• Journal of the Korean Society of Marine Environment & Safety
• /
• v.27 no.5
• /
• pp.574-583
• /
• 2021

Predicting shipping markets is an important issue. Such predictions form the basis for decisions on investment methods, fleet formation methods, freight rates, etc., which greatly affect the profits and survival of a company. To this end, in this study, we propose a shipping freight rate prediction model for container ships using gated recurrent units (GRUs) and long short-term memory structure. The target of our freight rate prediction is the China Container Freight Index (CCFI), and CCFI data from March 2003 to May 2020 were used for training. The CCFI after June 2020 was first predicted according to each model and then compared and analyzed with the actual CCFI. For the experimental model, a total of six models were designed according to the hyperparameter settings. Additionally, the ARIMA model was included in the experiment for performance comparison with the traditional analysis method. The optimal model was selected based on two evaluation methods. The first evaluation method selects the model with the smallest average value of the root mean square error (RMSE) obtained by repeating each model 10 times. The second method selects the model with the lowest RMSE in all experiments. The experimental results revealed not only the improved accuracy of the deep learning model compared to the traditional time series prediction model, ARIMA, but also the contribution in enhancing the risk management ability of freight fluctuations through deep learning models. On the contrary, in the event of sudden changes in freight owing to the effects of external factors such as the Covid-19 pandemic, the accuracy of the forecasting model reduced. The GRU1 model recorded the lowest RMSE (69.55, 49.35) in both evaluation methods, and it was selected as the optimal model.

• Journal of Energy Engineering
• /
• v.28 no.2
• /
• pp.1-17
• /
• 2019

This article analyzes causal relationships among gross domestic product(GDP), electricity consumption, carbon dioxide($CO_2$) emission and foreign direct investments(FDI) inflow of Korea over the period from 1976 to 2014, using unit root test, cointegration test, and vector error correction model(VECM). As the results, this article found (1) a long-run bi-directional causality between GDP and electricity consumption, which may imply a negative impact of electricity consumption-saving policy on economic growth, (2) uni-directional short- and long-run causalities running from $CO_2$ emission to GDP, and a uni-directional long-run causality running from $CO_2$ emission to electricity consumption, which can result in a negative impact of $CO_2$ emission reduction policy on economic growth and electricity consumption, (3) a uni-directional long-run causality running from FDI to GDP, and uni-directional short- and long-run causalities running from FDI to electricity consumption, which may result from relatively lower electricity prices than investing countries, (4) no causality between FDI and $CO_2$ emission, which is based on the characteristics of FDI composed of service industries. Considering the above causal relationships among the four variables, the policy implication needs to focus on the electricity demand management based on the relevant R&Ds, and on the gradual transition from fossil fuel- to renewable-energy. Adaptive policy to increase the FDI inflow is also needed.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.6
• /
• pp.453-459
• /
• 2021

The system marginal price of electricity is the amount paid to all the generating units, which is an important decision-making factor for the construction and maintenance of an electrical power unit. In this paper, we suggest a long-term forecasting model for calculating the system marginal price based on prices of natural gas and oil. As most variables used in the analysis are nonstationary time series, the long run relationship among the variables should be examined by cointegration tests. The forecasting model is similar to an error correction model which consists of a long run cointegrating equation and another equation for short run dynamics. To mitigate the robustness issue arising from the relatively small data sample, this study employs various testing and estimating methods. Compared to previous studies, this paper considers multiple fuel prices in the forecasting model of system marginal price, and provides greater emphasis on the robustness of analysis. As none of the cointegrating relations associated with system marginal price, natural gas price and oil price are excluded, three error correction models are estimated. Considering the root mean squared error and mean absolute error, the model based on the cointegrating relation between system marginal price and natural gas price performs best in the out-of-sample forecast.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.6
• /
• pp.188-196
• /
• 2021

