• Journal of Navigation and Port Research
• /
• v.48 no.1
• /
• pp.27-33
• /
• 2024

Following the COVID-19 pandemic, congestion within container terminals has led to a significant increase in waiting time and turnaround time for external trucks, resulting in a severe inefficiency in gate-in and gate-out operations. In response, port authorities have implemented a Vehicle Booking System (VBS) for external trucks. It is currently in a pilot operation. However, due to issues such as information sharing among stakeholders and lukewarm participation from container transport entities, its improvement effects are not pronounced. Therefore, this study proposed a deep learning-based predictive model for external trucks turnaround time as a foundational dataset for addressing problems of waiting time for external trucks' turnaround time. We experimented with the presented predictive model using actual operational data from a container terminal, verifying its predictive accuracy by comparing it with real data. Results confirmed that the proposed predictive model exhibited a high level of accuracy in its predictions.

• Smart Media Journal
• /
• v.13 no.4
• /
• pp.48-56
• /
• 2024

In the livestock industry, detecting environmental outliers and predicting data are crucial tasks. Outliers in livestock environment data, typically gathered through time-series methods, can signal rapid changes in the environment and potential unexpected epidemics. Prompt detection and response to these outliers are essential to minimize stress in livestock and reduce economic losses for farmers by early detection of epidemic conditions. This study employs two methods to experiment and compare performances in setting thresholds that define outliers in livestock environment data outlier detection. The first method is an outlier detection using Mean Squared Error (MSE), and the second is an outlier detection using a Dynamic Threshold, which analyzes variability against the average value of previous data to identify outliers. The MSE-based method demonstrated a 94.98% accuracy rate, while the Dynamic Threshold method, which uses standard deviation, showed superior performance with 99.66% accuracy.

• Computers and Concrete
• /
• v.34 no.1
• /
• pp.93-122
• /
• 2024

One well-known reason for using Fiber Reinforced Polymer (FRP) composites is to improve concrete strength and strain capacity via external confinement. Hence, various studies have been undertaken to offer a good illustration of the response of FRP-wrapped concrete for practical design intents. However, in such studies, the strength and strain of the confined concrete were predicted using regression analysis based on a limited number of test data. This study presents an approach based on artificial neural networks (ANNs) to develop models to predict the strength and strain at maximum stress enhancement of circular concrete cross-sections confined with different FRP types (Carbone, Glass, Aramid). To achieve this goal, a large test database comprising 493 axial compression experiments on FRP-confined concrete samples was compiled based on an extensive review of the published literature and used to validate the predicted artificial intelligence techniques. The ANN approach is currently thought to be the preferred learning technique because of its strong prediction effectiveness, interpretability, adaptability, and generalization. The accuracy of the developed ANN model for predicting the behavior of FRP-confined concrete is commensurate with the experimental database compiled from published literature. Statistical measures values, which indicate a better fit, were observed in all of the ANN models. Therefore, compared to existing models, it should be highlighted that the newly developed models based on FRP type are remarkably accurate.

• Nuclear Engineering and Technology
• /
• v.56 no.5
• /
• pp.1747-1753
• /
• 2024

The accurate determination of formation density and the physical properties of rocks is the most critical logging tasks which can be obtained using gamma-ray transport and detection tools. Though the simulation works published so far have considerably improved the knowledge of the parameters that govern the responses of the detectors in these tools, recent studies have found considerable differences between the results of using a conventional model of a homogeneous mixture of formation and fluid and an inhomogeneous fractured medium. It has increased concerns about the importance of the complexity of the model used for the medium in simulation works. In the present study, we have suggested two various models for the flow of the fluid in porous media and fractured rock to be used for logging purposes. For a typical gamma-gamma logging tool containing a ¹³⁷Cs source and two NaI detectors, simulated by using the MCNPX code, a simplified porous (SP) model in which the formation is filled with elongated rectangular cubes loaded with either mineral material or oil was investigated. In this model, the oil directly reaches the top of the medium and the connection between the pores is not guaranteed. In the other model, the medium is a large 3-D matrix of 1 cm³ randomly filled cubes. The designed algorithm to fill the matrix sites is so that this realistic random (RR) model provides the continuum growth of oil flow in various disordered directions and, therefore, fulfills the concerns about modeling the rock textures consist of extremely complex pore structures. For an arbitrary set of oil concentrations and various formation materials, the response of the detectors in the logging tool has been considered as a criterion to assess the effect of modeling for the distribution of pores in the formation on simulation studies. The results show that defining a RR model for describing heterogeneities of a porous medium does not effectively improve the prediction of the responses of logging tools. Taking into account the computational cost of the particle transport in the complex geometries in the Monte Carlo method, the SP model can be satisfactory for gamma-gamma logging purposes.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.44 no.2
• /
• pp.141-147
• /
• 2024

