폐기물 입지 및 경로 문제를 위한 협력적 공진화 알고리즘과 이웃해 탐색의 결합해법

한용호(Yongho Han) 한국SCM학회 2015 한국SCM학회지 Vol.15 No.2

This paper addresses with a waste location-routing problem with multiple capacitated depots and one uncapacitated vehicle per depots and formulates the problem into a binary integer programming model in order to minimize the total cost. We suggest a cooperative coevolutionary algorithm (CCEA) combined with a local search to make location and routing decisions simultaneously. First, a solution is represented using two populations of chromosomes. One population represents an assignment configuration that gives the set of open depots. The other population is a permutation fixing the rank of a customer on a given route. And fitness function, fitness evaluation pattern, and genetic operators are defined. Then, local search (LS) is adopted to generate a local solution by exploring the neighborhood of an initial solution given by CCEA. We use four neighborhood structures including insertion move and swap move in LS. Finally, we combine the CCEA with the LS and suggest a CCEA-based hybrid algorithm. A computational study is carried out to compare our algorithm with other algorithms. Numerical experiments show that our CCEA-based hybrid algorithms outperform GA-based algorithms and CCEA-based hybrid algorithms outperform pure CCEAs.

BPMS 기반 항만물류 가상기업의 구현

한용호(Yongho Han) 한국경영과학회 2008 經營 科學 Vol.25 No.3

First we provide an overview of how virtual enterprise works, including the lifecycle of a virtual enterprise, expected benefits from a virtual enterprise, and various types of collaborative networks. Next we focus on a problem encountered in the port logistics industry and suggest that implementation of a virtual enterprise can be a good solution to the problem. Then we suggest a prototype of the virtual enterprise in the port logistics industry and a freight transportation process as a business process in the prototype. And then we introduce major functions of the business process management system (BPMS) which, in general, can be regarded as a tool to create the virtual enterprise. Finally by adopting a specific BPMS, we illustrate how to design and implement the freight transportation process in the prototype.

다중 집단 기반의 협력적 공진화 알고리즘을 사용한 폐쇄루프 공급사슬 네트워크의 설계

한용호(Yongho Han) 한국SCM학회 2013 한국SCM학회지 Vol.13 No.1

Literature survey shows that only a few papers deal with comprehensive supply chain models with both forward and reverse flows. The goal of this study is to propose a comprehensive closed-loop model integrating forward and reverse logistics for the supply chain network design and to design a cooperative coevolutionary algorithm(CCEA) as a heuristic approach to get a good approximate solution. Mathematical model is formulated first, and then CCEA is designed as follows; First, the problem is broken down into eight subproblems. For each subproblem, a population of chromosomes is created, a chromosome is encoded using a permutation of integers, and a decoding method is suggested to get a partial solution. As genetic operators, binary tournament selection with elitist strategy, order crossover, and swap mutation are applied for evolving each of the subpopulations. To find feasible solutions satisfying a nonlinear constraint, a penalty method is adopted. The evaluation of a partial solution in a subpopulations is done by composing a complete solution with the best partial solutions from all the other subpopulations. Experimental results show that our CCEA in almost every case outperforms GA in terms of both the quality of solution obtained and the potential for solution improvement.

5개 모집단의 협력적 공진화 알고리즘을 사용한 다단계 공급사슬 네트워크 설계

한용호(Yongho Han) 한국SCM학회 2012 한국SCM학회지 Vol.12 No.2

This paper suggests an efficient cooperative coevolutionary algorithm(CCEA) consisting of five populations of chromosomes to solve the multi-stage supply chain network design problem. The mathematical model is described. The problem is broken down into five subproblems. For each subproblem, a population of chromosomes is created and each chromosome is represented as a permutation of integers. For each chromosome in a population, a decoding method is suggested to get a partial solution to the corresponding subproblem. Then the procedure of combining all chromosomes from each of the five populations to achieve a feasible solution is suggested. Both the method of deciding the collaborator of a population and the method of combining it with other collaborators is designed to evaluate the suitability of a given chromosome. An experimental study is carried out. The results show that our CCEA generates better performance than the previous CCEA for larger problems and that the difference in performance between two algorithms tends to get larger.

협력적 공진화 알고리즘과 퍼지 집합 이론의 병행 사용을 통한 친환경 공급사슬 네트워크의 설계

한용호(Yongho Han) 한국SCM학회 2014 한국SCM학회지 Vol.14 No.1

Recently green supply chain management has drawn the attention of researchers. Previous studies to design a generalized closed-loop supply chain(CLSC) network are very limited and do not consider the qualitative factors of the facilities to be open. In this paper, a new generalized supply chain network design problem is considered that includes facilities like suppliers, plants, distribution centers, dismantlers and customers. We propose a closed-loop supply chain model considering not only the quantitative costs but also the qualitative factors of facilities. The model has two objective functions. The first is to minimize the total cost, and the second is to maximize the total weight of all facilities to be open. Our approach is composed of two phases. In the first phase, a framework for facilities selection criteria is proposed based on fuzzy set theory. The output of this stage is the weight of each facility. In the second phase, a multi-objective mixed integer nonlinear programming model is formulated to choose the facilities to be open for each type and to optimize the amount of parts and products shipped between facilities in the CLSC network. A cooperative coevolutionary algorithm is then applied to get an approximate solution. A numerical example is presented and a sensitivity analysis is performed. The results show that both the proposed model and the solution approach are validated

