• The Journal of Bigdata
• /
• v.8 no.2
• /
• pp.149-163
• /
• 2023

Due to the continuous growth of the E-commerce market, the volume of orders that fulfillment centers have to process has increased, and various customer requirements have increased the complexity of order processing. Along with this trend, the operational efficiency of fulfillment centers due to increased labor costs is becoming more important from a corporate management perspective. Using historical performance data as training data, this study focused on real-time box recommendations applicable to packaging areas during fulfillment center shipping. Four types of data, such as product information, order information, packaging information, and delivery information, were applied to the machine learning model through pre-processing and feature-engineering processes. As an input vector, three characteristics were used as product specification information: width, length, and height, the characteristics of the input vector were extracted through a feature engineering process that converts product information from real numbers to an integer system for each section. As a result of comparing the performance of each model, it was confirmed that when the Gradient Boosting model was applied, the prediction was performed with the highest accuracy at 95.2% when the product specification information was converted into integers in 21 sections. This study proposes a machine learning model as a way to reduce the increase in costs and inefficiency of box packaging time caused by incorrect box selection in the fulfillment center, and also proposes a feature engineering method to effectively extract the characteristics of product specification information.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.24 no.4
• /
• pp.218-233
• /
• 2022

The long-term (1986~2020) predictability of the number of days of heat and cold damages for each growth stage of soybean is evaluated using the daily maximum and minimum temperature (Tmax and Tmin) data produced by Pusan National University Coupled General Circulation Model (PNU CGCM)-Weather Research and Forecasting (WRF). The Predictability evaluation methods for the number of days of damages are Normalized Standard Deviations (NSD), Root Mean Square Error (RMSE), Hit Rate (HR), and Heidke Skill Score (HSS). First, we verified the simulation performance of the Tmax and Tmin, which are the variables that define the heat and cold damages of soybean. As a result, although there are some differences depending on the month starting with initial conditions from January (01RUN) to May (05RUN), the result after a systematic bias correction by the Variance Scaling method is similar to the observation compared to the bias-uncorrected one. The simulation performance for correction Tmax and Tmin from March to October is overall high in the results (ENS) averaged by applying the Simple Composite Method (SCM) from 01RUN to 05RUN. In addition, the model well simulates the regional patterns and characteristics of the number of days of heat and cold damages by according to the growth stages of soybean, compared with observations. In ENS, HR and HSS for heat damage (cold damage) of soybean have ranged from 0.45~0.75, 0.02~0.10 (0.49~0.76, -0.04~0.11) during each growth stage. In conclusion, 01RUN~05RUN and ENS of PNU CGCM-WRF Chain have the reasonable performance to predict heat and cold damages for each growth stage of soybean in South Korea.

• Korean Journal of Remote Sensing
• /
• v.37 no.2
• /
• pp.337-357
• /
• 2021

Because the growth of paddy rice is affected by meteorological factors, the selection of appropriate meteorological variables is essential to build a rice yield prediction model. This paper examines the suitability of multiple meteorological datasets for the rice yield modeling in South Korea, 1996-2019, and a hindcast experiment for rice yield using a machine learning method by considering the nonlinear relationships between meteorological variables and the rice yield. In addition to the ASOS in-situ observations, we used CRU-JRA ver. 2.1 and ERA5 reanalysis. From the multiple meteorological datasets, we extracted the four common variables (air temperature, relative humidity, solar radiation, and precipitation) and analyzed the characteristics of each data and the associations with rice yields. CRU-JRA ver. 2.1 showed an overall agreement with the other datasets. While relative humidity had a rare relationship with rice yields, solar radiation showed a somewhat high correlation with rice yields. Using the air temperature, solar radiation, and precipitation of July, August, and September, we built a random forest model for the hindcast experiments of rice yields. The model with CRU-JRA ver. 2.1 showed the best performance with a correlation coefficient of 0.772. The solar radiation in the prediction model had the most significant importance among the variables, which is in accordance with the generic agricultural knowledge. This paper has an implication for selecting from multiple meteorological datasets for rice yield modeling.

