• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 1991년도 춘계공동학술대회 발표논문 및 초록집; 전북대학교, 전주; 26-27 Apr. 1991
• /
• pp.464-464
• /
• 1991

We address the problem of designing a physical CATV network with switched-star topology in which the broadband interactive service is provided. There are two types of decision variables: One is where to place conduit paths, and the other is how many cable units to be installed on each link traversed by an established conduit path. Due to the serious drawback of the conventional approach partitioning the problem into two subproblems, the unified approach handled in one setting is used here to attack the whole problem without dividing into two ones. In this paper, we present a mathematical design model and propose an efficient solution method exploiting the nice structure of it. In addition to this physical design, some results on logical network configuration have also been made. Finally, computaional experiments are conducted to illustrate the efficiency of our design approach.

• 교육시설 논문지
• /
• 제25권6호
• /
• pp.35-44
• /
• 2018

The purpose of this study is to predict more accurate construction costs and to support efficient decision making in the planning and design stages of smart education facilities. The higher the error in the projected cost, the more risk a project manager takes. If the manager can predict a more accurate construction cost in the early stages of a project, he/she can secure a decision period and support a more rational decision. During the planning and design stages, there is a limited amount of variables that can be selected for the estimating model. Moreover, since the number of completed smart schools is limited, there is little data. In this study, various artificial intelligence models were used to accurately predict the construction cost in the planning and design phase with limited variables and lack of performance data. A theoretical study on an artificial neural network and deep learning was carried out. As the artificial neural network has frequent problems of overfitting, it is found that there is a problem in practical application. In order to overcome the problem, this study suggests that the improved models of Deep Neural Network and Deep Belief Network are more effective in making accurate predictions. Deep Neural Network (DNN) and Deep Belief Network (DBN) models were constructed for the prediction of construction cost. Average Error Rate and Root Mean Square Error (RMSE) were calculated to compare the error and accuracy of those models. This study proposes a cost prediction model that can be used practically in the planning and design stages.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 2006년도 춘계공동학술대회 논문집
• /
• pp.1641-1646
• /
• 2006

This paper deals with a physical access network design problem arising from the deployment of fiber-to-the-home passive optical networks (FTTH-PON). The problem seeks to find an optimal location and capacity assignment of splitters in the network such that the total cost of fiber and splitter is minimized, while satisfying all the demand. We propose tabu search approach for finding a good feasible solution.

• 한국경영과학회지
• /
• 제22권3호
• /
• pp.23-36
• /
• 1997

In the design of telecommunication network, the network configuration using hubbing topology is useful for designing and managing the network efficiently : i. e. all of central offices (COs) are grouped into clusters. Each cluster has one hub consisting of large-scale transmission facilities like digital cross-connect systems and ATMs. In clustering process, the community of interest and geographical factor should be considered. However, there exists a tradeoff between two factors. One is to minimize total link costs for geographical factor and the other is to maximize the total intra-cluster traffics for community of interest. Hence, this can be solved by multiobjective linear programming techniques. In this paper, the problem under considerations is formulated as two p-median subproblems taking into considerations total costs and total intra-traffics, respectively. Then we propose the algorithm to solve the problem based on the concept of cost-traffic tradeoff. The algorithm enables to identify efficient cost-traffic tradeoff pairs. An illustration is also presented.

• 한국통신학회논문지
• /
• 제25권1B호
• /
• pp.15-30
• /
• 2000

광전송망은 망의 장애에 미리 대비할 수 있도록 신뢰성과 생존도를 고려하여 설계하여야 한다. 동기식 다중화의 국제표준인 SDH(Synchronous Digital Hierarchy) 방식의 광전송망은 장애 발생시 이를 자동적으로 복구할 수 있도록 생존도를 고려한 여러 가지 망 재구성 기법들을 제공한다. 그 중 SHR(Self-Healing Ring)은 링의 형태로 망을 구성한 시스템으로 뛰어난 생존도와 경제성으로 통신사업자들의 기간통신망 구조로 활발히 채택되고 있다. 이 때, 링들이 설치되는 지역적 범위가 넓어지고 수요가 증가되면, 다수의 링들이 중첩되어 상호연결되는 다중링(Multi-ring) 구조로 발전하게 된다. 본 연구에서는 수요의 분할처리를 허용하는 BSHR(Bidirectional SHR)들이 연접한 다중링 설계 문제를 다룬다. 이 문제는 망구축용량을 최소로하는 관점에서 생존도가 보장되는 부하 최적화 문제가 되며, 혼합정수계획법에 의한 정식화가 가능하다. 그러나, 현실문제에서는 망구축용량의 최소화 뿐만아니라 노드가 수요로 다계위 수요가 주어지며 중계노드에서의 다중화 번들링도 같이 고려되어야 하므로 수리모형으로는 해결할 수 없는 복잡한 문제가 된다. 따라서 이 叩걋\ulcorner고려사항들을 반영한 최적근사해를 실시간내에 구할 수 있는 휴리스틱 알고리즘을 개발하게 되었다. 사례연구에서는 휴리스틱 알고리즘을 적용한 실제 망설계 문제를 설명하였고, 망구성 방법에 따른 차이와 다중화 번들링 여부로 인한 실험 결과를 비교하였다.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회/대한산업공학회 2005년도 춘계공동학술대회 발표논문
• /
• pp.59-71
• /
• 2005

