Improvement for ranking method-based selection methods of government support R & D project

Young-Soo Park,Chang-Wook Kang,Sun-Geun Lee 한국산업경영시스템학회 2020 한국산업경영시스템학회 학술대회 Vol.2020 No.춘추계

Recognition that R&D (Research and Development) project is important for national competitiveness and sustainable economic growth has been extended all over the world, and the government financial support for R&D projects has increased. Therefore, government needs to make a decision about whether it invests in any project or not, and this decision-making complies with evaluation of the relevant experts. In this study, we improve project selection methods and procedures based on outranking methods which have mainly focused on general project environmental analysis in order to apply to the government support R&D project selection, and suggest R&D project selection methods and procedures that meet the purpose of government support. Therefore, it demonstrates that improved methods organize more profitable project groups to the purpose of the selecting government support R&D project than existing methods. Also, it illustrates that it is possible for decision makers to analyze variously by changing composition of selecting project group with variation of α.

LCC 평가를 통한 지붕방수공법선정에 관한 연구

최오영,김태희,김광희 한국건축시공학회 2008 한국건축시공학회지 Vol.8 No.5

The purpose of this study is to propose the decision making technique in roof waterproofing method at the early construction stage. Selecting the suitable construction method is difficult because of the complex interrelationships between many factors of influencing the construction method selection. This study presents an example of selecting suitable method by analyzing LCC (Life Cycle Cost) in roof waterproofing work. In this study, roof waterproofing method is analyzed by LCC(Life Cycle Cost) which is consists of the initial costs, running costs, and removal costs. Sheet waterproofing, membrane waterproofing and asphalt waterproofing costs are compared to select the most economic method. The result of this study revealed that considering LCC is useful in selecting the proper method in the construction work.

A Novel Machine Selection Method Combining Group Eigenvalue Method with TOPSIS Method

Tonghua Yang,Shenghua Xu,Neal N. Xiong 보안공학연구지원센터 2016 International Journal of Future Generation Communi Vol.9 No.6

Machine selection is an important step in the process of manufacturing. The selection process needs to consider some selection attributes simultaneously from a set of candidate machines. The attribute weights are important for the result of machine selection problem, but the most used AHP method has the shortcoming. Thus this paper will develop a new weighting method based on the concept of group eigenvalue method. Then the new machine selection method is proposed by combining group eigenvalue method with TOPSIS method. Two practical examples demonstrate that the proposed method has more effectiveness and feasibility.

Performance Evaluation of Feature Selection Methods on Large Dimensional Databases

Y. Leela Sandhya Rani,V. Sucharita,Debnath Bhattacharyya,Hye-Jin Kim 보안공학연구지원센터 2016 International Journal of Database Theory and Appli Vol.9 No.9

Data mining retrieves knowledge information from larger amounts of data. Clustering is an assemble of similar objects in to one class and dissimilar objects in to another class. When designing clustering ensemble on large dimensional data space, both time and space requirements for processing may be overinflated. This tends to impose feature selection methods to remove redundant features and handle the noise data. There are filter, wrapper and hybrid methods in feature selection. This paper shows a tour on types of feature selection techniques and numbers of experiments are conducted to compare feature selection techniques using different datasets with R tool, which gives better technique for clustering ensemble design.

