• 한국토양비료학회지
• /
• 제43권2호
• /
• pp.174-179
• /
• 2010

본 연구에서는 사탕무의 시비반응을 합리적으로 해석하고 경제성 있는 최적 시비량 및 최대 수확량의 도출을 위한 방법을 검토하기 위하여 간척지 토양에서 수준별 질소비료와 돈분퇴비의 시용에 따른 사탕무의 시비반응을 조사하였다. 시비반응은 Quadratic model, Exponential model, Square root model 및 Linear response & plateau model을 이용하여 통계적으로 분석하였다. 토양 EC가 증가함에 따라 사탕무 지하부 생체중이 지수함수적으로 감소하였다. 질소비료 시비수준별 사탕무 지하부 생체량의 시비반응은 Linear response에서 결정계수 & plateau model ($R^2$)가 $0.92^{**}$로 고도의 통계적 유의성을 보였으며, Linear response & plateau model에서 최적 질소시비량 ($N_{opt}$)은 138 kg $ha^{-1}$으로 나타났다. 돈분퇴비의 시비반응은 Quadratic model에서 $0.99^{**}$의 유의성 있는 결정계수 ($R^2$)를 보였으며, 최적 퇴비시비량은 각각 9.17 tons $ha^{-1}$으로 나타났다. 상기 결과로 볼 때 사탕무에서 질소와 돈분퇴비의 시비반응을 적절히 평가하기 위해서는 각각 Linear response & plateau model과 Quadratic model을 이용하는 것이 합리적인 것으로 판단되었다.

• 응용통계연구
• /
• 제32권4호
• /
• pp.631-642
• /
• 2019

다수의 항목무응답이 발생한 표본조사에서는 추정의 정확성이 떨어진다. 이를 해결하기 위한 많은 방법이 개발되었으나 응답률이 관심변수에 의해 영향을 받는 경우임에도 이를 고려하지 않고 랜덤으로 무응답이 발생한다는 가정 하에서 사용하는 무응답 처리 방법을 사용하게 되면 편향이 발생하는 것으로 알려져 있다. Chung과 Shin (2017)과 Min과 Shin (2018)은 응답률이 관심변수의 함수인 경우에서 발생된 편향을 적절히 처리하여 추정의 정확성을 향상시키는 방법을 제안하였다. 본 연구에서는 응답률 함수가 선형(linear)이면서 초모집단 모형의 오차가 정규분포를 따르는 경우를 살펴보았으며 층별 모집단 수가 편향 보정에 영향을 주는지도 살펴보았다. 모의실험을 통하여 제안된 추정량의 성능을 살펴보았으며 실제 자료 분석을 통해 이를 확인하였다.

• 산업경영시스템학회지
• /
• 제44권1호
• /
• pp.26-36
• /
• 2021

The topic of this study is the field of humanitarian logistics for disaster response. Many existing studies have revealed that compliance with the golden time in response to a disaster determines the success or failure of relief activities, and logistics costs account for 80% of the disaster response cost. Besides, the agility, responsiveness, and effectiveness of the humanitarian logistics system are emphasized in consideration of the disaster situation's characteristics, such as the urgency of life-saving and rapid environmental changes. In other words, they emphasize the importance of logistics activities in disaster response, which includes the effective and efficient distribution of relief supplies. This study proposes a mathematical model for establishing a transport plan to distribute relief supplies in a disaster situation. To determine vehicles' route and the amount of relief for cities suffering a disaster, it mainly considers the urgency, effectiveness (restoration rate), and uncertainty in the logistics system. The model is initially developed as a mixed-integer nonlinear programming (MINLP) model containing some nonlinear functions and transform into a Mixed-integer linear programming (MILP) model using a logarithmic transformation and piecewise linear approximation method. Furthermore, a minimax problem is suggested to search for breakpoints and slopes to define a piecewise linear function that minimizes the linear approximation error. A numerical experiment is performed to verify the MILP model, and linear approximation error is also analyzed in the experiment.

• 응용통계연구
• /
• 제33권1호
• /
• pp.11-23
• /
• 2020

표본조사는 비용과 시간을 절약하면서도 주어진 정확성을 만족하는 통계를 얻을 수 있다. 그러나 최근에는 다수의 무응답 발생으로 인해 조사의 정확성이 크게 떨어지고 있다. 무응답은 다양한 이유로 발생하고 있으나 무응답이 관심변수와 함수 관계가 있는 경우에는 이 정보를 이용하여 무응답을 적절히 처리해야 추정의 정확성이 유지될 수 있다. 최근 Chung과 Shin (2017, 2019), Min과 Shin (2018)은 응답률이 관심변수의 지수 또는 선형함수이고 초모집단모형의 오차가 정규분포를 따를 때 무응답으로 인해 발생한 편향을 제거함으로써 추정의 정확성이 향상되는 것을 확인하였다. 이에 본 연구에서는 사업체조사에서 초모집단모형의 오차가 감마분포 또는 로그-정규분포를 따르는 경우에서의 무응답 편향보정 추정량을 제안하였다. 또한 모의실험을 통하여 제안된 추정량의 우수성을 확인하였다.