In this study, micro CT images were used to confirm the tensile fracture strength according to the pore distribution characteristics of 3D printed concrete. Unlike general specimens, concrete structures printed by 3D printing techniques have the direction of pores (voids) depending on the stacking direction and the presence of filaments contact surfaces. Accordingly, the pore distribution of 3D printed concrete specimens was analyzed through quantitative and qualitative methods, and the tensile strength by direction was analyzed through a finite element technique. It was confirmed that the pores inside the 3D printed specimen had directionality, resulting in their anisotropic behavior. This study aims to analyze the characteristics of 3D concrete printing specimen and correlate them with simulation-based mechanical properties to improve performance of 3D printed material and structure.

• Clinical and Experimental Reproductive Medicine
• /
• v.35 no.4
• /
• pp.309-317
• /
• 2008

Objectives: The aim of this study was to evaluate the usefulness of Anti-mullerian hormone (AMH) as a predictive marker for ovarian response and cycle outcome in IVF cycles. Methods: From Jan., to Aug., 2007, 111 patients undergoing IVF/ICSI stimulated by short or antagonist protocol were selected. On cycle day 3, basal serum AMH level and FSH level were measured. The correlation between basal serum AMH or FSH, and COH outcome was analyzed and IVF outcome was compared according to the AMH levels. To determine the threshold value of AMH for poor- and hyper-response, ROC curve was analyzed. Results: Serum AMH showed higher correlation coefficient (r=0.792, p<0.001) with the number of retrieved mature oocyte than serum FSH (r=-0.477, p<0.001). According to ovarian response, FSH and AMH leves showed significant differences among poor, normal, and hyperresponder. For predicting poor (${\leq}2$ oocytes) and hyperresponse (${\geq}17$ oocyets), AMH cut-off values were 0.5 ng/ml (the sensitivity 88.9% and the specificity 89.5%) and 2.5 ng/ml (sensitivity 85.7%, specificity 87.0%), respectively. According to the AMH level, patients were divided into 3 groups: low (${\leq}0.60\;ng/ml$), normal ($0.60{\sim}2.60\;ng/ml$), and high AMH (${\geq}2.60\;ng/ml$). The number of retrieved mature oocytes was significantly higher ($2.7{\pm}2.2$, $8.1{\pm}4.8$, $16.5{\pm}5.7$) and total gonadotropin dose was lower ($3530.5{\pm}1251.0$, $2957.1{\pm}1057.6$, and $2219.2{\pm}751.9\;IU$) in high AMH group (p<0.001). There was no significant difference in fertilization rates and pregnancy rates (23.8%, 34.0%, 37.5%) among the groups. Conclusions: Basal serum AMH level correlated better with the number of retrieved mature oocytes than FSH level, suggesting its usefulness for predicting ovarian response. However, IVF outcome was not significantly different according to the AMH levels. Serum AMH level presented good cut-off value for poor- or hyper-responders, therefore it could be useful in prediction of cycle cancellation, gonadotropin dose, and OHSS risk in IVF cycles.

