• International conference on construction engineering and project management
• /
• 2013.01a
• /
• pp.375-385
• /
• 2013

As climate change and environmental pollution become one of the biggest global issues today, new renewable energy, especially solar photovoltaic (PV) system, is getting great attention as a sustainable energy source. However, initial investment cost of PV system is considerable, and thus, it is crucial to predict electricity generation accurately before installation of the system. This study analyzes the loss ratio of solar photovoltaic electricity generation from the actual PV system monitoring data to predict electricity generation more accurately in advance. This study is carried out with the following five steps: (i) Data collection of actual electricity generation from PV system and the related information; (ii) Calculation of simulation-based electricity generation; (iii) Comparative analysis between actual electricity generation and simulation-based electricity generation based on the seasonality; (iv) Stochastic approach by defining probability distribution of loss ratio between actual electricity generation and simulation-based electricity generation ; and (v) Case study by conducting Monte-Carlo Simulation (MCS) based on the probability distribution function of loss ratio. The results of this study could be used (i) to estimate electricity generation from PV system more accurately before installation of the system, (ii) to establish the optimal maintenance strategy for the different application fields and the different season, and (iii) to conduct feasibility study on investment at the level of life cycle.

• Journal of the Architectural Institute of Korea Planning & Design
• /
• v.35 no.1
• /
• pp.19-28
• /
• 2019

The purpose of this study is to evaluate the walking behavior analysis methodology used in the previous studies, paying attention to the demand for empirical data collecting for urban and neighborhood planning. The preceding researches are divided into (1)Recording, (2) Surveys, (3)Statistical data, (4)Global positioning system (GPS) devices, and (5)Mobile Big Data analysis. Next, we analyze the precedent research and identify the changes of the walkability research. (1)being required empirical data on the actual walking and moving patterns of people, (2)beginning to be measured micro-walking behaviors such as actual route, walking facilities, detour, walking area. In addition, according to the trend of research, it is analyzed that the use of GPS device and the mobile big data are newly emerged. Finally, we analyze pedestrian data based on mobile big data in terms of 'application' and distinguishing it from existing survey methodology. We present the possibility of mobile big data. (1)Improvement of human, temporal and spatial constraints of data collection, (2)Improvement of inaccuracy of collected data, (3)Improvement of subjective intervention in data collection and preprocessing, (4)Expandability of walking environment research.

• Journal of the Korea Convergence Society
• /
• v.13 no.1
• /
• pp.23-30
• /
• 2022

The failure of the reefer container causes a great loss of cost, but the current reefer container alarm system is inefficient. Existing studies using simulation data of refrigeration systems exist, but studies using actual operation data of refrigeration containers are lacking. Therefore, this study classified the causes of failure using actual refrigerated container operation data. Data imbalance occurred in the actual data, and the data imbalance problem was solved by comparing the logistic regression analysis with ENN-SMOTE and class weight with the 2-stage algorithm developed in this study. The 2-stage algorithm uses XGboost, LGBoost, and DNN to classify faults and normalities in the first step, and to classify the causes of faults in the second step. The model using LGBoost in the 2-stage algorithm was the best with 99.16% accuracy. This study proposes a final model using a two-stage algorithm to solve data imbalance, which is thought to be applicable to other industries.

• Smart Media Journal
• /
• v.13 no.2
• /
• pp.9-15
• /
• 2024

In this paper, a Diffusion Multi-step Classifier (DMC) is proposed to address the imbalance issue in credit prediction. DMC utilizes a Diffusion Model to generate continuous numerical data from credit prediction data and creates categorical data through a Multi-step Classifier. Compared to other algorithms generating synthetic data, DMC produces data with a distribution more similar to real data. Using DMC, data that closely resemble actual data can be generated, outperforming other algorithms for data generation. When experiments were conducted using the generated data, the probability of predicting delinquencies increased by over 20%, and overall predictive accuracy improved by approximately 4%. These research findings are anticipated to significantly contribute to reducing delinquency rates and increasing profits when applied in actual financial institutions.

• Journal of Preventive Medicine and Public Health
• /
• v.10 no.1
• /
• pp.109-117
• /
• 1977

The improvement of civil registration reguires continuous study rather than periodic efforts. More and better statistics, however, are urgently required to formulate development programs and planning. Data obtainable from the civil registration are usually marred by errors of omission which are difficult to correct. This study aimed at finding out the problems occuring when a set of crude birth registration data in a rural area is used. Data Sources of this study are: 1) For birth registration: government birth registration records obtained from myun office and other government offices. 2) For the actual number of births: birth and child records from the Kang Wha Community Health Project The study area is Sunwon Myun and Naega Myun in Kang Wha Gun, Gyunggido. The referrance period for the accumulated data is one full year: Jan. 1st 1975-Dec. 31st 1975 Major findings are as follows: If the number of registered births is compared with the actual number of births which occured in the target area, the former is far greater than the latter. The general assumption usually is, that the actual number of births exceeds the registered number of birth in Korea. The observation from this specific study in this specific target area, shows the opporsite trend. The number of births which were registered is 550. This is more than two times as much as the number of births which actually occured during the year of 1975 in the study area namely 256. The difference comes mainly from the fact that many cases of births from other areas were registered in the target area. In other words birth is not registered where it occured but where the permanent residence adress is. Among 550 births registered in the target area 66% did not occur in the target area. Only one third of all registered births were registered within the legal period for birth registration which is 2 weeks. 34% of the registered births actually occured in 1974, but were registered in 1975. In 55% of the cases a difference was observed between the actual date of birth and the registered date of birth. From the 256 births which occured in the target area, only 153 births (59%) were registered at the myun office and the remaining 130 births (41%) were not resistered there in the year of study. 6% of the 550 cases listed as registered have no seperate registration sheets. Nevertheless, they definitly have been registered in the birth list at the myun office. 3% of the 550 cases are not recorded in this list but have a separate registration sheet at the myun office. In conclusion, birth registration data have many errors and problems. Their usefulness as. a source for vital and other statistics should be reconsidered. A series of sound methological studies will be necessary to establish their actual usefulness. A continuous and permanent compulsory system of birth recording is needed.

