operator but additional genetic operator for shortening the distance of found path. In this way, proposed genetic algorithm find reasonable path in time-limited situation.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.11 no.2 s.54
• /
• pp.1-9
• /
• 2007

The method of generating the artificial acceleration time histories for seismic analysis based on genetic algorithms is presented. For applying to the genetic algorithms, the frequencies are selected as the decision variables eventually to be genes. An arithmetic average crossover operator and an arithmetic ratio mutation operator are suggested in this study. These operators as well as the typical simple crossover operator are utilized in generating the artificial acceleration time histories corresponding to the specified design response spectrum. Also these generated artificial time histories are checked whether their outward features are to be coincident with the recorded earthquake motion or not. The features include envelope shape, correlation condition between 2 horizontal components of motion, and the relationship of max. acceleration, max. velocity and max. displacement of ground.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.37 no.5
• /
• pp.29-44
• /
• 2000

Adaptation of operator probabilities is one of the most important and promising issues in evolutionary computation areas. This is because the setting of appropriate probabilities is not only very tedious and difficult but very important to the performance improvement of genetic algorithms. Many researchers have introduced their algorithms for setting or adapting operator probabilities. Experimental results in most previous works, however, have not been satisfiable. Moreover, Tuson have insisted that “the adaptation is not necessarily a good thing” in his papers［$^1$$^2$］. In this paper, we propose a self-tuning scheme for adapting operator probabilities in genetic algorithms. Our scheme was extensively tested on four function optimization problems and one combinational problem； and compared to simple genetic algorithms with constant probabilities and adaptive genetic algorithm proposed by Srinivas et al［$^3$］. Experimental results showed that our scheme was superior to the others. Our scheme compared with previous works has three advantages： less computational efforts, co-evolution without additional operations for evolution of probabilities, and no need of additional parameters.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.6 s.38
• /
• pp.85-92
• /
• 2005

In this Paper, we develope an effective web-based learning contents using interactive genetic algorithms with modified mutation operation. In the interactive genetic algorithm, reciprocal exchange mutation is used. But. we modify the mutation operator to improve the learning effects. The new web-based learning contents using interactive genetic algorithm provide the dynamic learning contents providing and real-time test system. Especially, learners can execute the interactive genetic algorithm according to the learners' characters and interests to select the efficient learning environments and contents sequences.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10c
• /
• pp.42-44
• /
• 1998

Sequential Ordering Problem(SOP)은 여러 개의 도시를 방문함에 있어 '어떤 도시를 다른 도시보다 먼저 방문해야 한다'는 선행제약이 있는 비대칭 순회 세일즈맨 문제(Traveling Salesman Problem)로서, 주어진 선행 제약을 만족하면서 모든 도시를 한번씩만 경유하는 가장 짧은 경로를 찾는 NP-Complete에 속하는 문제이다. 유전자 알고리즘은 SOP와 같은 조합 최적화문제에 대해 유용한 메타휴리스틱의 한가지이다. 본 논문에서는 SOP에 유전자 알고리즘을 적용할 때, 선행제약을 만족하는 해를 생성하는데 사용할 수 있는 선행관계유지 유전 연산자를 소개하고 이를 비교한다. 비교하는 유전 연산자는 선행관계유지 교차연산자, 선행관계유지 순서기반 교차연산자, 최대부분순서/임의삽입 연산자, 선행관계유지 간선재결합 연산자이다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.04b
• /
• pp.643-645
• /
• 2002

본 논문은 영상의 isophote정보를 constraint로 사용만 유전자 알고리즘 (Genetic Algorithms) 기반의 컬러 영상 복원방법을 제안한다. 제안만 방법은 오염된 관측 영상으로부터 원 영상으로 복원하기 위해. 영상 복원 문제인 illposed 문제를 유전자 알고리즘을 이용하여 비용함수(cost funcition)가 최소가 되도록 하는 최적화 문제로 모델링 한다. 본 논문에서 제안만 방법은 영상에서 같은 밝기 값을 가진 영역의 경계선을 나타내는 isophole 를 비용함수의 정칙화(regularization) 연산자로 사용하여 영상을 복원한다. 사용자가 복원할 영역을 지정만 후, 유전자 알고리즘을 사용하여 복원될 영역치 isophote 를 자연스럽게 유지하도록 복원한다. 제안한 방법은 디지털 비디오에서 상업적인 광고나, 자막 측은 로고등을 제거하는데 사용될 수 있으며, 실험 결과, 일반적으로 영상 복원에 많이 사용하는 Constraint 로 라플라시안(Laplaoian) 연산자보다 isophote를 정칙화 연산자로 사용함으로써 효율적으로 영상이 복원됨을 알 수 있다.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.5 s.37
• /
• pp.179-186
• /
• 2005

This paper describes a genetic algorithm and compares three crossover operators for Rural Postman Problem with Time Windows (RPPTW). The RPPTW which is a multiobjective optimization problem, is an extension of Rural Postman Problem(RPP) in which some service places (located at edge) require service time windows that consist of earliest time and latest time. Hence, RPM is a m띤tieect optimization Problem that has minimal routing cost being serviced within the given time at each service Place. To solve the RPPTW which is a multiobjective optimization problem, we obtain a Pareto-optimal set that the superiority of each objective can not be compared. This Paper performs experiments using three crossovers for 12 randomly generated test problems and compares the results. The crossovers using in this Paper are Partially Matched Exchange(PMX) Order Exchange(OX), and Modified Order Exchange(MOX) which is proposed in this paper. For each test problem, the results show the efficacy of MOX method for RPPTW.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.12
• /
• pp.55-61
• /
• 2020

Genetic Algorithm(GA) is applied to a problem that could not figure out its solution in a straightway. It is called as NP-hard problem. GA requires a high-performance system to be run on since the high-cost operations are needed such as crossover, selection, and mutation. Moreover, the scale of the problem domain is normally huge. That is why the straightway cannot be applied. To reduce the drawback of high-cost requirements, we try to answer if all the operations including mutation are necessary for all cases. In the experiment, we set up two cases of with/without mutation operations and gather the number of generations and the fitness of a solution. The subject in the experiment is Travelling Salesman Problem(TSP), which is one of the popular problems solved by GA. As a result, the cases with mutation operation are not faster and the solution is fitter than the case with mutation operation. From the result, the conclusion is that mutation operation does not always need for a better solution in a faster way.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.04a
• /
• pp.341-344
• /
• 2007

본 논문에서는 계층적 공정 경쟁 개념을 병렬 유전자 알고리즘에 적용하여 계층적 공정 경쟁 기반 병렬유전자 알고리즘 (Hierarchical Fair Competition Genetic Algorithm: HFCGA)을 구현하였을 뿐만 아니라 실수코딩 유전자 알고리즘(Real-Coded Genetic Algorithm: RCGA)에서 좋은 성능을 갖는 산술교배(Arithmetic crossover), 수정된 단순교배(modified simple crossover) 그리고 UNDX(unimodal normal distribution crossover)등의 다양한 교배연산자들을 적용, 분석함으로써 개선된 병렬 유전자 알고리즘을 제안하였다. UNDX연산자는 다수의 부모(multiple parents)를 이용하여 부모들의 기하학적 중심(geometric center)에 근접하게 정규분포를 이루며 생성된다. 본 논문은 UNDX를 이용한 HFCGA모델을 구현하고 함수파라미터 최적화 문제에 많이 쓰이는 함수들에 적용시킴으로써 그 성능의 우수성을 증명 한다.