• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.18 no.1
• /
• pp.21-34
• /
• 2006

The simulation of tsunami inundation using detailed bathymetry and topography is required to establish the countermeasure of disaster mitigation and the tsunami hazard map. In this study, a simulation of the 1983 tsunami event in the East Sea using parallel finite element model, which is possible to simulate with suitable accuracy by the Beowulf parallel computation method, is performed to produce detailed features of coastal inundation. Results of simulations are compared with measured data. The evolution of statistic distribution of tsunami heights is studied numerically and the distribution functions of tsunami heights show a tendency to the log-normal curve along coastal area.

• Journal of the Korea Society for Simulation
• /
• v.17 no.2
• /
• pp.75-82
• /
• 2008

The times of service commencement and service completion had been used for inferring the queueing systems. However, the service commencement times are difficult to measure because of unobservable nature in queueing systems. In this paper, for inferring queueing systems, the service commencement times are replaced for arrival times which can be easily observed. Determining the service commencement time is very important in our methods. The methods for first come first served(FCFS), last come first served(LCFS) queueing discipline are already developed in our previous work. In this paper, we extend to random selection for service(RSS) queueing discipline. The performance measures we used are mean queueing time and mean service time, the variances of two. The simulation results verify our proposed methods to infer queueing systems under RSS discipline.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.10 no.3 s.49
• /
• pp.35-45
• /
• 2006

The simulation of tsunami using detailed bathymetry and topography is required to establish the countermeasure of disaster mitigation and the tsunami hazard map. In this study, a simulation of the 1993 tsunami event in the East Sea using parallel finite element model, which is possible to simulate with suitable accuracy by the Beowulf parallel computation method, is performed to produce detailed features of coastal inundation. Results of simulation are compared with measured data. The evolution of statistic distribution of tsunami heights is studied numerically and the distribution functions of tsunami heights show a tendency to the log-normal curve along coastal area.

• Fire Science and Engineering
• /
• v.31 no.3
• /
• pp.137-143
• /
• 2017

A sufficient number of ambulances are critical for preventing delayed vehicle dispatch for emergency patients. This study presents effective methodologies for evaluating the effects of ambulance quantities on availability. The statistical properties of the emergency requests and responses were analyzed for a city in Korea. The inter-request times were modeled by statistical distributions. The ambulance dispatch was modeled using simulation, reflecting the shared dispatch among the city districts. The simulation results revealed that the existing ambulance quantity could successfully meet the majority of the requests, but more vehicles were desirable for improvement. The locations of the additional vehicles were determined efficiently by simulations with a greedy approach. The simulations with added vehicles showed a significantly better coverage of the emergency calls. This research can help design improved emergency vehicle operations, and help save lives.

• Journal of Internet Computing and Services
• /
• v.5 no.4
• /
• pp.63-76
• /
• 2004

For efficient design and operation of a communication network, precise simulation of network characteristics is essential. This issue has been the focus of research by several groups. In this study, we first modeled the HTTP traffic which would be employed on simulation on the level of application using the real collected traffic data. There are two different viewpoints on the characteristics of web traffic pattern, Poisson distribution and self-similar characteristics. In our study, the results show that web traffic characteristics do not depend on only one type of distribution, but the traffic can be modeled as composition of these depending on the size of response of Web server. This implicates that the web traffic can be modeled as the combination of two characteristics. We also found that the characteristics of Web traffic rely on the properties of web servers. This result was deployed as a traffic generator in implementing the network simulator (NetDAS).

• Journal of the Korea Society for Simulation
• /
• v.30 no.1
• /
• pp.75-85
• /
• 2021

In this paper, we present a deep neural network-based prediction model that processes and analyzes the corporate credit and personal credit information of individual business owners as a new method to predict the default rate of individual business more accurately. In modeling research in various fields, feature selection techniques have been actively studied as a method for improving performance, especially in predictive models including many features. In this paper, after statistical verification of macroeconomic indicators (macro variables) and credit information (micro variables), which are input variables used in the default rate prediction model, additionally, through the credit information feature selection method, the final feature set that improves prediction performance was identified. The proposed credit information feature selection method as an iterative & hybrid method that combines the filter-based and wrapper-based method builds submodels, constructs subsets by extracting important variables of the maximum performance submodels, and determines the final feature set through prediction performance analysis of the subset and the subset combined set.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.47 no.6
• /
• pp.12-18
• /
• 2010

