• Information Systems Review
• /
• v.15 no.1
• /
• pp.45-68
• /
• 2013

Despite improved technologies, procedures, and regulations, consumers are still uncertain about purchasing online. The objective of this study is to understand uncertainty factors in online shopping and their relationships with the consumers' intention to purchase. For this objective we derived seller anonymity, lack of product transparency, and lack of process transparency as uncertainty factors from previous researches which may affect consumers' perceived uncertainty on online shopping. Then, a causal model was developed to conceptualize the relationships between these uncertainty factors as antecedent variables and consumer's intention to purchase as consequent variable with perceived uncertainty as an intermediary variable. Purchase involvement was used as a moderating variable on the relationship between perceived uncertainty and the intention to purchase online. The model was tested empirically to find meaningful relationships among these variables. The findings indicate that all antecedent variables affect perceived uncertainty significantly and perceived uncertainty negatively affects consumers' intention to purchase. Moreover, the results of analysis show purchase involvement has a significant moderating effect on the relationship between perceived uncertainty and intention to purchase online.

• Journal of Korea Water Resources Association
• /
• v.49 no.5
• /
• pp.411-422
• /
• 2016

Rainfall-runoff modeling in conjunction with rainfall frequency analysis has been widely used for estimating design floods in South Korea. However, uncertainties associated with underlying distribution and sampling error have not been properly addressed. This study applied a Bayesian method to quantify the uncertainties in the rainfall frequency analysis along with Gumbel distribution. For a purpose of comparison, a probability weighted moment (PWM) was employed to estimate confidence interval. The uncertainties associated with design rainfalls were quantitatively assessed using both Bayesian and PWM methods. The results showed that the uncertainty ranges with PWM are larger than those with Bayesian approach. In addition, the Bayesian approach was able to effectively represent asymmetric feature of underlying distribution; whereas the PWM resulted in symmetric confidence interval due to the normal approximation. The use of long period data provided better results leading to the reduction of uncertainty in both methods, and the Bayesian approach showed better performance in terms of the reduction of the uncertainty.

• Journal of Korean Society of Transportation
• /
• v.21 no.1
• /
• pp.77-90
• /
• 2003

This paper explores the way and the extent to which drivers' route choice was influenced by uncertain information. In particular, this paper investigates the effect of qualitative information on route choice when drivers face a choice with different degrees of uncertain information. The SP survey was conducted and route choice legit models were estimated. We also applied Prospect Theory to the analysis of drivers' decision making under uncertain information. The main findings are firstly, drivers tend to prefer a route with information than(to) one without information. This indicated that providing charge information encouraged drivers to choose the routes for which information is provided in preference to those for which it is not provided. Secondly, drivers also prefer a route with a certain and precise information over one with uncertain and imprecise information. Thirdly, when the information is given as a range, the size of the range of the information influenced route choice slightly and as the range of the charge increases, the route becomes slightly less unattractive. Fourthly, when the information is given as a range, drivers' route choices are influenced more by the median value of the ranges than by the size of the overall ranges of the information. Application of Prospect Theory to the results explains the way drivers may be interpreting the choice situation and how they make a route choice in response to uncertain information. The results of this paper implicate that drivers' decision making under uncertainty seem to be very complicated and flexible, depending on the way drivers interpret the choice situation. Therefore, it is recommended to apply wider related theories to the analysis of the drivers' behaviour.

• Journal of Korea Water Resources Association
• /
• v.49 no.8
• /
• pp.657-671
• /
• 2016

It is expected that water resources will be changed spatially and temporally due to the global climate change. The quantitative assessment of change in water availability and appropriate water resources management measures are needed for corresponding adaptation. However, there are large uncertainties in climate change impact assessment on water resources. For this reason, development of technology to evaluate the uncertainties quantitatively is required. The objectives of this study are to develop the climate change uncertainty assessment method and to apply it. The 5 RCMs (HadGEM3-RA, RegCM4, MM5, WRF, and RSM), 5 statistical post-processing methods (SPP) and 2 hydrological models (HYM) were applied for evaluation. The results of the uncertainty analysis showed that the RCM was the largest sources of uncertainty in Spring, Summer, Autumn (29.3~68.9%), the hydrological model was the largest source of uncertainty in Winter (46.5%). This method can be possible to analyze the changes in the total uncertainty according to the specific RCM, SPP, HYM model. And then it is expected to provide the method to reduce the total uncertainty.

• Environmental and Resource Economics Review
• /
• v.26 no.3
• /
• pp.359-398
• /
• 2017

Uncertainty is central to energy and climate policy. A growing number of literature show that almost all components of energy and climate models are, to some extent, uncertain and that the effect of uncertainty on the model outputs, in turn policy recommendations, is significantly large. Most existing energy and climate-economy models developed and used in Korea, however, do not take uncertainty into account explicitly. Rather, many models conduct a deterministic analysis or do a simple (limited) sensitivity analysis. In order to help social planners to make more robust decisions (across various plausible situations) on energy and climate change issues, an uncertainty analysis should be conducted. As a first step, this paper reviews the theory of decision making under uncertainty and the method for addressing uncertainty of existing probabilistic energy and climate-economy models. In addition, the paper proposes a strategy to apply an uncertainty analysis to energy and climate-economy models used in Korea. Applying the uncertainty analysis techniques, this paper revises the FUND model and investigates the impacts of climate change in Korea.