In this paper we address a design problem for hub and spoke transportation networks and then consider a capacitated hub locations problem with direct shipment (CHLPwD). We determine the location of hubs, the allocation of nodes to hubs, and direct shipment paths in the network, with the objective of minimizing the total cost in the network. An evolutionary algorithm is developed here to solve the CHLPwD. To do this, we propose the representation and the genetic operators suitable for the problem and adopt a heuristic method for the allocation of nodes to hubs. To enhance the search capability, problem-specific information is used in our evolutionary algorithm. The proposed algorithm is compared with the heuristic method in terms of solution quality and computation time. The experimental results show that our algorithm can provide better solutions than the heuristic.

• 한국경영과학회지
• /
• 제19권1호
• /
• pp.69-83
• /
• 1994

This paper deals with a capacitated ring-star network design problem (CRSNDP) with node capacity constraints. The CRSNDP is formulated as a mixed 0-1 integer problem, and a 2-phase heuristic solution procedure, ADD & VAM and RING, is developed, in which the CRSNDP is decomposed into two subproblems : the capacitated facility location problem (CFLP) and the traveling sales man problem (TSP). To solve the CFLP in phase I the ADD & VAM procedure selects hub nodes and their appropriate capacity from a candidate set and then assigns them user nodes under node capacity constraints. In phase II the RING procedure solves the TSP to interconnect the selected hubs to form a ring. Finally a solution of the CRSNDP can be achieved through combining two solution of phase I & II, thus a final design of the capacitated ring-star network is determined. The analysis of computational results on various random problems has shown that the 2-phase heuristic procedure produces a solution very fast even with large-scale problems.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권8호
• /
• pp.3701-3727
• /
• 2016

Recently Service Function Chaining (SFC) is promising to innovate the network service mode in modern networks. However, a feasible implementation of SFC is still difficult due to the need to achieve functional equivalence with traditional modes without sacrificing performance or increasing network complexity. In this paper, we present a configurable network service chaining (CNSC) mechanism to provide services for network traffics in a flexible and optimal way. Firstly, we formulate the problem of network service chaining and design an effective service chain construction framework based on integrating software-defined networking (SDN) with network functions virtualization (NFV). Then, we model the service path computation problem as an integer liner optimization problem and propose an algorithm named SPCM to cooperatively combine service function instances with a network utility maximum policy. In the procedure of SPCM, we achieve the service node mapping by defining a service capacity matrix for substrate nodes, and work out the optimal link mapping policies with segment routing. Finally, the simulation results indicate that the average request acceptance ratio and resources utilization ratio can reach above 85% and 75% by our SPCM algorithm, respectively. Upon the prototype system, it is demonstrated that CNSC outperforms other approaches and can provide flexible and scalable network services.

• 한국정보통신학회논문지
• /
• 제21권1호
• /
• pp.144-150
• /
• 2017

최근 데이터의 지능적 처리 및 정확도 향상을 위해 딥러닝 기술이 응용되고 있다. 이 기술은 다층의 데이터 처리 레이어들로 구성된 계산 모델을 통해 이루어지는데, 이 모델은 여러 수준의 추상화를 거쳐 데이터의 표현을 학습한다. 딥러닝의 한 부류인 컨볼루션 신경망은 인간 행동 추정, 얼굴 인식, 이미지 분류, 음성 인식 같은 연구 분야에서 많이 활용되고 있다. 이미지 분류에 좋은 성능을 보여주는 컨볼루션 신경망은 깊은 학습망과 많은 부류를 이용하면 효과적으로 분류율을 높일수 있지만, 적은 부류의 데이터를 사용할 경우, 과적합 문제가 발생할 확률이 높아진다. 따라서 본 논문에서는 컨볼루션 신경망기반의 소부류의 분류을 위한 학습망을 제작하여 자체적으로 구축한 이미지 DB를 학습시키고, 객체를 분류하는 연구를 실험 하였으며, 1000개의 부류를 분류하기 위해 제작된 기존 공개된 망들과 비교 실험을 통해 기존 망보다 평균 7.06%이상의 상승된 분류율을 보여주었다.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2021년도 봄 학술논문 발표대회
• /
• pp.223-224
• /
• 2021

Recently, the demand for high fluidity concrete has been increased due to skyscrapers. However, it has its own limits. First of all, high fluidity concrete has large variation and through trial & error it costs lots of money and time. Neural network model has repetitive learning process which can solve the problem while training the data. Therefore, the purpose of this study is to predict optimum mix design of 40MPa, 60MPa high fluidity concrete by using neural network model and verifying compressive strength by applying real data. As a result, comparing collective data and predicted compressive strength data using MATLAB, 40MPa mix design error rate was 1.2%~1.6% and 60MPa mix design error rate was 2%~3%. Overall 40MPa mix design error rate was less than 60MPa mix design error rate.