• Journal of Intelligence and Information Systems
• /
• v.23 no.1
• /
• pp.95-108
• /
• 2017

Recently, AlphaGo which is Bakuk (Go) artificial intelligence program by Google DeepMind, had a huge victory against Lee Sedol. Many people thought that machines would not be able to win a man in Go games because the number of paths to make a one move is more than the number of atoms in the universe unlike chess, but the result was the opposite to what people predicted. After the match, artificial intelligence technology was focused as a core technology of the fourth industrial revolution and attracted attentions from various application domains. Especially, deep learning technique have been attracted as a core artificial intelligence technology used in the AlphaGo algorithm. The deep learning technique is already being applied to many problems. Especially, it shows good performance in image recognition field. In addition, it shows good performance in high dimensional data area such as voice, image and natural language, which was difficult to get good performance using existing machine learning techniques. However, in contrast, it is difficult to find deep leaning researches on traditional business data and structured data analysis. In this study, we tried to find out whether the deep learning techniques have been studied so far can be used not only for the recognition of high dimensional data but also for the binary classification problem of traditional business data analysis such as customer churn analysis, marketing response prediction, and default prediction. And we compare the performance of the deep learning techniques with that of traditional artificial neural network models. The experimental data in the paper is the telemarketing response data of a bank in Portugal. It has input variables such as age, occupation, loan status, and the number of previous telemarketing and has a binary target variable that records whether the customer intends to open an account or not. In this study, to evaluate the possibility of utilization of deep learning algorithms and techniques in binary classification problem, we compared the performance of various models using CNN, LSTM algorithm and dropout, which are widely used algorithms and techniques in deep learning, with that of MLP models which is a traditional artificial neural network model. However, since all the network design alternatives can not be tested due to the nature of the artificial neural network, the experiment was conducted based on restricted settings on the number of hidden layers, the number of neurons in the hidden layer, the number of output data (filters), and the application conditions of the dropout technique. The F1 Score was used to evaluate the performance of models to show how well the models work to classify the interesting class instead of the overall accuracy. The detail methods for applying each deep learning technique in the experiment is as follows. The CNN algorithm is a method that reads adjacent values from a specific value and recognizes the features, but it does not matter how close the distance of each business data field is because each field is usually independent. In this experiment, we set the filter size of the CNN algorithm as the number of fields to learn the whole characteristics of the data at once, and added a hidden layer to make decision based on the additional features. For the model having two LSTM layers, the input direction of the second layer is put in reversed position with first layer in order to reduce the influence from the position of each field. In the case of the dropout technique, we set the neurons to disappear with a probability of 0.5 for each hidden layer. The experimental results show that the predicted model with the highest F1 score was the CNN model using the dropout technique, and the next best model was the MLP model with two hidden layers using the dropout technique. In this study, we were able to get some findings as the experiment had proceeded. First, models using dropout techniques have a slightly more conservative prediction than those without dropout techniques, and it generally shows better performance in classification. Second, CNN models show better classification performance than MLP models. This is interesting because it has shown good performance in binary classification problems which it rarely have been applied to, as well as in the fields where it's effectiveness has been proven. Third, the LSTM algorithm seems to be unsuitable for binary classification problems because the training time is too long compared to the performance improvement. From these results, we can confirm that some of the deep learning algorithms can be applied to solve business binary classification problems.

• Journal of Astronomy and Space Sciences
• /
• v.26 no.4
• /
• pp.425-438
• /
• 2009

We reevaluate the Burton equation (Burton et al. 1975) of predicting Dst index using high quality hourly solar wind data supplied by the ACE satellite for the period from 1998 to 2006. Sixty magnetic storms with monotonously decreasing main phase are selected. In order to determine the injection term (Q) and the decay time ($\tau$) of the equation, we examine the relationships between $Dst^*$ and $VS_s$, ${\Delta}Dst^*$ and $VS_s$, and ${\Delta}Dst^*$ and $Dst^*$ during the magnetic storms. For this analysis, we take into account one hour of the propagation time from the ACE satellite to the magnetopause, and a half hour of the response time of the magnetosphere/ring current to he solar wind forcing. The injection term is found to be $Q(nT/h)\;=\;-3.56VS_s$ for $VS_s$ > 0.5mV/m and Q(nT=h) = 0 for $VB_s\;{\leq}\;0.5mV/m$. The $\tau$ (hour) is estimated as $0.060Dst^*\;+\;16.65$ for $Dst^*$ > -175nT and 6.15 hours for $Dst^*\;{\leq}\;-175nT$. Based on these empirical relationships, we predict the 60 magnetic storms and find that the correlation coefficient between the observed and predicted $Dst^*$ is 0.88. To evaluate the performance of our prediction scheme, the 60 magnetic storms are predicted again using the models by Burton et al. (1975) and O'Brien & McPherron (2000a). The correlation coefficients thus obtained are 0.85, the same value for both of the two models. In this respect, our model is slightly improved over the other two models as far as the correlation coefficients is concerned. Particularly our model does a better job than the other two models in predicting intense magnetic storms ($Dst^*\;{< \atop \sim}\;-200nT$).

