• The Journal of Bigdata
• /
• v.5 no.2
• /
• pp.111-120
• /
• 2020

Despite the recent economic contraction caused by the Corona 19 incident, interest in the residential environment is growing as more people live at home due to the increase in telecommuting, thereby increasing demand for remodeling. In addition, the government's real estate policy is also expected to have a visible impact on the sales of the interior and furniture industries as it shifts from regulatory policy to the expansion of housing supply. Accurate demand forecasting is a problem directly related to inventory management, and a good demand forecast can reduce logistics and inventory costs due to overproduction by eliminating the need to have unnecessary inventory. However, it is a difficult problem to predict accurate demand because external factors such as constantly changing economic trends, market trends, and social issues must be taken into account. In this study, LSTM model and 1D-CNN model were compared and analyzed by artificial intelligence-based time series analysis method to produce reliable results for manufacturers producing furniture components.

• Journal of Environmental Science International
• /
• v.32 no.8
• /
• pp.595-611
• /
• 2023

The purpose of this study is to improve the daily prediction results of PM2.5 from the air quality diagnosis and evaluation system operated by the Busan Institute of Health and Environment in real time. The air quality diagnosis and evaluation system is based on the photochemical numerical model, CMAQ (Community multiscale air quality modeling system), and includes a 3-day forecast at the end of the model's calculation. The photochemical numerical model basically has limitations because of the uncertainty of input data and simplification of physical and chemical processes. To overcome these limitations, this study applied DNN (Deep Neural Network), a deep learning technique, to the results of the numerical model. As a result of applying DNN, the r of the model was significantly improved. The r value for GFS (Global forecast system) and UM (Unified model) increased from 0.77 to 0.87 and 0.70 to 0.83, respectively. The RMSE (Root mean square error), which indicates the model's error rate, was also significantly improved (GFS: 5.01 to 6.52 ug/m³ , UM: 5.76 to 7.44 ug/m³ ). The prediction results for each concentration grade performed in the field also improved significantly (GFS: 74.4 to 80.1%, UM: 70.0 to 77.9%). In particular, it was confirmed that the improvement effect at the high concentration grade was excellent.

• Journal of Korean Society of Transportation
• /
• v.22 no.5
• /
• pp.89-97
• /
• 2004

The Korea Air Force(KAF) has operated freight flights based on the prefixed time and route schedule, which is adjusted once in a month. The major purpose of the operation of freight flights in the KAF is to distribute necessary supplies from the home air base to other air bases. The secondary purpose is to train the young pilots to get more experiences in navigation. Each freight flight starts from and returned to the home air base everyday except holidays, while it visits several other air bases to accomplish its missions. The study aims to forecast freight demand at each base by using time series analysis, and then it tried to optimize the cost of operating flights by solving vehicle routing problem. For more specifically, first, several constraints in operating cargos were defined by reviewing the Korea Air Force manuals and regulation. With such constraints, an integer programming problem was formulated for this specific routing problem allowing several visits in a tour with limitation of maximum number of visits. Then, an algorithm to solve the routing problem was developed. Second, the time series analysis method was applied to find out the freight demand at each air base from the mother air base in the next month. With the forecasted demands and the developed solution algorithm, the oprimum routes are calculated for each flight. Finally, the study compared the solved routing system by the developed algorithm with the existing routing system of the Korea Air Force. Through this comparison, the study proved that the proposed method can provide more (economically) efficient routing system than the existing system in terms of computing and monetary cost. In summary, the study suggested objective criteria for air routing plan in the KAF. It also developed the methods which could forecast properly the freight demands at each bases by using time series analysis and which could find the optimum routing which minimizes number of cargo needed. Finally, the study showed the economical savings with the optimized routing system by using real case example.

• Journal of Korea Water Resources Association
• /
• v.44 no.7
• /
• pp.523-536
• /
• 2011

The objective of this study is to develop the data driven model for the flood forecasting that are improved the problems of the existing hydrological model for flood forecasting in medium and small streams. Neuro-Fuzzy flood forecasting model which linked the Takagi-Sugeno fuzzy inference theory with neural network, that can forecast flood only by using the rainfall and flood level and discharge data without using lots of physical data that are necessary in existing hydrological rainfall-runoff model is established. The accuracy of flood forecasting using this model is determined by temporal distribution and number of used rainfall and water level as input data. So first of all, the various combinations of input data were constructed by using rainfall and water level to select optimal input data combination for applying Neuro-Fuzzy flood forecasting model. The forecasting results of each combination are compared and optimal input data combination for real-time flood forecasting is determined.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.9 no.3
• /
• pp.171-182
• /
• 2006

In this study, rainfall-runoff analysis was performed for Yongdam watershed($930km^2$) using KOWACO flood analysis model based on Storage Function Method as lumped hydrologic model and Vflo which was developed for real-time flood prediction by University of Oklahoma. The results shows that, the hydrographs of lumped and distributed model with uncalibrated parameters which estimated from physical or experimental relationship show significant biases from observed hydrographs. However, the hydrograph at Cheoncheon site from the distributed model follows the actual hydrograph to an extent that no more calibration is necessary. It encourages that distributed model can have advantages for application in real-time flood forecasting as physically based distributed hydrologic model which can construct event-independent basin parameter group.

