• Journal of the Korea Institute of Building Construction
• /
• v.18 no.6
• /
• pp.621-631
• /
• 2018

The repair and maintenance cost of domestic public rental housing is an issue of considerable interest and growing financial concern. This paper suggests a quantity-based model as an alternative method for predicting costs, instead of the conventional model which is based on actual cost data. Furthermore, this paper provides a forecast of the repair costs incurred each year during the multi family house's maintenance phase (40 years). The recently changed the long-term repair plan and quality-improved interior materials were considered into the research. In order to estimate the cost of maintenance work, 5 sample apartments were selected and analyzed. The repair and maintenance cost from the case studies was converted to cost per household and per floor area for general use. On the other hand, the net present value method was applied to reflect the effect of time. We expect that the results will help to establish expenditure plans that are more effective for public rental housing in the maintenance stage.

• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.120-126
• /
• 2019

Recently, accurate prediction of power consumption based on machine learning techniques in Internet of Energy (IoE) has been actively studied using the large amount of electricity data acquired from advanced metering infrastructure (AMI). In this paper, we propose a deep learning model based on Gated Recurrent Unit (GRU) as an artificial intelligence (AI) network that can effectively perform pattern recognition of time series data such as the power consumption, and analyze performance of the prediction based on real household power usage data. In the performance analysis, performance comparison between the proposed GRU-based learning model and the conventional learning model of Long Short Term Memory (LSTM) is described. In the simulation results, mean squared error (MSE), mean absolute error (MAE), forecast skill score, normalized root mean square error (RMSE), and normalized mean bias error (NMBE) are used as performance evaluation indexes, and we confirm that the performance of the prediction of the proposed GRU-based learning model is greatly improved.

• The Korean Journal of Applied Statistics
• /
• v.29 no.3
• /
• pp.525-537
• /
• 2016

The paper considers a hybrid model to analyze and forecast time series data based on an empirical mode decomposition (EMD) that accommodates complex characteristics of time series such as nonstationarity and nonlinearity. We aggregate IMFs using the concept of cumulative energy to improve the interpretability of intrinsic mode functions (IMFs) from EMD. We forecast aggregated IMFs and residue with a hybrid model that combines the ARIMA model and an exponential smoothing method (ETS). The proposed method is applied to forecast KOSPI time series and is compared to traditional forecast models. Aggregated IMFs and residue provide a convenience to interpret the short, medium and long term dynamics of the KOSPI. It is also observed that the hybrid model with ARIMA and ETS is superior to traditional and other types of hybrid models.

• Journal of the Korean association of regional geographers
• /
• v.15 no.2
• /
• pp.282-291
• /
• 2009

This research predicts the spatial discord which relates with a restriction using 1915 and 2005 spatial data. In this research, difference of relative spatial density is measured and predicted the discord which relates with an environmental restriction in Pyeongchang. First, research area Pyeonchang's relative crowding degree of the building was strengthened from 1915 to 2005. When classifies a change type, formed the strong hold with new regulation and grew types and at the strong hold where contiguous concentration is progress types, general the influence weakening types and the change almost nil types. The next is the result which analyzes the long and short terms discord for the environmental restriction which is forecast from the research area. That is forecast with the fact that the discord between of development and preservation will be big with long and short terms in Jinbu-Myeon, Pyeongchang-Eup city center angles and 31 national road circumferences. And in Daegwanryeong-Myeon the discord is big short-term but with the fact that the discord will be weakened long-term. Bangrim-Myeon, now the discord is weak but the discord will be strengthened long-term. This result could be applied with fundamental data for weakening the spatial discord of the area.

• Korean Management Science Review
• /
• v.25 no.3
• /
• pp.27-43
• /
• 2008

This study analyzes the changes of Gyeongju local society because of setting up low and intermediate level radioactive waste disposal site by using urban dynamics model. Specifically, after examining 'Gyeongju Long-Term Development Plan' announced in 2007, I establish the number of industries, population, gross local product, residents' income, and the long term employment condition as essential change-causing factors in Gyeongju local society based on the Big3 government project, and forecast it by using 'Gyeongju long-Term Development Plan' and all sorts of statistical data. In this stage, I assume 3 scenarios(basic, optimistic, and pessimistic view) to estimate the changes of local society more exquisitely, and scenarios are composed through mediation about variables of a growth rate and an inflow or outflow rate. The result shows that Gyeonaju local society would have growing changes by 2020. The essential change-causing factors are as follows. The case of population is estimated that it starts going down at the level of approximately 270 thousand by 2009, starts going up continuously after 2009, the year of completion of low and intermediate level radioactive waste disposal site, and increases from the level of about 300 thousand as minimum to 340 thousand as maximum in 2020. The estimates of other cases are made that the number of Industries has about 10 thousand increases, gross local product has almost 6 trillion increases, nominal gross national income doubles, as well as residences have approximately 280 thousand increases, and also made that employment condition also improves continuously, and diffusion ratio of house starts going up but the amount of supplies is a little bit insufficient in the long view.