For the tactical vehicles operated by the Korean army, the hub-reduction portal axle was applied considering Korea's topographical characteristics. Hub-reduction was applied to a Korean military vehicle to increase the vehicle body to secure ground clearance and improve the driving capability on rough roads, such as unpaved and field land by increasing the torque. The Korean military is operating tactical vehicles after various performance tests, including durability driving, but wheel damage occurred in one of the vehicles operating in the front units. Failure analysis revealed many damaged parts, including the hub, making it difficult to determine the cause. Therefore, an analysis of the failure occurrence mechanism for each damaged part was conducted, which confirmed that the cause of wheel breakage was a hub. Furthermore, the root cause of the hub breakage was a crack due to internal pores and foreign matters. In addition, a realistic improvement plan that can be applied throughout the design, manufacture, and shipping stages was presented using the fishbone diagram analysis. The derived improvement plan was verified through unit performance tests, including CAE and actual vehicle tests, and by reflecting this, the driving safety of Korean tactical vehicles was improved. Finally, it is expected that the proposed method for analyzing the failure occurrence mechanism will be used as reference material when analyzing the quality problems of similar military vehicles in the future.

• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.3
• /
• pp.71-80
• /
• 2021

The selection of an appropriate neural network algorithm is an important step for accurate data prediction in machine learning. Many algorithms based on basic artificial neural networks have been devised to efficiently predict future data. These networks include deep neural networks (DNNs), recurrent neural networks (RNNs), long short-term memory (LSTM) networks, and gated recurrent unit (GRU) neural networks. Developers face difficulties when choosing among these networks because sufficient information on their performance is unavailable. To alleviate this difficulty, we evaluated the performance of each algorithm by comparing their errors and processing times. Each neural network model was trained using a tax dataset, and the trained model was used for data prediction to compare accuracies among the various algorithms. Furthermore, the effects of activation functions and various optimizers on the performance of the models were analyzed The experimental results show that the GRU and LSTM algorithms yields the lowest prediction error with an average RMSE of 0.12 and an average R2 score of 0.78 and 0.75 respectively, and the basic DNN model achieves the lowest processing time but highest average RMSE of 0.163. Furthermore, the Adam optimizer yields the best performance (with DNN, GRU, and LSTM) in terms of error and the worst performance in terms of processing time. The findings of this study are thus expected to be useful for scientists and developers.

• The Journal of the Korea Contents Association
• /
• v.22 no.1
• /
• pp.214-223
• /
• 2022

The purpose of this study is to present a new industrial land demand prediction method that can consider external economic factors. The analysis model used ARIMA-X, which can consider exogenous variables. Exogenous variables are composed of macroeconomic variable, Business Survey Index, and Composite Economic Index variables to reflect the economic and industrial structure. And, among the exogenous variables, only variables that precede the supply of industrial land are used for prediction. Variables with precedence in the supply of industrial land were found to be import, private and government consumption expenditure, total capital formation, economic sentiment index, producer's shipment index, machinery for domestic demand and composite leading index. As a result of estimating the ARIMA-X model using these variables, the ARIMA-X(1,1,0) model including only the import was found to be statistically significant. The industrial land demand forecast predicted the industrial land from 2021 to 2030 by reflecting the scenario of change in import. As a result, the future demand for industrial land was predicted to increase by 1.91% annually to 1,030.79 km². As a result of comparing these results with the existing exponential smoothing method, the results of this study were found to be more suitable than the existing models. It is expected to b available as a new industrial land forecasting model.

• The Korea Journal of Herbology
• /
• v.27 no.6
• /
• pp.37-42
• /
• 2012

Objectives : Angelica Gigantis Radix is prescribed as the root of different Angelica species on the pharmacopoeia in Korea, Japan and China. Chemical components and their biological activities were also different according to their species. A study for the development of simple method to compare Angelica roots was needed. In order to classify them, the methods such as DNA sequencing analysis and taste sensor were applied to three Angelica species like Angelica gigas, Angelica acutiloba and Angelica sinensis. Methods : PCR amplification of intergenic transcribed spacer (ITS) region was performed using ITS1 and ITS4 primer from nine Angelica roots, and then nucleotide sequence was determined. Taste pattern of samples were measured using the taste-sensing system SA402B equipped with a sensing unit, which consists of artificial lipid membrane sensor probes of anionic bitterness, astringency, saltiness, umami, and cationic bitterness (C00, AE1, CT0, AAE, and AN0, respectively). Results : As a result of comparing the similarity of the ITS region sequences, A. sinensis was discriminated from the others (A. gigas and A. acutiloba). Equally this genetic result, A. gigas and A. acutiloba showed similar taste pattern as compared to A. sinensis. Sourness, bitterness, aftertaste of bitterness, astringency, and aftertaste of astringency of A. sinensis were significantly high as compared with A. gigas and A. acutiloba. In contrast, richness was significantly low. Conclusions : These taste pattern can be used as a way of comparison of Angelica species and this technic could be applied to establish a taste pattern marker for standardization of herbs in various purposes.

