• 정보처리학회논문지B
• /
• 제11B권1호
• /
• pp.77-82
• /
• 2004

강화학습에서 temporal-credit 할당 문제 즉, 에이전트가 현재 상태에서 어떤 행동을 선택하여 상태전이를 하였을 때 에이전트가 선택한 행동에 대해 어떻게 보상(reward)할 것인가는 강화학습에서 중요한 과제라 할 수 있다. 본 논문에서는 조합최적화(hard combinational optimization) 문제를 해결하기 위한 새로운 메타 휴리스틱(meta heuristic) 방법으로, greedy search뿐만 아니라 긍정적 반응의 탐색을 사용한 모집단에 근거한 접근법으로 Traveling Salesman Problem(TSP)를 풀기 위해 제안된 Ant Colony System(ACS) Algorithms에 Q-학습을 적용한 기존의 Ant-Q 학습방범을 살펴보고 이 학습 기법에 다양화 전략을 통한 상태전이와 TD-오류를 적용한 학습방법인 Ant-TD 강화학습 방법을 제안한다. 제안한 강화학습은 기존의 ACS, Ant-Q학습보다 최적해에 더 빠르게 수렴할 수 있음을 실험을 통해 알 수 있었다.

• Industrial Engineering and Management Systems
• /
• 제7권2호
• /
• pp.171-181
• /
• 2008

Home care, known also as domiciliary care, is part of the community care service that is a responsibility of the local government authorities in the UK as well as many other countries around the world. The aim is to provide the care and support needed to assist people, particularly older people, people with physical or learning disabilities and people who need assistance due to illness to live as independently as possible in their own homes. It is performed primarily by care workers visiting clients' homes where they provide help with daily activities. This paper is concerned with the dispatching of care workers to clients in an efficient manner. The optimized routine for each care worker determines a schedule to achieve the minimum total cost (in terms of distance traveled) without violating the capacity and time window constraints. A collaborative population-based meta-heuristic called Particle Swarm Optimization (PSO) is applied to solve the problem. A particle is defined as a multi-dimensional point in space which represents the corresponding schedule for care workers and their clients. Each dimension of a particle represents a care activity and the corresponding, allocated care worker. The continuous position value of each dimension determines the care worker to be assigned and also the assignment priority. A heuristic assignment scheme is specially designed to transform the continuous position value to the discrete job schedule. This job schedule represents the potential feasible solution to the problem. The Earliest Start Time Priority with Minimum Distance Assignment (ESTPMDA) technique is developed for generating an initial solution which guides the search direction of the particle. Local improvement procedures (LIP), insertion and swap, are embedded in the PSO algorithm in order to further improve the quality of the solution. The proposed methodology is implemented, tested, and compared with existing solutions for some 'real' problem instances.

• Industrial Engineering and Management Systems
• /
• 제13권2호
• /
• pp.163-171
• /
• 2014

Under amazing increase in markets and certain demand on qualified service in the delivery system, global logistic optimization is becoming a keen interest to provide an essential infrastructure coping with modern competitive prospects. As a key technology for such deployment, we have been engaged in the practical studies on vehicle routing problem (VRP) in terms of Weber model, and developed a hybrid approach of meta-heuristic methods and the graph algorithm of minimum cost flow problem. This paper extends such idea to multi-depot VRP so that we can give a more general framework available for various real world applications including those in green or low carbon logistics. We show the developed procedure can handle various types of problem, i.e., delivery, direct pickup, and drop by pickup problems in a common framework. Numerical experiments have been carried out to validate the effectiveness of the proposed method. Moreover, to enhance usability of the method, Google Maps API is applied to retrieve real distance data and visualize the numerical result on the map.