선택적 부족분 공급방식에 따른 댐 하류하천의 유황 변화 분석

최영제,박문형 한국수자원학회 2022 한국수자원학회논문집 Vol.55 No.12

Currently, South Korea implements water resources management policies focusing on integrated water quantity, quality and hydro-ecology management. In particular, rehabilitation of natural rivers has become a major issue. As for reservoir operation during non-flood season, efforts have been made continuously to apply the Deficit Supply Method that can maximize water supply to address droughts and increase in water demand. When Deficit Supply Method is applied, the water supply capacity of reservoir can be maximized. However, downstream water flow would remain constant. In consideration that a natural stream, a long-time-created hydro-ecology, can be significantly influenced by flow variability, the Deficit Supply Method-based reservoir operation can generate effective water supply. Still, it may trigger adverse effects from the aspects of natural rehabilitation and hydro-ecology recovery. The main objective of this study is to analyze impacts on downstream flow duration through reservoir simulation by comparing the Firm Supply Method, the Deficit Supply Method and the Selective Deficit Supply Method, and examining each method’s effects on reservoir operation. This study found that the Firm Supply Method could maintain water flow variability, but could not maximize water supply capacity. When the Deficit Supply Method was applied, water supply capacity could be increased while remaining vulnerable regarding water flow variability, as a difference between average flow and low flow was negligible at downstream. In comparison, the Selective Deficit Supply Method was found to sustain time-based reliability at 95% or higher, whereas downstream flow duration could be maintained at a level similar to the level generated by the Firm Supply Method. 최근 우리나라의 물 관련 정책은 수량-수질-수생태 통합관리 방향으로 진행되고 있으며, 특히 하천의 자연성 회복이 주요한 이슈가 되고 있다. 이수기 댐 운영에 있어서는 가뭄, 물 수요 증가 등으로 용수공급 효과를 극대화시킬 수 있는 부족분 공급방식을 적용하고자하는 시도들이 이어지고 있다. 댐 운영에 부족분 공급방식을 적용하면 댐의 용수공급능력을 극대화 시킬 수는 있지만 하류 하천의 유량이 일정하게 유지된다는 특징이 있다. 자연하천은 오랜 시간동안 형성된 하나의 생태계로 유량의 변동성에 큰 영향을 받는다. 결국 부족분 공급방식을 적용한 댐 운영은 수량 관리에서는 효과적이지만 하천의 자연성 회복 및 수생태 측면에서는 부정적 영향을 미칠 수있다. 본 연구에서는 저수지 모의를 통해 보장량 공급방식, 부족분 공급방식, 선택적 부족분 공급방식 등의 댐 운영이 하류 하천 유황에 미치는 영향을 분석하고, 각 운영방식의 적용 효과에 대해 분석하고자 하였다. 그 결과 보장량 공급방식을 적용하면 하천의 유량 변동성은 크게 유지할 수 있으나 댐의 용수공급능력은 크지 않은 것으로 나타났다. 부족분 공급방식을 활용하면 용수공급능력을 증대시킬 수는 있으나 하류의 평수량과 갈수량의 차이가 매우 작아 유량의 변동성 측면에서는 매우 취약한 것으로 확인되었다. 선택적 부족분 공급방식을 적용할 경우 기간신뢰도를 95% 이상으로 유지하며, 하류 하천의 유황은 보장량 공급방식을 적용할 때와 유사하게 유지할 수 있는 것으로 분석되었다.

노인장기요양보험 대상자 선정도구의 타당성 검증: 독일과 일본의 장기요양대상자 선정도구를 기준으로

이윤경 ( Yun Kyung Lee ) 한국사회복지정책학회 2015 사회복지정책 Vol.42 No.3

노인장기요양보험에서 급여 대상자 선정은 대상자에게 서비스 이용권을 부여하는 것으로, 제도의 신뢰 도를 결정하는 중요한 기준이다. 본 연구는 독일과 일본의 장기요양보험 선정도구를 기준으로 현재 한국에서 사용하고 있는 노인장기요양 대상자 선정도구의 타당성을 검증하고, 또한 3국의 대상자 보장성 규모를 평가하기 위한 연구이다. 이를 위해 한국, 독일, 일본의 선정도구를 조사대상 551명에게 적용하여 국가별 장기요양필요도 산정 알고리즘을 활용하여 각각의 점수를 산정하였다. 분석결과 한국, 독일, 일본의 요양필요도 수준은 매우 높은 상관관계를 갖고 있어 한국의 장기요양대상자 선정도구는 독일과 일본의 도구를 기준으로 할 때 타당도 높은 도구임이 입증되었다. 또한 국가별 장기요양보험의 보장성 수준은 독일이 가장 높으며, 일본, 한국의 순으로 나타났다. 한국은 독일의 약 85%이며, 일본의 약 93%수준으로 나타났다. 즉, 한국의 장기요양보험의 대상자 보장 수준은 최근 계속적인 보장성 확대를 통해 향상되었음을 알 수 있다. 본 연구결과를 통해 향후 한국은 노인장기요양보험 대상자 선정도구의 계속적 신뢰도 유지를 위해 제도도입이후 변화된 요양서비스 현황이 반영된 도구 개편이 이루어질 수 있도록 노력해야 할 것이다. 또한 노인장기요양보험은 향후 대상자의 보장성 확대보다는 서비스의 질 향상, 급여 내용의 전문화, 급여 범위의 확대 등의 질적 보장성 확대를 위한 노력이 요구되어진다. The client selection process for the long-term care insurance is to provide a right of using service to them. For efficient policy management, accordingly, it is very important to select them through understanding their exact needs. In this study, validity for the long-term care client selection method being applied in Korea was verified. And, validity for Korea`s selection method was evaluated based on Germany and Japan`s selection method, with applying Korea`s method, newly developed Germany`s selection method and japan`s method to the same client group. As a result, it was shown that the Korea`s selection method had high validity, but the client diversity is low, when evaluated based on selection method of Germany and japan. To supplement the policy as shown in this result, an alternative to achieve appropriate client selection method was proposed, and also an alternative to overcome the limitations of the present client selection method was suggested.