• Journal of Korea Water Resources Association
• /
• v.39 no.3 s.164
• /
• pp.261-274
• /
• 2006

This study performs a daily long-term runoff analysis for 30 years to forecast medium- and long-term probabilistic reservoir inflows on the Soyang River basin. Snowmelt is computed by Anderson's temperature index snowmelt model and potenetial evaporation is estimated by Penman-combination method to produce input data for a rainfall-runoff model. A semi-distributed TOPMODEL which is composed of hydrologic rainfall-runoff process on the headwater-catchment scale based on the original TOPMODEL and a hydraulic flow routing model to route the catchment outflows using by kinematic wave scheme is used in this study It can be observed that the time variations of the computed snowmelt and potential evaporation are well agreed with indirect observed data such as maximum snow depth and small pan evaporation. Model parameters are calibrated with low-flow(1979), medium-flow(1999), and high-flow(1990) rainfall-runoff events. In the model evaluation, relative volumetric error and correlation coefficient between observed and computed flows are computed to 5.64% and 0.91, respectively. Also, the relative volumetric errors decrease to 17% and 4% during March and April with or without the snowmelt model. It is concluded that the semi-distributed TOPMODEL has well performance and the snowmelt effects for the long-term runoff computation are important on the study area.

• The Korean Journal of Applied Statistics
• /
• v.29 no.7
• /
• pp.1213-1229
• /
• 2016

Cho and Shin (2016) recently demonstrated that an integrated HAR model has a forecast advantage over the HAR model of Corsi (2009). Recalling that realized volatilities of financial assets have asymmetries, we add a leverage term to the integrated HAR model, yielding the LIHAR model. Out-of-sample forecast comparisons show superiority of the LIHAR model over the HAR and IHAR models. The comparison was made for all the 20 realized volatilities in the Oxford-Man Realized Library focusing specially on the DJIA, the S&P 500, the Russell 2000, and the KOSPI. Analysis of the realized volatility data sets reveal apparent long-memory and asymmetry. The LIHAR model takes advantage of the long-memory and asymmetry and produces better forecasts than the HAR, IHAR, LHAR models.

• The Korean Journal of Applied Statistics
• /
• v.32 no.1
• /
• pp.69-81
• /
• 2019

We consider bivariate long range dependent (LRD) time series forecasting using a deep learning method. A long short-term memory (LSTM) network well-suited to time series data is applied to forecast bivariate time series; in addition, we compare the forecasting performance with bivariate fractional autoregressive integrated moving average (FARIMA) models. Out-of-sample forecasting errors are compared with various performance measures for functional MRI (fMRI) data and daily realized volatility data. The results show a subtle difference in the predicted values of the FIVARMA model and VARFIMA model. LSTM is computationally demanding due to hyper-parameter selection, but is more stable and the forecasting performance is competitively good to that of parametric long range dependent time series models.

• Proceedings of the Safety Management and Science Conference
• /
• 2006.04a
• /
• pp.471-484
• /
• 2006

The focus of analysis is effect on Non-electrical Machinery and Equipment of Macroeconomic variables through long-term and short-term periods. Also, this paper is related with implication on steady growth possibility of Non-electrical Machinery and Equipment. The period of variables is from 1985 to April in 2005. In case of not-available data is treated as missing figures. As spatial scope, these data are Non-electrical Machinery and Equipment on the basis of KSIC. In case of items, it composes MTI 1&3 digit of Korea International Trade Association (KITA), on the basis of HSK & classification of Korea Machinery industries. According to Granger causality test, yield of Cooperate Bond and export amount of Machinery have a meaning at statistical Confidence level of 10%. In case of index of the unit cost of export and export amount of Machinery, they have an interactive Granger cause. In yen dollar exchange rate and export amount of Machinery, former variable gives an unilateral Granger cause to latter that.

• Journal of the Korea Society for Simulation
• /
• v.5 no.1
• /
• pp.53-66
• /
• 1996

Container handling facilities in Korean ports have increased rapidly according to Korean industrialization and the worldwide containerization. Over 98% of total containers handles in Korean ports are handled in Puan ports. This paper presents the estimation method of annual container handling capacity of container terminals by the computer simulation models. Simulation models are developed utilizing SIMAN IV simulation package. Annual handling capacity of real container terminals such as BCTOC and PECT was estimated by the proposed simulation models. Also, Annual handling capaicty of planned or expected terminals in Puan port was estimated. The comparisons between container forecast demand and estimated handling capacity of Pusan port from 1996 through 2001 were made. It showed that Pusan port will have over two million TEU handling capacity shortage during that period and will face enormous port congestion. Lastly, mid term and long-term capacity expansion plansof container terminals in korean ports were discussed.