• 정보관리학회지
• /
• 제24권3호
• /
• pp.43-65
• /
• 2007

심사자 자동추천시스템은 심사 대상에 대한 포괄성, 전문성, 공정성, 타당성을 확보할 수 있도록 설계되어야 한다. 이를 위해 본 연구는 다면적인 학문분야분류표의 각 범주 간 연관성을 자동으로 산출할 수 있는 확률적 온톨로지를 적용하여 포괄적으로 심사자 추천 범위를 넓히고 전문성을 반영한 심사자 랭킹을 가능하도록 한다. 또한 연구자 간의 멘터, 공저역, 공동연구를 포함하는 연구자 네트워크를 구축하고 이를 심사자 배제 규칙으로 활용함으로써 공정한 심사자 추천이 이루어질 수 있도록 한다. 아울러, 전문가들을 통해 상기 방법론과 패널 결과를 검증 받아 타당성 있는 시스템이 갖추어야 할 방향을 제시한다.

• 한국군사과학기술학회지
• /
• 제15권1호
• /
• pp.93-100
• /
• 2012

In this paper, the deterministic optimal design for the tail plane made of composite materials is conducted under the deterministic loading condition and compared with that of the metallic materials. Next, the reliability analysis with five random variables such as loading and material properties of unidirectional prepreg is conducted to examine the probability of failure for the deterministic optimal design results. The MATLAB programing is used for reliability analysis combined with FEA S/W(COMSOL) for structural analysis. The laminated composite is assumed to the equivalent orthotropic material using classical laminated plate theory. The response surface methodology and importance sampling technique are adopted to reduce computational cost with satisfying the accuracy in reliability analysis. As a result, structural weight of composite materials is lighter than that of metals in deterministic optimal design. However, the probability of failure for the deterministic optimal design of the tail plane structures is too high to be neglected. The sensitivity of each variable is also estimated using probabilistic sensitivity analysis to figure out which variables are sensitive to failure. The computational cost is considerably reduced when response surface methodology and importance sampling technique are used. The study of the computationally inexpensive method for reliability-based design optimization will be necessary in further work.

• 한국물환경학회지
• /
• 제22권6호
• /
• pp.1068-1074
• /
• 2006

For receiving water quality protection a control systems of urban drainage for CSOs reduction is needed. Examples in combined sewer systems include downstream storage facilities that detain runoff during periods of high flow and allow the detained water to be conveyed by an interceptor sewer to a centralized treatment plant during periods of low flow. The design of such facilities as storm-water detention storage is highly dependant on the temporal variability of storage capacity available as well as the infiltration capacity of soil and recovery of depression storage. For the continuous long-term analysis of urban drainage system this study used analytical probabilistic model based on derived probability distribution theory. As an alternative to the modeling of urban drainage system for planning or screening level analysis of runoff control alternatives, this model has evolved that offers much ease and flexibility in terms of computation while considering long-term meteorology. This study presented rainfall and runoff characteristics of the subject area using analytical probabilistic model. Runoff characteristics manifested the unique characteristics of the subject area with the infiltration capacity of soil and recovery of depression storage and was examined appropriately by sensitivity analysis. This study presented the average annual CSOs, number of CSOs and event mean CSOs for the decision of storage volume.

• 한국물환경학회지
• /
• 제23권4호
• /
• pp.490-497
• /
• 2007

A Combined sewer overflows (CSOs) are themselves a significant source of water pollution. Therefore, the control of urban drainage for CSOs reduction and receiving water quality protection is needed. Examples in combined sewer systems include downstream storage facilities that detain runoff during periods of high flow and allow the detained water to be conveyed by an interceptor sewer to a centralized treatment plant during periods of low flow. The design of such facilities as stormwater detention storage is highly dependant on the temporal variability of storage capacity available (which is influenced by the duration of interevent dry periods) as well as the infiltration capacity of soil and recovery of depression storage. As a result, a continuous approach is required to adequately size such facilities. This study for the continuous long-term analysis of urban drainage system used analytical probabilistic model based on derived probability distribution theory. As an alternative to the modeling of urban drainage system for planning or screening level analysis of runoff control alternatives, this model have evolved that offer much ease and flexibility in terms of computation while considering long-term meteorology. This study presented rainfall and runoff characteristics of the subject area using analytical probabilistic model. This study presented the average annual COSs and number of COSs when the interceptor capacity is in the range $3{\times}DWF$ (dry weather flow). Also, calculated the average annual mass of pollutant lost in CSOs using Event Mean Concentration. Finally, this study presented a decision of storage volume for CSOs reduction and water quality protection.

• 한국산업정보학회논문지
• /
• 제23권6호
• /
• pp.79-94
• /
• 2018

마케팅 분야에서는 제품품질, 고객만족, 고객추천을 바탕으로 구매행동과의 영향 유무 및 상관관계를 통계적 Regression 방법으로 가설 검증하는 것을 주요한 연구 대상으로 하고 있다. 또한 최근에는 ASCI와 같은 고객만족지수 혹은 라이켈트의 NPS와 같은 고객추천지수를 바탕으로 실제 기업성과와 연관되는 시장 지분에 어떠한 영향을 미치는 지에 대한 통계적 분석 연구도 활발히 이루어지고 있다. 본 연구에서는 실제 고객이 매장을 방문하여, 과거 고객카드에 명품을 구매하던 구매하지 않던 간에 만족/불만족을 표시한 체인 및 고객 추천의향을 검토하여 Hidden Markov Model을 이용한 고객의 최상의 구매패턴을 분석하는 확률적 기법에 대하여 연구하는 것을 목적으로 하고 있다. 이를 바탕으로 고객만족 -> 고객추천의향 -> 고객추천행동->구매 및 재구매 체인에 대응하는 실제 소비자의 구매패턴을 고객만족과 NPS(순추천지수) 및 여러 수리통계적 이론-Hidden Markov Model, Bayesian Inference, Maximum Likelihood Estimation을 이용하여 확률적 추적 메카니즘을 구현하는 것을 목표로 한다. 제시된 목표는 인공지능을 구현하는 이론과 알고리듬을 사용하여 달성되었기에 이론적 추적 메카니즘을 여러 인공지능망 -DNN, CNN, GAN등을 사용하여 기업에서 사용할 수 있는 고객의 구매패턴 앱으로 발전시키는 것을 후속연구에서 기대한다.