• 한국정밀공학회지
• /
• 제14권9호
• /
• pp.124-129
• /
• 1997

The purpose of this study is to predict the stability and frequency response of multipurpose vehicle. The vehicle model has seven degrees of freedom. The motion equations are derived by using Lagrangian equation and linearized. The positions of eigenvalues of model which are dominated by lateral velocity, yaw rate, roll rate of sprung mass are used to predict the stability of motion. The resonse of sprung mass to steering wheel is simulated in time domain. It is predicted that the roll response of sprung mass would rather be improved by modifying the position of eigenvalues. The responses of sprung mass to steering wheel are also simulated in frequency domain. The magnitude and phase plots of gains are evaluated in driver's steering input frequency range.

• Computers and Concrete
• /
• 제7권4호
• /
• pp.317-329
• /
• 2010

The paper describes localization of deformation in a bar under tensile loading. The material of the bar is considered as non-linear viscous elastic and the bar consists of two symmetric halves. It is assumed that the model represents behavior of the quasi-brittle viscous material under uniaxial tension with different loading rates. Besides that, the bar could represent uniaxial stress-strain law on a single plane of a microplane material model. Non-linear material property is taken from the microplane material model and it is coupled with the viscous damper producing non-linear Maxwell material model. Mathematically, the problem is described with a system of two partial differential equations with a non-linear algebraic constraint. In order to obtain solution, the system of differential algebraic equations is transformed into a system of three partial differential equations. System is subjected to loadings of different rate and it is shown that localization occurs only for high loading rates. Mathematically, in such a case two solutions are possible: one without the localization (unstable) and one with the localization (stable one). Furthermore, mass is added to the bar and in that case the problem is described with a system of four differential equations. It is demonstrated that for high enough loading rates, it is the added mass that dominates the response, in contrast to the viscous and elastic material parameters that dominated in the case without mass. This is demonstrated by several numerical examples.

• 산업경영시스템학회지
• /
• 제47권3호
• /
• pp.68-74
• /
• 2024

This study develops a model to determine the input rate of the chemical for coagulation and flocculation process (i.e. coagulant) at industrial water treatment plant, based on real-world data. To detect outliers among the collected data, a two-phase algorithm with standardization transformation and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) is applied. In addition, both of the missing data and outliers are revised with linear interpolation. To determine the coagulant rate, various kinds of machine learning models are tested as well as linear regression. Among them, the random forest model with min-max scaled data provides the best performance, whose MSE, MAPE, R² and CVRMSE are 1.136, 0.111, 0.912, and 18.704, respectively. This study demonstrates the practical applicability of machine learning based chemical input decision model, which can lead to a smart management and response systems for clean and safe water treatment plant.

• 응용통계연구
• /
• 제30권6호
• /
• pp.993-1004
• /
• 2017

표본조사에서는 추정의 정확성 및 정밀성 향상을 위해 흔히 층화추출법을 사용하며 층 내에서는 동일한 표본 가중치를 이용하여 표본을 추출한다. 그러나 실제 응답률은 관심변수 값에 영향을 받을 수 있기 때문에 주어진 동일한 가중치는 응답률을 반영하여 보정되어야 한다. 또한 관심변수가 연속형 보조변수와 선형 관계가 있고 보조변수를 기준으로 층이 나누어진 경우에는 층 내에서 동일한 가중치를 사용하는 것 보다 층을 세분화한 후 얻어진 가중치를 사용하는 것이 효과적일 수 있다. 본 연구에서는 응답률이 관심변수 자료 값의 지수함수이고, 관심변수가 보조변수와 선형 관계가 있을 때 정보적 표본설계 기법을 이용하여 추정의 정확성과 정밀성을 높이는 방법을 제안하였다. 또한 모의실험을 통하여 제안된 방법의 우수성을 확인하였다.

• 한국안전학회지
• /
• 제35권1호
• /
• pp.62-69
• /
• 2020

Most people suffer from stress during day life because modernized society is very complex and changes fast. Because stress can affect to many kind of physiological phenomena it is even considered as a disease. Therefore, it should be detected earlier, then must be released. When a person is being stressed several bio-signals such as heart rate, etc. are changed. So, those can be detected using medical electronics techniques. In this paper, stress assessment system is studied using random forest algorithm based on heart rate, RR interval and Galvanic skin response. The random forest model was trained and tested using the data set obtained from the bio-signals. It is found that the stress assessment procedure developed in this paper is very useful.

• The Plant Pathology Journal
• /
• 제26권1호
• /
• pp.17-24
• /
• 2010

A logistic model for describing combined effects of both temperature and wetness period on appressorium formation was developed using laboratory data on percent appressorium formation of Colletotrichum acutatum. In addition, the possible use of the logistic model for forecasting infection risks was also evaluated as compared with a first-order linear model. A simplified equilibrium model for enzymatic reactions was applied to obtain a temperature function for asymptote parameter (A) of logistic model. For the position (B) and the rate (k) parameters, a reciprocal model was used to calculate the respective temperature functions. The nonlinear logistic model described successfully the response of appressorium formation to the combined effects of temperature and wetness period. Especially the temperature function for asymptote parameter A reflected the response of upper limit of appressorium formation to temperature, which showed the typical temperature response of enzymatic reactions in the cells. By having both temperature and wetness period as independent variables, the nonlinear logistic model can be used to determine the length of wetness periods required for certain levels of appressorium formation under different temperature conditions. The infection model derived from the nonlinear logistic model can be used to calculate infection risks using hourly temperature and wetness period data monitored by automated weather stations in the fields. Compared with the nonlinear infection model, the linear infection model always predicted a shorter wetness period for appressorium formation, and resulted in significantly under- and over-estimation of response at low and high temperatures, respectively.