• Clean Technology
• /
• v.28 no.1
• /
• pp.79-93
• /
• 2022

In this study, an electronics industrial wastewater activated sludge model (e-ASM) to be used as a Water Digital Twin was calibrated based on real high-tech electronics industrial wastewater treatment measurements from lab-scale and pilot-scale reactors, and examined for its treatment performance, effluent quality prediction, and optimal process selection. For specialized modeling of a high-tech electronics industrial wastewater treatment system, the kinetic parameters of the e-ASM were identified by a sensitivity analysis and calibrated by the multiple response surface method (MRS). The calibrated e-ASM showed a high compatibility of more than 90% with the experimental data from the lab-scale and pilot-scale processes. Four electronics industrial wastewater treatment processes-MLE, A2/O, 4-stage MLE-MBR, and Bardenpo-MBR-were implemented with the proposed Water Digital Twin to compare their removal efficiencies according to various electronics industrial wastewater characteristics. Bardenpo-MBR stably removed more than 90% of the chemical oxygen demand (COD) and showed the highest nitrogen removal efficiency. Furthermore, a high concentration of 1,800 mg L^-1 T MAH influent could be 98% removed when the HRT of the Bardenpho-MBR process was more than 3 days. Hence, it is expected that the e-ASM in this study can be used as a Water Digital Twin platform with high compatibility in a variety of situations, including plant optimization, Water AI, and the selection of best available technology (BAT) for a sustainable high-tech electronics industry.

• Journal of IKEEE
• /
• v.27 no.1
• /
• pp.116-121
• /
• 2023

In this paper, we develop a deep learning structure for a complex microbial incubator that applies deep learning prediction result information. The proposed complex microbial incubator consists of pre-processing of complex microbial data, conversion of complex microbial data structure, design of deep learning network, learning of the designed deep learning network, and GUI development applied to the prototype. In the complex microbial data preprocessing, one-hot encoding is performed on the amount of molasses, nutrients, plant extract, salt, etc. required for microbial culture, and the maximum-minimum normalization method for the pH concentration measured as a result of the culture and the number of microbial cells to preprocess the data. In the complex microbial data structure conversion, the preprocessed data is converted into a graph structure by connecting the water temperature and the number of microbial cells, and then expressed as an adjacency matrix and attribute information to be used as input data for a deep learning network. In deep learning network design, complex microbial data is learned by designing a graph convolutional network specialized for graph structures. The designed deep learning network uses a cosine loss function to proceed with learning in the direction of minimizing the error that occurs during learning. GUI development applied to the prototype shows the target pH concentration (3.8 or less) and the number of cells (10⁸ or more) of complex microorganisms in an order suitable for culturing according to the water temperature selected by the user. In order to evaluate the performance of the proposed microbial incubator, the results of experiments conducted by authorized testing institutes showed that the average pH was 3.7 and the number of cells of complex microorganisms was 1.7 × 10⁸. Therefore, the effectiveness of the deep learning structure for the complex microbial incubator applying the deep learning prediction result information proposed in this paper was proven.

• Korean Journal of Construction Engineering and Management
• /
• v.13 no.4
• /
• pp.141-151
• /
• 2012

Infrastructure are the essential element in the country for they are the basic facilities forming the basis of economic activity. In Korea Infrastructure which are subject to the management of government are increasing annually, and subsequently the budget of maintenance costs is expected to rise significantly. For the effective management of the constructed and accumulated infrastructure, the integrated management of the future assets will be needed which includes the determination of the asset status, management subjects, and the location, the maintenance of their performance and state, and the prediction of the cost required to increase their useful lifetime. However, in the domestic cases the road facilities valuation has not been done systematically, and the preparation and research on this is scarce. Thus, the systematic procedure for the road facilities valuation is required. In this paper the following study was conducted to derive the reasonable asset valuation methods. First, the valuation methods was investigated and summarized throughout the domestic and international research literature. Second, to apply the investigated valuation methods to the road facilities the valuation process that reflects domestic conditions and characteristics has been developed. Third, a working Bridge, Highways, and General national ways were applied to the general valuation process, and the results were analyzed. As a final step a schematic diagram of the asset management support by WDRC valuation method was presented.