• Journal of Korea Water Resources Association
• /
• v.43 no.2
• /
• pp.221-231
• /
• 2010

A regression water level forecasting model using data from stage and rainfall monitoring stations is developed to solve the difficulties which real-time forecasting models could not get the reliabilities by assuming future rainfall duration and intensity. The model could forecast future water levels of maximum 2 hours after using data from monitoring stations in Daejeon area. It shows stable forecasts by its maximum standard deviation is 5 cm, average standard deviations are 1~4 cm and most of coefficients of determination are larger than 0.95. It shows also more researches about the stationary of watershed which assumed in this regression method are necessary.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.3 no.3
• /
• pp.156-162
• /
• 2001

In order to monitor local climatic information, twelve automated weather stations (AWS) were installed in alpine area by the Alpine Agricultural Experiment Station, Rural Development Administration (RDA), at the field of major crop located in around highland area, and collected data from 1993 to 2000. Hourly measurements of air and soil temperature (underground 10 cm,20 cm), relative humidity, wind speed and direction, precipitation, solar radiation and leaf wetness were automatically performed and the data could be collected through a public phone line. Datalogger was selected as CR10X (Campbell scientific, LTD, USA) out of consideration for sensers' compatibility, economics, endurance and conveniences. All AWS in alpine area were combined for net work and daily climatic data were analyzed in text and graphic file by program (Chumsungdae, LTD) on 1 km $\times$ 1 km grid tell basis. In this analysis system, important multi-functionalities, monitoring and analysis of local climatic information in alpine area was emphasized. The first objective was to obtain the output of a real time data from AWS. Secondly, daily climatic normals for each grid tell were calculated from geo-statistical relationships based on the climatic records of existing weather stations as well as their topographical informations. On 1 km $\times$ 1 km grid cell basis, real time climatic data from the automated weather stations and daily climatic normals were analyzed and graphed. In the future, if several simulation models were developed and connected with this system it would be possible to precisely forecast crop growth and yield or plant disease and pest by using climatic information in alpine area.

• Korean Journal of Ecology and Environment
• /
• v.46 no.1
• /
• pp.1-9
• /
• 2013

Algal blooms in potable water supplies are becoming an increasingly prevalent and serious water quality problem around the world. In addition to precipitating taste and odor problems, blooms damage the environment, and some classes like cyanobacteria (blue-green algae) release toxins that can threaten human health, even causing death. There is a recognized need in the water industry for models that can accurately forecast in real-time algal bloom events for planning and mitigation purposes. In this study, using data for an interconnected system of rivers and reservoirs operated by a New Jersey water utility, various ANN models, including both discrete prediction and classification models, were developed and tested for forecasting counts of three different algal classes for one-week and two-weeks ahead periods. Predictor model inputs included physical, meteorological, chemical, and biological variables, and two different temporal schemes for processing inputs relative to the prediction event were used. Despite relatively limited historical data, the discrete prediction ANN models generally performed well during validation, achieving relatively high correlation coefficients, and often predicting the formation and dissipation of high algae count periods. The ANN classification models also performed well, with average classification percentages averaging 94 percent accuracy. Despite relatively limited data events, this study demonstrates that with adequate data collection, both in terms of the number of historical events and availability of important predictor variables, ANNs can provide accurate real-time forecasts of algal population counts, as well as foster increased understanding of important cause and effect relationships, which can be used to both improve monitoring programs and forecasting efforts.

• Journal of Korean Society for Geospatial Information Science
• /
• v.18 no.2
• /
• pp.21-28
• /
• 2010

Air pollution monitoring has attracted a lot of interests because it affects directly to the human life quality. The most of the current air pollution monitoring stations use the expensive and bulky instruments and are only installed in the specific area. Therefore, it is difficult to install them to as many places as people need. In this work, we constructed a low price and small size Radio Frequency(RF) sensor system to solve this problem. This system also had the measurement range similar to the ones used in the air pollution forecast systems. This system had the sensor unit to measure the air quality, the central processing unit for air quality data acquisition, the power unit to supply the power to every units, and the RF unit for the wireless transmission and reception of the data. This system was easy to install in the field. We also added a GPS unit to track the position of the RF-sensor in real time by wireless communication. For the various measurements of the air pollution, we used CO, $O_3$, $NO_2$ sensors as gas sensors and also installed a dust sensor.

• The Korean Journal of Applied Statistics
• /
• v.22 no.5
• /
• pp.1019-1032
• /
• 2009

The city gas demand data has strong seasonality. Thus, the seasonality factor is the majority for the development of forecasting model for city gas supply amounts. Also, real city gas demand amounts can be affected by other factors; weekday effect, holiday effect, the number of validity day, and the number of consumptions. We examined the degree of effective power of these factors for the city gas demand and proposed a time-series model for efficient forecasting of city gas supply. We utilize the liner regression model with autoregressive regression errors and we have excellent forecasting results using real data.