• Structural Engineering and Mechanics
• /
• 제29권4호
• /
• pp.391-410
• /
• 2008

In this article, a harmony search algorithm is presented for optimum design of steel frame structures. Harmony search is a meta-heuristic search method which has been developed recently. It is based on the analogy between the performance process of natural music and searching for solutions of optimization problems. The design algorithms obtain minimum weight frames by selecting suitable sections from a standard set of steel sections such as American Institute of Steel Construction (AISC) wide-flange (W) shapes. Stress constraints of AISC Load and Resistance Factor Design (LRFD) and AISC Allowable Stress Design (ASD) specifications, maximum (lateral displacement) and interstorey drift constraints, and also size constraint for columns were imposed on frames. The results of harmony search algorithm were compared to those of the other optimization algorithms such as genetic algorithm, optimality criterion and simulated annealing for two planar and two space frame structures taken from the literature. The comparisons showed that the harmony search algorithm yielded lighter designs for the design examples presented.

• Advances in Computational Design
• /
• 제9권2호
• /
• pp.115-136
• /
• 2024

Stochastic optimization methods have been extensively studied for structural optimization in recent decades. In this study, a novel algorithm named the CA-SA method, is proposed for topology optimization of steel double-layer grid structures. The CA-SA method is a hybridized algorithm combining the Simulated Annealing (SA) algorithm and the Cellular Automata (CA) method. In the CA-SA method, during the initial iterations of the SA algorithm, some of the preliminary designs obtained by SA are placed in the cells of the CA. In each successive iteration, a cell is randomly chosen from the CA. Then, the "local leader" (LL) is determined by selecting the best design from the chosen cell and its neighboring ones. This LL then serves as the leader for modifying the SA algorithm. To evaluate the performance of the proposed CA-SA algorithm, two square-on-square steel double-layer grid structures are considered, with discrete cross-sectional areas. These numerical examples demonstrate the superiority of the CA-SA method over SA, and other meta-heuristic algorithms reported in the literature in the topology optimization of large-scale skeletal structures.

• 한국인터넷방송통신학회논문지
• /
• 제21권6호
• /
• pp.33-41
• /
• 2021

본 논문은 NP-난제로 알려진 능력한정 최소신장트리 문제(CMST)의 해를 다항시간으로 찾을 수 있는 규칙을 가진 휴리스틱 탐욕 알고리즘을 제안하였다. CMST는 다항시간으로 해를 구하는 방법인 EW 알고리즘의 성능이 좋지 않아 컴퓨터 프로그램의 도움을 받는 메타휴리스틱 기법들을 적용하고 있다. 그러나 메타휴리스틱 기법들도 최적 해를 찾지 못하는 성능의 한계를 보였다. 본 논문에서는 컴퓨터 도움 없이 시각적으로 손으로 CMST의 해를 찾는 규칙을 제시하였다. 제안된 방법은 먼저 MST를 작도하고, MST로부터 초기 CMST의 실현 가능 해를 구하고, CMST의 해를 개선하기 위해 서브트리의 게이트들이 근 노드에 보다 근접하도록 설정하는 최적화 과정을 수행하였다. 제안된 알고리즘을 OR-LIB의 10개 데이터, Q=3,5,10의 30개 경우에 대해 적용한 결과 최상의 성능을 보였다.

• Journal of Information Processing Systems
• /
• 제13권4호
• /
• pp.1000-1013
• /
• 2017

Clustering is a NP-hard problem that is used to find the relationship between patterns in a given set of patterns. It is an unsupervised technique that is applied to obtain the optimal cluster centers, especially in partitioned based clustering algorithms. On the other hand, cat swarm optimization (CSO) is a new meta-heuristic algorithm that has been applied to solve various optimization problems and it provides better results in comparison to other similar types of algorithms. However, this algorithm suffers from diversity and local optima problems. To overcome these problems, we are proposing an improved version of the CSO algorithm by using opposition-based learning and the Cauchy mutation operator. We applied the opposition-based learning method to enhance the diversity of the CSO algorithm and we used the Cauchy mutation operator to prevent the CSO algorithm from trapping in local optima. The performance of our proposed algorithm was tested with several artificial and real datasets and compared with existing methods like K-means, particle swarm optimization, and CSO. The experimental results show the applicability of our proposed method.