• Geotechnical Engineering
• /
• v.12 no.4
• /
• pp.157-178
• /
• 1996

Current design methods for reinforced earth structures take no account of the magnitude of the strains induced in the tensile members as these are invariably manufactured from high modulus materials, such as steel, where straits are unlikely to be significant. With fabrics, however, large strains may frequently be induced and it is important to determine these to enable the stability of the structure to be assessed. In the present paper internal design method of analysis relating to the use of fabric reinforcements in reinforced earth structures for both stress and strain considerations is presented. For the internal stability analysis against rupture and pullout of the fabric reinforcements, a strain compatibility analysis procedure that considers the effects of reinforcement stiffness, relative movement between the soil and reinforcements, and compaction-induced stresses as studied by Ehrlich 8l Mitchell is used. I Bowever, the soil-reinforcement interaction is modeled by relating nonlinear elastic soil behavior to nonlinear response of the reinforcement. The soil constitutive model used is a modified vertsion of the hyperbolic soil model and compaction stress model proposed by Duncan et at., and iterative step-loading approach is used to take nonlinear soil behavior into consideration. The effects of seepage pressures are also dealt with in the proposed method of analy For purposes of assessing the strain behavior oi the fabric reinforcements, nonlinear model of hyperbolic form describing the load-extension relation of fabrics is employed. A procedure for specifying the strength characteristics of paraweb polyester fibre multicord, needle punched non-woven geotHxtile and knitted polyester geogrid is also described which may provide a more convenient procedure for incorporating the fablic properties into the prediction of fabric deformations. An attempt to define improvement in bond-linkage at the interconnecting nodes of the fabric reinforced earth stracture due to the confining stress is further made. The proposed method of analysis has been applied to estimate the maximum tensions, deformations and strains of the fabric reinforcements. The results are then compared with those of finite element analysis and experimental tests, and show in general good agreements indicating the effectiveness of the proposed method of analysis. Analytical parametric studies are also carried out to investigate the effects of relative soil-fabric reinforcement stiffness, locked-in stresses, compaction load and seepage pressures on the magnitude and variation of the fabric deformations.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.1 no.1
• /
• pp.1-11
• /
• 1999

Rice yield and primary productivity (NPP) are dependent upon the variability of climate and soil. The variability and regionality of the rice yield and net primary productivity were evaluated with the meteorological data collected from Korea Meteorology Administration and the actual rice yield data from the Ministration of Agriculture and Forestry, Korea. The estimated NPP using the three models, dependent upon temperature(NPP-T), precipitation(NPP-P) and net radiation(NPP-R), ranged from 10.87 to 17.52 Mg ha$^{-1}$ with average of 14.69 Mg ha$^{-1}$ in the South Korea and was ranged 6.47 to 15.58 Mg ha$^{-1}$ with average of 12.59 Mg ha$^{-1}$ in the North Korea. The primary limiting factor of NPP in Korea was net radiation, and the secondary limiting factor was temperature. Spectral analysis on the long term change in air temperature in July and August showed periodicity. The short periodicity was 3 to 7 years and the long periodicity was 15 to 43 years. The coefficient of variances, CV, of the rice yield from 1989 to 1998 ranged 3.23 percents to 12.37 percents which were lower than past decades. The CV's in Kangwon and Kyeongbuk were high while that in Chonbuk was the lowest. The prediction model based on th e yield index and yield response to temperature obtain ed from the field crop situation showed reasonable results and thus the spatial distributions of rice yield and predicted yield could be expressed in the maps. The predicted yields was well fitted with the actual yield except Kyungbuk. For better prediction, modification should be made considering radiation factor in further development.