• Journal of Korean Society of Environmental Engineers
• /
• v.38 no.2
• /
• pp.87-95
• /
• 2016

Turbidity is a key indicator to the user that the 'Discolored Water' phenomenon known to be caused by corrosion of the pipeline in the water supply system. 'Discolored Water' is defined as a state with a turbidity of the degree to which the user visually be able to recognize water. Therefore, this study used data mining techniques in order to estimate turbidity changes in water supply system. Decision tree analysis was applied in data mining techniques to develop estimation models for turbidity changes in the water supply system. The pH and residual chlorine dataset was used as variables of the turbidity estimation model. As a result, the case of applying both variables(pH and residual chlorine) were shown more reasonable estimation results than models only using each variable. However, the estimation model developed in this study were shown to have underestimated predictions for the peak observed values. To overcome this disadvantage, a high-pass filter method was introduced as a pretreatment of estimation model. Modified model using high-pass filter method showed more exactly predictions for the peak observed values as well as improved prediction performance than the conventional model.

• Atmosphere
• /
• v.23 no.2
• /
• pp.171-186
• /
• 2013

Recently, the number of observations used in a data assimilation system is increasing due to the enormous amount of observations, including satellite data. However, it is not clear that all of these observations are always beneficial to the performance of the numerical weather prediction (NWP). Therefore, it is important to evaluate the effect of observations on these forecasts so that the observations can be used more usefully in NWP process. In this study, the adjoint-based Forecast Sensitivity to Observation (FSO) method with the KMA Unified Model (UM) is applied to two high-impact weather events which occurred in summer and winter in Korea in an effort to investigate the effects of observations on the forecasts of these events. The total dry energy norm is used as a response function to calculate the adjoint sensitivity. For the summer case, TEMP observations have the greatest total impact while BOGUS shows the greatest impact per observation for all of the 24-, 36-, and 48-hour forecasts. For the winter case, aircraft, ATOVS, and ESA have the greatest total impact for the 24-, 36-, and 48-hour forecasts respectively, while ESA has the greatest impact per observation. Most of the observation effects are horizontally located upwind or in the vicinity of the Korean peninsula. The fraction of beneficial observations is less than 50%, which is less than the results in previous studies. As an additional experiment, the total moist energy norm is used as a response function to measure the sensitivity of 24-hour forecast error to observations. The characteristics of the observation impact with the moist energy response function are generally similar to those with the dry energy response function. However, the ATOVS observations were found to be sensitive to the response function, showing a positive (a negative) effect on the forecast when using the dry (moist) norm for the summer case. For the winter case, the dry and moist energy norm experiments show very similar results because the adjoint of KMA UM does not calculate the specific humidity of ice properly such that the dry and moist energy norms are very similar except for the humidity in air that is very low in winter.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.2
• /
• pp.93-98
• /
• 2016

For accurate precipitation forecasts the choice of weather factors and prediction method is very important. Recently, machine learning has been widely used for forecasting precipitation, and artificial neural network, one of machine learning techniques, showed good performance. In this paper, we suggest a new method for forecasting precipitation using DBN, one of deep learning techniques. DBN has an advantage that initial weights are set by unsupervised learning, so this compensates for the defects of artificial neural networks. We used past precipitation, temperature, and the parameters of the sun and moon's motion as features for forecasting precipitation. The dataset consists of observation data which had been measured for 40 years from AWS in Seoul. Experiments were based on 8-fold cross validation. As a result of estimation, we got probabilities of test dataset, so threshold was used for the decision of precipitation. CSI and Bias were used for indicating the precision of precipitation. Our experimental results showed that DBN performed better than MLP.

• Journal of Digital Convergence
• /
• v.15 no.11
• /
• pp.245-250
• /
• 2017

Flash floods is defined as the flooding of intense rainfall over a relatively small area that flows through river and valley rapidly in short time with no advance warning. So that it can cause damage property and casuality. This study is to establish the flash-flood warning system using 38 accident data, reported from the National Disaster Information Center and Land Surface Model(TOPLATS) between 2009 and 2012. Three variables were used in the Land Surface Model: precipitation, soil moisture, and surface runoff. The three variables of 6 hours preceding flash flood were reduced to 3 factors through factor analysis. Decision tree, random forest, Naive Bayes, Support Vector Machine, and logistic regression model are considered as big data methods. The prediction performance was evaluated by comparison of Accuracy, Kappa, TP Rate, FP Rate and F-Measure. The best method was suggested based on reproducibility evaluation at the each points of flash flood occurrence and predicted count versus actual count using 4 years data.