• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2008년도 춘계학술대회논문집
• /
• pp.113-116
• /
• 2008

This work presents numerical optimization for design of staggered arrays of dimples printed on opposite surfaces of a cooling channel with a fast and elitist Non-Dominated Sorting of Genetic Algorithm (NSGA-II) of multi-objective optimization. As Pareto optimal front produces a set of optimal solutions, the trends of objective functions with design variables are predicted by hybrid multi-objective evolutionary algorithm. The problem is defined by three non-dimensional geometric design variables composed of dimpled channel height, dimple print diameter, dimple spacing and dimple depth to maximize heat transfer rate compromising with pressure drop. Twenty designs generated by Latin hypercube sampling were evaluated by Reynolds-averaged Navier-Stokes solver and the evaluated objectives were used to construct Pareto optimal front through hybrid multi-objective evolutionary algorithm. The optimum designs were grouped by k-mean clustering technique and some of the clustered points were evaluated by flow analysis. With increase in dimple depth, heat transfer rate increases and at the same time pressure drop also increases, while opposite behavior is obtained for the dimple spacing. The heat transfer performance is related to the vertical motion of the flow and the reattachment length in the dimple.

• 한국컴퓨터정보학회논문지
• /
• 제10권5호
• /
• pp.179-186
• /
• 2005

본 논문에서는 유전자 알고리즘을 이용한 rural Postman problem with Time windows(RPPTW) 해법을 위해 유전자 알고리즘에 사용되는 교차 연산자를 제안하고, 기존의 교차 연산자와 비교한다. RPPTW는 다중목적 최적화 문제로서, Rural Postman Problem(RPP)에 서비스 시간 제한을 위한 시간 윈도우(Time Windows)를 두고 제한된 시간 내에 서비스를 받을 수 있도록 구성된 문제이다. 따라서, RPPTW는 주어진 시간 내에 서비스를 받으면서 최소 비용으로 라우팅을 하는 다중 목적 최적화 문제이다. 다중 목적 최적화 문제인 RPPTW를 해결하기 위해서는 Pareto-optimal 집합을 구해야 한다. Pareto-optimal 집합은 각 목적값들의 우수성을 비교할 수 없는 집합이다. 본 논문에서는 12개의 임의로 생성된 문제들에 대해 3개의 교차 연산자를 사용하여 실험을 하여 그 결과를 비교하였다. 본 논문에서 사용된 교차 연산자들은 PMX(Partially Matched Exchange), OX(Order Exchange), 그리고 본 논문에서 제안한 MOX(Modified Order Exchange)이다. 각 문제들에 대한 실험 결과를 통해서 RPPTW를 위한 교차 연산자 중에 본 논문에서 제안한 MOX방법이 효율적임를 알 수 있었다.

• 한국컴퓨터산업학회논문지
• /
• 제6권3호
• /
• pp.539-550
• /
• 2005

본 논문은 연결비용 및 메시지 지연시간을 최소화하는 최적 LAM 설계를 위한 파레토 계층-적소 영역 유전자 알고리즘의 응용이다. 이러한 연구과제를 수행하기 위하여 LAN 설계를 위한 적합한 가능해의 표현 방법을 제안하고, LAN 설계 시 선택되는 여러 위상(topology) 형태는 지역적이고, 내선망에 가장 보편적으로 쓰이는 스패닝 트리 형태에 한정하여 유전자 알고리즘을 적용한다. 스패닝 트리를 초기 모집단의 염색체로 표현하기 위해 $Pr\ddot{u}fer$ 수 개념을 이용한다. $Pr\ddot{u}fer$ 수는 간결하면서도 스패닝 트리의 특징을 잘 반영하며 염색체 표현에 적합하다. 유전자 알고리즘에 의해 이중 목적에 맞는 적응도 평가함수의 산출하고, 파레토 계층-적소영역 선별 알고리즘을 사용하며, 적합한 유전연산자를 적용함으로써 다양한 해공간을 탐색하여 최적 LAN을 하여, 제안되었던 알고리즘이 꽤 짧은 시간에 다목적 LAN 설계 문제의 좋은 해답들을 제공할 수 있는 것으로 나타났다.

• Management Science and Financial Engineering
• /
• 제13권2호
• /
• pp.29-46
• /
• 2007

This paper makes an attempt to compare the two important methods for finding solutions of multi-item inventory problem with more than one conflicting objectives. Panda et al.[9] discusses a distance-based method to find the best possible compromise solution with variation of priority under the given weight structure. In this paper, the problem in [9] is revisited through the Pareto-optimal front of genetic algorithm with the help of a situation of retail stocking of FMCG business. The advantages of using the solutions from the perspective of the decision maker obtained through multi-objective optimization are highlighted in terms of population search, weighted goals and priority structure, cost, set of compromise solutions along with prevention of stock-out situation.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회/대한산업공학회 2003년도 춘계공동학술대회
• /
• pp.1071-1076
• /
• 2003

Evolutionary algorithm is recognized as a promising approach to solving multi-objective combinatorial optimization problems. When no preference information of decision makers is given, multi-objective optimization problems have been commonly used to search for diverse and good Pareto optimal solution. In this paper we propose a new multi-objective evolutionary algorithm based on competitive coevolutionary algorithm, and demonstrate the applicability of the algorithm. The proposed algorithm is designed to promote both population diversity and rapidity of convergence. To achieve this, the strategies of fitness evaluation and the operation of the Pareto set are developed. The algorithm is applied to job shop scheduling problems (JSPs). The JSPs have two objectives: minimizing makespan and minimizing earliness or tardiness. The proposed algorithm is compared with existing evolutionary algorithms in terms of solution quality and diversity. The experimental results reveal the effectiveness of our approach.