• 산업경영시스템학회지
• /
• 제44권2호
• /
• pp.58-65
• /
• 2021

Meta-heuristic algorithms have been developed to efficiently solve difficult problems and obtain a global optimal solution. A common feature mimics phenomenon occurring in nature and reliably improves the solution through repetition. And at the same time, the probability is used to deviate from the regional optimal solution and approach the global optimal solution. This study compares the algorithm created based on the above common points with existed SA and HS to show advantages in time and accuracy of results. Existing algorithms have problems of low accuracy, high memory, long runtime, and ignorance. In a two-variable polynomial, the existing algorithms show that the memory increases and the accuracy decrease. In order to improve the accuracy, the new algorithm increases the number of initial inputs and increases the efficiency of the search by introducing a direction using vectors. And, in order to solve the optimization problem, the results of the last experiment were learned to show the learning effect in the next experiment. The new algorithm found a solution in a short time under the experimental conditions of long iteration counts using a two-variable polynomial and showed high accuracy. And, it shows that the learning effect is effective in repeated experiments.

• 한국정보전자통신기술학회논문지
• /
• 제15권5호
• /
• pp.299-307
• /
• 2022

Harmony Search(HS)는 비교적 최근에 개발된 메타 휴리스틱 최적화 알고리즘으로 최근 이에 관한 연구가 다양하게 진행되고 있다. HS는 음악인의 즉홍 연주를 기반으로 하고 있으며 목적변수는 악기의 역할을 한다. 그러나 각 악기는 음대역만 주어질 뿐 음악의 기본이라 할 수 있는 음계의 개념이 없다. 본 연구에서는 기존 HS에 음계를 도입하고 대역폭을 양자화하여 알고리즘의 성능을 향상시키고자 한다. 도입한 음계는 음대역 범위에서 무작위로 초기화되던 기존 방식을 대신하여 HM 초기화에 적용하였다. 양자화 단계는 임의로 정할 수 있도록 하였으며 이를 통해 알고리즘 초반에는 상대적으로 큰 대역폭을 사용하여 알고리즘의 탐색성을 향상시키고 후반에는 작은 대역폭을 통해 탐지성을 향상시키고자 하였다. 음계 도입과 대역폭 양자화를 통하여 기존 HS보다 초기값에 따른 알고리즘 성능 편차를 줄이고 알고리즘 수렴속도 및 성공률을 향상시킬 수 있었다. 본 연구의 성과는 여러 함수에 대한 최적화 수치 예를 종래의 방식과 비교하여 확인하였다. 구체적인 비교 수치는 모의실험에 서술하였다.

• 전자공학회논문지CI
• /
• 제40권6호
• /
• pp.87-94
• /
• 2003

휴리스틱 알고리즘 연구에 있어서 중요한 분야 중 하나가 강화(Intensification)와 다양화(Diversification)의 조화를 맞추는 문제이다. 개미 집단 최적화(Ant Colony Optimization, ACO)는 최근에 제안된 조합 최적화 문제를 해결하기 위한 메타휴리스틱 탐색 방법으로, 그리디 탐색(greedy search)뿐만 아니라 긍정적 반응의 탐색을 사용한 모집단에 근거한 접근법으로 순회 판매원 문제(Traveling Salesman Problem, TSP)를 풀기 위해 처음으로 제안되었다. 본 논문에서는 ACO접근법의 하나인 개미 집단 시스템(Ant Colony System ACS)에서 강화와 다양화의 조화를 통한 성능향상기법에 대해 알아본다. 먼저 에이전트들의 방문 횟수 적용을 통한 상태전이는 탐색 영역을 넓힘으로써 에이전트들이 더욱 다양하게 탐색하게 한다. 그리고, 전역 갱신 규칙에서 전역 최적 경로만 갱신하는 전통적인 ACS알고리즘에서 대하여, 경로 사이클을 구성한 후 각 경로에 대해 긍정적 강화를 받는 엘리트 경로를 구분하는 기준을 정하고, 그 기준에 의해 추가 강화하는 방법을 제안한다. 그리고 여러 조건 하에서 TSP문제를 풀어보고 그 성능에 대해 기존의 ACS 방법과 제안된 방법을 비교 평가해, 해의 질과 문제를 해결하는 속도가 우수하다는 것을 증명한다.