Automatic Target Detection (ATD) systems that use forward-looking infrared (FLIR) consists of three stages. preprocessing, detection, and clutter rejection. All potential targets are extracted in preprocessing and detection stages. But, this results in a high false alarm rates. To reduce false alarm rates of ATD system, true targets are extracted in the clutter rejection stage. This paper focuses on clutter rejection stage. This paper presents a new clutter rejection technique using PCA features and stochastic features of clutters and targets. PCA features are obtained from Euclidian distances using which potential targets are projected to reduced eigenspace selected from target eigenvectors. CV is used for calculating stochastic features of edges in targets and clutters images. To distinguish between target and clutter, LDA (Linear Discriminant Analysis) is applied. The experimental results show that the proposed algorithm accurately classify clutters with a low false rate compared to PCA method or CV method

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.42 no.1
• /
• pp.1-9
• /
• 2022

This study was carried out to compare the DMY (dry matter yield) of IRG (Italian ryegrass) in the southern coastal regions of Korea due to seasonal climate scenarios such as the Kaul-Changma (late monsoon) in autumn, extreme winter cold, and drought in the next spring. The IRG data (n = 203) were collected from various Reports for Collaborative Research Program to Develop New Cultivars of Summer Crops in Jeju, 203 Namwon, and Yeungam from the Rural Development Administration - (en DASH). In order to define the seasonal climate scenarios, climate variables including temperature, humidity, wind, sunshine were used by collected from the Korean Meteorological Administration. The discriminant analysis based on 5% significance level was performed to distinguish normal and abnormal climate scenarios. Furthermore, the DMY comparison was simulated based on the information of sample distribution of IRG. As a result, in the southern coastal regions, only the impact of next spring drought on DMY of IRG was critical. Although the severe winter cold was clearly classified from the normal, there was no difference in DMY. Thus, the DMY comparison was simulated only for the next spring drought. Under the yield comparison simulation, DMY (kg/ha) in the normal and drought was 14,743.83 and 12,707.97 respectively. It implies that the expected damage caused by the spring drought was about 2,000 kg/ha. Furthermore, the predicted DMY of spring drought was wider and slower than that of normal, indicating on high variability. This study is meaningful in confirming the predictive DMY damage and its possibility by spring drought for IRG via statistical simulation considering seasonal climate scenarios.

• Journal of the Korea Concrete Institute
• /
• v.26 no.3
• /
• pp.343-350
• /
• 2014

A probabilistic construction stage analysis using the Monte Carlo Simulation was performed to address the effects of uncertainty regarding the material properties, environmental factors, and applied forces. In the previous research, creep and shrinkage were assumed to be completely independent random variables. However, because of the common influencing factors in the material models for the creep and shrinkage estimation, strong correlation between creep and shrinkage can be presumed. In this paper, an Monte Carlo Simulation using CEB-FIB creep and shrinkage equations were performed to actually evaluate the correlation coefficient between two phenomena, and then another Monte Carlo Simulation to evaluate the statistical properties of axial strain affected by partially correlated random variables including the material properties, environmental factors, and applied forces. The results of Monte Carlo Simulation were compared with measured strains of a column on a first story in a 58-story building. Comparison indicated that the variation due to the uncertainty related with the material properties were most severe. And measured strains was within the range of mean+standard deviation.

• The Journal of the Korea Contents Association
• /
• v.10 no.5
• /
• pp.268-284
• /
• 2010

The purpose of this study was to evaluate the effects of horseback riding simulation training(HRST) on gross motor function of children with cerebral palsy. Twenty-four children with cerebral palsy participated were divided into two groups randomly. Control group and experimental group both received 30 minutes of regular neuro-rehab exercises for three times a week per twelve weeks, but experimental group received additional 15 minutes of the HRST. Gross Motor Function Measure(GMFM) was used to evaluate the changes before and after HRST. Normalized GMFM scores were compared using Mann-Whitney U test, statistical significance was set at a=.05. The results were as followings: First, in each compared before and after the experiment, the control group showed significantly increased GMFM score in dimensions A and B. The experimental group showed significant increasing in all dimensions of GMFM test. Second, the experimental group showed significant differences comparing to control group in dimensions C, D and E of GMFM test. Consequently, HRST should be considered as a therapeutic method for physical therapy for the children with cerebral palsy to improve the functional movements.