Risk Response Analysis Model for Construction Method Using the Forced-Decision Method and Binary Weighting Analysis

Lee, Jongsik,Chun, Jaeyoul Architectural Institute of Japan 2009 JOURNAL OF ASIAN ARCHITECTURE AND BUILDING ENGINEE Vol.8 No.1

<P>Although the selection of a construction method for each type of work is an important factor in determining the quality of a building, the construction duration, costs, and methods are currently being selected according to the subjective judgment of the person in charge, without sufficient consideration of the characteristics of the work type. As a response to this issue, this study proposes a process model to support decision making when selecting the most suitable construction method for major types of construction work. The study used the risk response level model, which connects the conditions of the order and the site, the constructability of the type of work within the site, a review of the economic efficiency of the work site, the forced-decision method, and binary weighting analysis in the construction planning phase. As a result of the application of this model to the construction method selection process for the construction of soil-retaining walls, the factors to be preferably considered in the field cases were analyzed in the order of the environmental, ground, design, construction, structural, and site conditions. Further, the risk response level of each construction method was calculated via risk response level analysis, subject to four applicable construction methods.</P>

A Study on Unbiased Methods in Constructing Classification Trees

Lee, Yoon-Mo,Song, Moon Sup The Korean Statistical Society 2002 Communications for statistical applications and me Vol.9 No.3

we propose two methods which separate the variable selection step and the split-point selection step. We call these two algorithms as CHITES method and F&CHITES method. They adapted some of the best characteristics of CART, CHAID, and QUEST. In the first step the variable, which is most significant to predict the target class values, is selected. In the second step, the exhaustive search method is applied to find the splitting point based on the selected variable in the first step. We compared the proposed methods, CART, and QUEST in terms of variable selection bias and power, error rates, and training times. The proposed methods are not only unbiased in the null case, but also powerful for selecting correct variables in non-null cases.

A component mode selection method based on a consistent perturbation expansion of interface displacement

Kim, Soo Min,Kim, Jin-Gyun,Park, K.C.,Chae, Soo-Won Elsevier 2018 Computer methods in applied mechanics and engineer Vol.330 No.-

<P><B>Abstract</B></P> <P>A mode selection method is presented for the reduced-order modeling (ROM) of structural systems in conjunction with the Craig–Bampton component mode synthesis technique. The proposed method is derived by using a consistent expansion of the interface displacement in terms of a frequency-dependent small parameter as applied to a Craig–Bampton-like ROM formulation. It is found that this procedure yields a coupling mechanism of the modes of the full model to those of substructures. The present mode selection method employs this coupling mechanism as an indicator, labeled as the CMS <SUB> σ </SUB> method, for the substructural modal contributions to the full model. The performance of the proposed method is demonstrated by various numerical examples and compared favorably with existing method such as the CMS <SUB> χ </SUB> method.</P> <P><B>Highlights</B></P> <P> <UL> <LI> We propose a coupling-matrix based mode selection scheme of the CMS method. </LI> <LI> The proposed method is derived by using the moment-matching approach with perturbation analysis. </LI> <LI> Its excellent performances are demonstrated by numerical examples. </LI> </UL> </P>

고차원 선형모형에서 벌점화우도 방법을 이용한 변수 선택방법 연구

주아림,전수영 한국자료분석학회 2015 Journal of the Korean Data Analysis Society Vol.17 No.5

If the sample size is greater than the number of covariates or parameters, typically forward selection, backward elimination and stepwise selection methods are used among several variable selection methods. However, in the case of high-dimensional linear models, it is difficult to use such methods. In order to overcome this problem, penalized likelihood, also called regularization or shrinkage, methods may be used in high-dimensional linear models. This paper considers several variable selection methods used in high-dimensional linear models, and investigates the applicability of penalized likelihood methods. We briefly review ridge, LASSO, LARS, elastic net, adaptive LASSO, and apply these methods to the real data, prostate and Korea economic index data. Numerical results indicate that the best method was the LARS algorithm among five penalized likelihood methods based on RSS, AIC and BIC. 선형회귀모형에서 표본 크기 이 공변량의 수 보다 큰 경우 여러 가지 변수 선택방법 중 일반적으로 전진적 선택방법, 후진적 제거방법, 단계적 선택방법 등이 사용된다. 하지만, 가 에 비해 매우 큰 경우인 고차원문제를 가지는 선형모형에서는 이러한 방법들을 사용하는 데에 어려움이 있다. 이러한 변수선택의 어려운 점을 극복하기 위해 유의하지 않은 변수들을 제거하는 다양한 regularization 또는 shrinkage 방법이라고도 불리는 벌점화우도(penalized likelihood) 방법들이 고차원 선형모형에서 제안되었다. 본 논문은 고차원 선형모형에서 사용이 용이한 변수 선택방법에 대한 연구로, 여러 가지 벌점화우도 방법들에 대해 알아보고 그 응용성을 알아본다. 그 중에서 ridge, LASSO, LARS, elastic net, adaptive LASSO 등 다섯 가지 벌점화우도 방법의 특징을 간단히 살펴보고, 실제 자료인 전립선(prostate) 자료와 한국 경제 수치자료에 적용해 본다. RSS, AIC, BIC를 평가기준으로 다섯 가지 벌점화우도 방법의 실 자료 분석 결과 고차원 선형모형에서 가장 적합한 방법은 LARS 알고리즘이었다.