• The Journal of Information Systems
• /
• v.23 no.1
• /
• pp.185-202
• /
• 2014

ERP, a typical enterprise application, helps companies to increase their productivity and to support their decision makings. ERP is composed of diverse functions that are optimized under PC environment, whereas the ERP applications on a mobile platform have many constraints such as a small screen, limited resolution, and computing power. Because all the functions of a ERP legacy system are not required for ERP on a mobile device, the core functions of the ERP system should be selected to increase system efficiency. In this study, two main methods were used; interviews and log analyses. The end users using a ERP system were interviewed for their perceptions, and log data analyses were made for the hitting number of specific ERP functions. The differences between the actual usage based on log data and users' cognitive preferences about ERP functions were analysed. Finally, the functional differences between users' perception and actual usage were suggested for some practical implications.

• The Research Journal of the Costume Culture
• /
• v.12 no.1
• /
• pp.90-103
• /
• 2004

The purpose of the present study was to investigate images pursued and purchase behaviors for clothing as determined by self-image. Data were collected through a self-administered questionnaire survey from March 3 to March 11, 2003 from 600 female students attending universities in Seoul; 514 were used for the data analysis. Data were analyzed by chi-square analysis, t-test, ANOVA, correlation analysis, tics, cluster analysis and Duncan's multiple range test. Self image was defined six factors: social image, gay image, intellectual image, girlish image, iron nerves image, image like a man and was classified three group avail of six factor: commonness type, social brilliance type, immature boldness type. The results showed significant differences in images of actual clothing worn by self and in the clothing image pursued among the groups determined by the self image. Significant differences were also found in clothing purchase behaviors such as monthly clothing expenditure, shopping frequency, store types, and the clothing items often used for expressing self-image among the groups divided by self-image.

• Korean Institute of Interior Design Journal
• /
• v.17 no.4
• /
• pp.66-74
• /
• 2008

A form of consumption has been changed to a form of pursuing rather qualitative sides than quantitative sides according to an improvement of the living level followed by an increase in national income and to an increase in women's advance to society in modern society. The above change had a lot of effects on children play space, a subsidiary means within commercial facilities. However, partial children play facilities have steady children injury cases by a lack of safety management facilities. Therefore, the study investigated indoor children play facilities installed in large-scale distribution stores as department stores and discount stores out of commercial facilities in Seoul. The study investigated actual conditions and collected picture data with a checklist drawn up on the basis of the safety management standard for three months from April, 2008 to June, 2008 by visiting five places under cooperation. The purpose of the study was to grasp actual conditions of safety management of play space within commercial facilities through the contents and picture data of the investigated checklist and to offer basic data in the side of safety management of commercial facilities and various indoor children play spaces by proposing security measures for the problems.

• Journal of KIBIM
• /
• v.9 no.3
• /
• pp.1-7
• /
• 2019

Digital twin technology has been actively developed to monitor and assess the current state of actual structures. The digital twin changes the traditional observation method performed in the field to the real-time observation and detection system using virtual online model. Thus, this study designed a digital twin model for a beam and examined the feasibility of the digital twin for bridges. To reflect the current state of the bridge, model updating was performed according to the field test data to construct an analysis model. Based on the constructed bridge analysis model, the relationship between strain and displacement was used to represent a virtual model that behaves in the same way as the actual structure. The strain and displacement relationship was expressed as a matrix derived using an approximate analytical theory. Then, displacements can be obtained using the measured data obtained from strain sensors installed on the bridge. The coordinates of the obtained displacements are used to construct a virtual digital model for the bridge. For verification, a beam was fabricated and tested to evaluate the digital twin model constructed in this study. The displacements obtained from the strain and displacement relationship agrees well with the actual displacements of the beam. In addition, the displacements obtained from the virtual model was visualized at the locations of the strain sensor.

• KEPCO Journal on Electric Power and Energy
• /
• v.7 no.1
• /
• pp.171-177
• /
• 2021

In terms of distribution planning, accurate electric load prediction is one of the most important factors. The future load prediction has manually been performed by calculating the maximum electric load considering loads transfer/switching and multiplying it with the load increase rate. In here, the risk of human error is inherent and thus an automated maximum electric load forecasting system is required. Although there are many existing methods and techniques to predict future electric loads, such as regression analysis, many of them have limitations in reflecting the nonlinear characteristics of the electric load and the complexity due to Photovoltaics (PVs), Electric Vehicles (EVs), and etc. This study, therefore, proposes a method of predicting future electric loads on distribution lines by using Machine Learning (ML) method that can reflect the characteristics of these nonlinearities. In addition, predictive models were developed based on actual data collected at KEPCO's existing distribution lines and the adequacy of developed models was verified as well. Also, as the distribution planning has a direct bearing on the investment, and amount of investment has a direct bearing on the maximum electric load, various baseline such as maximum, lowest, median value that can assesses the adequacy and accuracy of proposed ML based electric load prediction methods were suggested.