• Journal of Technology Innovation
• /
• v.31 no.2
• /
• pp.59-82
• /
• 2023

The purpose of this research is to conduct the empirical analysis of the short- and long-term causal relationship between public R&D investment, corporate R&D investment, and university R&D investment on economic growth in Korea. To this end, based on the time series data from 1976 to 2020, a causality test was conducted through the unit root test, cointegration test, and vector error correction model (VECM). As a result, it was found that there is a long-run equilibrium relationship between economic growth in Korea, public R&D investment, corporate R&D investment, and university R&D investment, in which a causal relationship exists in the long run. Also, while public R&D investment has a short-term effect on economic growth, corporate and university R&D investment does not have a short-term effect on economic growth. In addition, the results shows that there is a bidirectional causal relationship between economic growth and public R&D investment, corporate R&D investment and public R&D investment, and university R&D investment and public R&D investment in the short term. Through this research, it was empirically found that a highly mutual relationship exists between public R&D investment, corporate R&D investment, university R&D investment and economic growth. In order to increase the ripple effect of R&D investment on economic growth in the future, R&D investment between universities and corporations should be mutually promoted, and R&D investment by corporations should have a positive effect on public R&D investment so that public R&D investment can contribute to future economic growth.

• Land and Housing Review
• /
• v.14 no.3
• /
• pp.55-75
• /
• 2023

Based on prior studies on real estate policy, tax policy, and financial policy, this study examined how tax policy and financial policy affected real estate prices using monthly data from January 2014 to December 2021. We performed a VAR model using unit root tests, cointegration tests, as well as conducted impulse response analysis and variance decomposition analysis. The results are as follows. First, the tax regulation index and the financial regulation index had no discernible impact on housing prices. Specifically, a one-sided stabilizing regulatory policy was ineffective and, instead, led to unintended side effects, such as price increases resulting from reduced transaction volume. Secondly, mortgage rates had a negative impact on the housing sale price index. In other words, an increase in interest rates might led to a decrease in housing prices. Thirdly, an increase in the transfer difference, which involves capital gains tax, has a positive effect on housing prices. This led to rising housing prices because the transfer taxes were shifted to buyers, causing them to hesitate to make purchases due to the increased tax burden. Fourthly, both acquisition taxes and mortgage loans had relatively little impact on housing prices.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.5A
• /
• pp.737-746
• /
• 2008

Recently, NDTs (Non-Destructive Techniques) using electromagnetic(EM) properties are applied to the performance evaluation for RC (Reinforced Concrete) structures. Since nonmetallic materials which are cement-based system have their unique dielectric constant and conductivity, they can be characterized and changed with different mixture conditions like W/C (water to cement) ratios and unit cement weight. In a room condition, cement mortar is generally dry so that porosity plays a major role in EM properties, which is determined at early-aged stage and also be affected by curing condition. In this paper, EM properties (dielectric constant and conductivity) in cement mortar specimens with 4 different W/C ratios are measured in the wide region of 0.2 GHz~20 GHz. Each specimen has different submerged curing period from 0 to 28 days and then EM measurement is performed after 4 weeks. Furthermore, porosity at the age of 28 days is measured through MIP (Mercury Intrusion Porosimeter) and saturation is also measured through amount of water loss in room condition. In order to evaluate the porosity from the initial curing stage, numerical analysis based on the modeling for the behavior in early-aged concrete is performed and the calculated results of porosity and measured EM properties are analyzed. For the convenient comparison with influencing parameters like W/C ratios and curing period, EM properties from 5 GHz to 15 GHz are averaged as one value. For 4 weeks, the averaged dielectric constant and conductivity in cement mortar are linearly decrease with higher W/C ratios and they increase in proportion to the square root of curing period regardless of W/C ratios.