• Journal of the Korean Society for Precision Engineering
• /
• v.9 no.1
• /
• pp.18-28
• /
• 1992

시스템의 모델링 과정에서 발생될 수 있는 불확실성(uncertainty) 혹은 미지의 가반중량을 비롯한 외란에 의해 발생되는 불확실성 등을 갖고있는 로봇의 강건 추적제어기 설계를 위해 가변구조시스템(variable structure system) 이론을 적용하였다. 시스템 방정식과 연계하여 슬라이딩 모드가 존재하기 위한 조건을 구했으며, 입력 에 대한 불확실성은 매칭조건(matching condition)을 가정하여 다루었다. 기존의 방법에 비해 제어기 설계과정이 간단 명료하며 요구되는 궤적에 대한 추적제어 효과 또한 매우 우수함을 컴퓨터 시뮬레이션을 통해 입증하였다.

• Proceedings of the KSRS Conference
• /
• 2006.03a
• /
• pp.311-314
• /
• 2006

본 연구에서는 서로 다른 센서간의 영상 자료 융합을 위하여 Dempster-Shafer 기법을 제안하고 있다. 제안 된 Dempster-Shafer 기법은 불확실성의 최소 값을 대표하는 Belief 함수와 불확실성의 최대 값을 나타내는 Plausibility 함수를 사용한다. 이러한 두 함수의 차이는 Belief Interval 로 정의되며 이 값은 분석 대상에 존재하는 불확실 정도의 Measure 로 사용되며 Evidence Combination의 이론에 근거하여 서로 다른 센서간의 자료 융합이 가능하며 분류 결과로 클래스 맵 뿐 만 아니라 분류 결과에 대한 불확실성 정도를 나타내는 Belief 함수 값과 Plausibility 함수 값을 생성하여 분류 결과에 대한 보충적인 분석을 가능하게 하여 사용자의 분석 정확성을 증대 시킬 수 있다.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2011.11a
• /
• pp.243-247
• /
• 2011

본 논문에서는 원자력발전소의 화재 안전성평가에서 제기되는 화재모델링 불확실성 분석 방법을 검토하고 논의하였다. 원자력발전소의 성능기반 화재 안전성평가에 대해서는 NUREG-1934를, 확률론적 화재 안전성 평가에 대해서는 NUREG/CR-6850를 중심으로 화재 모델링 불확실성 분석 방법을 소개하고 몬테칼로 시뮬레이션을 이용한 불확실성 분석 방법에 대해 논의하였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.29-29
• /
• 2011

SWAT (Soil and Water Assessment Tool) 모형은 물리적 기반의 준분포형 강우-유출 모형으로서, 대규모의 복잡한 유역에서 장기간에 걸친 다양한 종류의 토양과 토지이용 및 토지관리 상태에 따른 유출과 유사 및 오염물질의 거동에 대한 토지관리 방법의 영향을 예측이 가능하여, 수자원 관리 계획 및 유역관리를 위한 의사결정 지원 등 그 적용 범위가 매우 광범위하다. 이러한 모형의 적용성 검증을 위해서는 매개변수 민감도 분석 및 검 보정, 예측 불확실성 분석을 필요로 한다. 최근 수문 모델의 불확실성을 분석하기 위한 다양한 기법들이 개발 되었는데, 본 연구는 충주댐 유역(6,581.1 m)을 대상으로 유역출구점의 실측 일 유출량 자료(1998~2003)를 바탕으로 SWAT 모형의 유출관련 매개변수(총 18개)에 대한 불확실성 분석을 실시하였다. 이때 사용된 분석 기법으로는 SUFI2 (Sequential Uncertainty FItting algorithm 2), GLUE (Generalized Likelihood Uncertainty Estimation), ParaSol (Parameter Solution)등을 적용 하였다. 이러한 기법은 모두 SWAT-CUP (SWAT-Calibration Uncertainty Program, Abbaspour, 2007) 모형에 탑재되어있으며, 모형의 결과로써 검 보정, 매개변수의 민감도 분석, 각종 목적 함수 및 불확실성의 범위 등이 자동으로 산출 되므로 모형의 사용자가 불확실성 평가 기법의 분석 및 비교를 손쉽게 할 수 있다. 그 결과 대표적인 목적 함수인 결정 계수( $^2$)와 NSE (Nash-Sutcliffe Model Efficiency)는 모두 0.65에서 0.92사이의 값을 나타내어 대체적으로 모의가 잘 이루어졌음을 알 수 있었다. 그러나 불확실성의 범위를 나타내는 지표인 p-factor 및 r-factor에서는 평가 기법 별로 그 차이가 확연하게 드러났다. 여기서 p-factor는 불확실성 범위에 실측치가 포함되는 비율이며, r-factor는 불확실성의 상대적인 범위로 각각 1과 0에 가까울수록 모의 기법의 성능이 우수함을 의미한다. 세 가지 알고리듬 중에서 SUFI2의 p-factor가 약 0.51로 가장 높게 나타났으며, ParaSol의 r-factor가 0.00으로 가장 작게 나타났다. 여기서 p-factor는 불확실성 범위에 실측치가 포함되는 비율이며, r-factor는 불확실성의 상대적인 범위를 의미한다. 본 연구의 결과는 SWAT 모형을 이용한 수문모델링에서 수문분석에 따른 예측결과의 불확실성을 정량적으로 평가함으로서, 모형의 적용성 평가 및 모의결과의 신뢰성 확보에 근거자료로 활용이 가능할 것으로 판단된다.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.9
• /
• pp.111-120
• /
• 2013

Uncertainty in Context-aware computing is mainly a consequence of the complexity of context acquisition mechanisms and context processing. The presence of uncertainty may harm the users' confidence in the application, rendering it useless. This paper describes a three-phase strategy to manage uncertainty by identifying its possible sources, representing uncertain information, and determining how to proceed, once uncertain context is detected. The level of effort that is necessary to eliminate the uncertainty of context information affects the reliability of the system, because Sensor network system have no intervention of humans. In this paper, We applied proposed method to the development for the sensor network system, Uncertainty management can be applied a part of the system development life-cycle. It confirmed that result of testing show that detection performance is stable.