• 한국경영과학회지
• /
• 제35권3호
• /
• pp.25-43
• /
• 2010

We consider a multi-objective balancing and sequencing problem in mixed model assembly lines, which is important for an efficient use of the assembly lines. In this paper, we present a neighborhood symbiotic evolutionary algorithm to simultaneously solve the two problems of balancing and model sequencing under multiple objectives. We aim to find a set of well-distributed solutions close to the true Pareto optimal solutions for decision makers. The proposed algorithm has a two-leveled structure. At Level 1, two populations are operated : One consists of individuals each of which represents a partial solution to the balancing problem and the other consists of individuals for the sequencing problem. Level 2, which is an upper level, works one population whose individuals represent the combined entire solutions to the two problems. The process of Level 1 imitates a neighborhood symbiotic evolution and that of Level 2 simulates an endosymbiotic evolution together with an elitist strategy to promote the capability of solution search. The performance of the proposed algorithm is compared with those of the existing algorithms in convergence, diversity and computation time of nondominated solutions. The experimental results show that the proposed algorithm is superior to the compared algorithms in all the three performance measures.

• 산업공학
• /
• 제24권3호
• /
• pp.220-230
• /
• 2011

The multi-reservoir operation problem for efficient utilization of water resources involves conflicting objectives, and the problem can be solved by varying weight coefficient on objective functions. Accordingly, decision makers need to choose appropriate weight coefficients balancing the trade-offs among multiple objectives. Although the appropriateness of the weight coefficients may depend on the total amount of water inflow, reservoir operating policy may not be changed to a certain degree for different hydrological conditions on inflow. Therefore, we propose to use fuzzy Data Envelopment Analysis (DEA) to rank the weight coefficients in consideration of the inflow variation. In this approach, we generate a set of Paretooptimal solutions by applying different weight coefficients on Coordinated Multi-reservoir Operating Model. Then, we rank the Pareto-optimal solutions or the corresponding weight coefficients by using Fuzzy DEA model. With the proposed approach, we can suggest the best weight coefficients that can produce the appropriate Pareto-optimal solution considering the uncertainty of inflow, whereas the general DEA model cannot pinpoint the best weight coefficients.

• 한국경영과학회지
• /
• 제32권1호
• /
• pp.77-91
• /
• 2007

In this paper, we present a symbiotic evolutionary algorithm for multi-objective optimization. The goal in multi-objective evolutionary algorithms (MOEAs) is to find a set of well-distributed solutions close to the true Pareto optimal solutions. Most of the existing MOEAs operate one population that consists of individuals representing the entire solution to the problem. The proposed algorithm has a two-leveled structure. The structure is intended to improve the capability of searching diverse and food solutions. At the lower level there exist several populations, each of which represents a partial solution to the entire problem, and at the upper level there is one population whose individuals represent the entire solutions to the problem. The parallel search with partial solutions at the lower level and the Integrated search with entire solutions at the upper level are carried out simultaneously. The performance of the proposed algorithm is compared with those of the existing algorithms in terms of convergence and diversity. The optimization problems with continuous variables and discrete variables are used as test-bed problems. The experimental results confirm the effectiveness of the proposed algorithm.

• 설비공학논문집
• /
• 제17권10호
• /
• pp.938-946
• /
• 2005

Shape optimization of internally finned circular tube has been peformed for periodically fully developed turbulent flow and heat transfer. The physical domain considered in this study is very complicated due to periodic boundary conditions both streamwise and circumferential directions. Therefore, Pareto frontier sets of a heat exchanger can be acquired by coupling the CFD and the multi-objective genetic algorithm, which is a global optimization technique. The optimal values of fin widths $(d_1,\;d_2)$ and fin height (H) are numerically obtained by minimizing the pressure loss and maximizing the heat transfer rate within ranges of $d_1=0.2\sim1.5\;mm,\;d_2=0.2\sun1.5\;mm,\;and\;H=0.2\sim1.5\;mm$. The optimal values of the design variables are acquired after the fifth generation and also compared to those of a local optimization algorithm for the same geometry and conditions.

• Steel and Composite Structures
• /
• 제7권1호
• /
• pp.53-70
• /
• 2007

This paper presents a analysis of the problem of optimal design of the beams with two I-type cross section shapes. These types of beams are simply supported and subject to pure bending. The strength and stability conditions were formulated and analytically solved in the form of mathematical equations. Both global and selected types of local stability forms were taken into account. The optimization problem was defined as bicriteria. The cross section area of the beam is the first objective function, while the deflection of the beam is the second. The geometric parameters of cross section were selected as the design variables. The set of constraints includes global and local stability conditions, the strength condition, and technological and constructional requirements in the form of geometric relations. The optimization problem was formulated and solved with the help of the Pareto concept of optimality. During the numerical calculations a set of optimal compromise solutions was generated. The numerical procedures include discrete and continuous sets of the design variables. Results of numerical analysis are presented in the form of tables, cross section outlines and diagrams. Results are discussed at the end of the work. These results may be useful for designers in optimal designing of thin-walled beams, increasing information required in the decision-making procedure.