• Journal of Mechanical Science and Technology
• /
• 제20권12호
• /
• pp.2124-2135
• /
• 2006

In this paper, the reliability estimation of pipelines is performed by employing the probabilistic method, which accounts for the uncertainties in the load and resistance parameters of the limit state function. The FORM (first order reliability method) and the SORM (second order reliability method) are carried out to estimate the failure probability of pipeline utilizing the FAD (failure assessment diagram). And the reliability of pipeline is assessed by using this failure probability and analyzed in accordance with a target safety level. Furthermore, the MCS (Monte Carlo Simulation) is used to verify the results of the FORM and the SORM. It is noted that the failure probability increases with the increase of dent depth, gouge depth, operating pressure, outside radius, and the decrease of wall thickness. It is found that the FORM utilizing the FAD is a useful and is an efficient method to estimate the failure probability in the reliability assessment of a pipeline. Furthermore, the pipeline safety assessment technique with the deterministic procedure utilizing the FAD only is turned out more conservative than those obtained by using the probability theory together with the FAD. The probabilistic method such as the FORM, the SORM and the MCS can be used by most plant designers regarding the operating condition and design parameters.

• 한국항해항만학회지
• /
• 제33권4호
• /
• pp.255-261
• /
• 2009

Prediction of the stability for ships is very complex in reality. In this paper, risk based approach is applied to predict the probability of capsize for a certified ship, which is effected by the forces of sea especially the wave loading Safety assessment and risk analysis process are also applied for the probabilistic prediction of stability for ships. The probability of shipsencountering different waves at sea is calculated by the existed statistics data and risk based models. Finally, ship capsizing probability is calculated according to single degree of freedom(SDF) rolling differential equation and basin erosion theory of nonlinear dynamics. Calculation results show that the survival probabilities of ship excited by the forces of the seas, especially in the beam seas status, can be predicted by the risk based method.

• 지질공학
• /
• 제5권1호
• /
• pp.45-58
• /
• 1995

한반도의 계기지진 자료를 이용하여 확률론적 지진위험분포를 계산하였다. 이 목적을 위하여 여러 자료를 수집, 검토하여 통일성과 균질성을 갖도록 지진요소들을 재조정한 1905년 이래의 계기지진 목록을 작성하였다. 한반도에서 100년, 1000년, 4000년 동안에 발생가능한 최대 잠재지진의 규모는 극한치 이론을 적용하여 계산할 때 각각 6.3, 7.2, 7.8로 나타난다. 한편 100년, 1000년에 기대되는 최대지반가속도를 재래주기법으로 계산하여, 등치선도를 작성, 제시하였다. 또한 수개의 인구밀집지역에 대해서는 최대 가속도가 발생확률의 함수로 나타나는 지진재해예상곡선이 작성되었다.

• Nuclear Engineering and Technology
• /
• 제35권2호
• /
• pp.108-120
• /
• 2003

In assessing the long term post closure radiological safety assessment of a potential HLW repository in Korea, three categories of uncertainties exist. The first one is the scenario uncertainty where series of different natural events are translated into written statements. The second one is the modeling uncertatinty where different mathematical models are applied for an identical scenario. The last one is the data uncertainty which can be expressed in terms of probabilistic density functions. In this analysis, three different scenarios are seleceted; a small well scenario, a radiolysis scenario, and a naturally discharged scenario. The MASCOT-K and the AMBER, probabilistic safety assessment codes based on connection of sub-modules and a compartment theory respectively, are applied to assess annual individual doses for a generic biosphere. Results illustrate that for a given scenario, predictions from two different codes fairly match well each other But the discrepancies for the different scenarios are significant. However, total doses are still well below the guideline of 2 mRem/yr. Detailed analyses with model and data uncertainties are underway to further assure the safety of a Korean reference dispsoal concept.

• 한국정보통신학회논문지
• /
• 제23권10호
• /
• pp.1195-1200
• /
• 2019

본 논문은 그룹검사(Group Testing)에 대한 검출 알고리즘을 제안한다. 그룹검사는 다수의 샘플 중 극히 일부의 결함 샘플을 찾는 문제로써 이것은 압축센싱 문제와 유사하다. 본 논문에서는 잡음이 없는 그룹검사를 정의하고, 결함 샘플을 검출하기 위한 확률 기반의 알고리즘을 제안한다. 제안하는 알고리즘은 입력과 출력 신호 간 외부확률들이 서로 교환하여 출력 신호의 사후 확률이 최대가 되도록 구성한다. 그리고 검출 알고리즘에 대한 모의실험을 통해 그룹검사 문제에서 결함 샘플을 찾는다. 본 연구에 대한 모의시험 결과는 정보이론의 하한치와 비교하여 입력과 출력 신호 크기에 따라 실패확률이 얼마나 차이가 있는지 확인한다.