역물류의 특성을 지닌 이종 차량군 친환경 경로 문제에 대한 진화 알고리즘

한용호(Yongho Han) 한국SCM학회 2017 한국SCM학회지 Vol.17 No.2

This paper addresses a heterogeneous fleet vehicle routing problem (VRP) having simultaneous pickup and delivery, which plays an important role in the green logistics. The decision making is on both the fleet composition and the vehicle routings. We formulate the problem into a mathematical model and propose an evolutionary algorithm (EA) hybridized with the nearest neighborhood algorithm (NNA). The EA has two populations. The first one is employed for representing vehicles to be used to travel to customers and the second one is for representing each route which is assigned to each vehicle. Permutation representation is used for encoding individuals in both of two populations. After we obtain a solution by combining and decoding the information of each individual from both populations, the NNA is applied to get a better route for each vehicle. Two populations evolve separately in each generation. Truncation selection, order crossover, and swap mutation operation are commonly adopted for each of both populations. The developed EA can be implemented into one of two approach, either GA-based one or CCEA-based one according to the pattern of evaluating individuals from each of both populations. A computational study is carried out to compare GA-based approach with CCEA-based one. Numerical experiments show that our GA-based approach tends to outperform CCEA-based one in terms of the quality of solutions generated.

크로스도킹 환경에서의 차량경로계획을 위한 진화 알고리즘

한용호(Yongho Han) 한국SCM학회 2018 한국SCM학회지 Vol.18 No.2

Cross-docking is one of the effective ways applied in supply chain management in order to minimize total transportation cost while satisfying the customer demands. This paper addresses the transportation problem of cross-docking network design where products are transferred from suppliers to customers through distribution center without storing them for long. The objective of the problem is to find vehicle routings that minimize total transportation cost with heterogeneous vehicles of different capacities. Evolutionary algorithms (EA) are designed to solve the problem. The EA needs three types of chromosomes. The first is employed for representing the assignment of vehicles to travel to suppliers or customers, the second for representing suppliers to be assigned to a vehicle, and the third for representing customers to be assigned to a vehicle. Permutation representation is used for encoding the individuals of all the three types. Depending on how to evaluate the individuals of different types, EAs can be implemented in different forms: genetic algorithm (GA), cooperative coevolutionary algorithm with two populations (CCEA-2p), or CCEA with three populations (CCEA-3P). A computational study is carried out to compare the performances of these approaches. Numerical experiments show that both CCEA-2P and CCEA-3P tend to outperform GA in terms of the quality of the best solutions generated, while the performance difference between CCEA-2P and CCEA-3P looks insignificant.

순열 표현 기반의 협력적 공진화 알고리즘을 사용한 다단계 공급사슬 네트워크의 설계

한용호(Yongho Han) 한국경영과학회 2012 經營 科學 Vol.29 No.2

This paper addresses a network design problem in a supply chain system that involves locating both plants and distribution centers, and determining the best strategy for distributing products from the suppliers to the plants, from the plants to the distribution centers and from the distribution centers to the customers. This paper suggests a cooperative coevolutionary algorithm (CCEA) approach to solve the model. First, the problem is decomposed into three subproblems for each of which the chromosome population is created correspondingly. Each chromosome in each population is represented as a permutation denoting the priority. Then an algorithm generating a solution from the combined set of chromosomes from each population is suggested. Also an algorithm evaluating the performance of a solution is suggested. An experimental study is carried out. The results show that our CCEA tends to generate better solutions than the previous CCEA as the problem size gets larger and that the permutation representation for chromosome used here is better than other representation.

공급사슬 네트워크 설계를 위한 협력적 공진화 알고리즘에서 집단들간 상호작용방식에 관한 연구

한용호(Yongho Han) 한국경영과학회 2014 經營 科學 Vol.31 No.3

Cooperative coevolutionary algorithm (CCEA) has proven to be a very powerful means of solving optimization problems through problem decomposition. CCEA implies the use of several populations, each population having the aim of finding a partial solution for a component of the considered problem. Populations evolve separately and they interact only when individuals are evaluated. Interactions are made to obtain complete solutions by combining partial solutions, or collaborators, from each of the populations. In this respect, we can think of various interaction modes. The goal of this research is to develop a CCEA for a supply chain network design (SCND) problem and identify which interaction mode gives the best performance for this problem. We present general design principle of CCEA for the SCND problem, which require several co-evolving populations. We classify these populations into two groups and classify the collaborator selection scheme into two types, the random-based one and the best fitness-based one. By combining both two groups of population and two types of collaborator selection schemes, we consider four possible interaction modes. We also consider two modes of updating populations, the sequential mode and the parallel mode. Therefore, by combining both four possible interaction modes and two modes of updating populations, we investigate seven possible solution algorithms. Experiments for each of these solution algorithms are conducted on a few test problems. The results show that the mode of the best fitness-based collaborator applied to both groups of populations combined with the sequential update mode outperforms the other modes for all the test problems.

협력적 공진화 알고리즘에 기반한 다단계 공급사슬 네트워크의 설계

한용호(Yongho Han) 한국SCM학회 2011 한국SCM학회지 Vol.11 No.2

We consider a network design problem in a four tier supply chain consisting of suppliers, plants, distribution centers and customers. We suggest a cooperative coevolutionary algorithm(CCEA) to solve the problem. First, the problem is decomposed into three subproblems for each of which the chromosome population is created correspondingly. Each chromosome in a population is represented as an integer denoting a supply node. Then an algorithm generating a solution from a chromosome representation is suggested. Also an algorithm evaluating the performance of a solution is suggested. Finally we set the operator of selection, crossover, and mutation. An experimental study is carried out to compare the performance of the CCEA with that of a genetic algorithm. The results show that the CCEA tends to generate better solutions than genetic algorithms.