• Journal of Information Technology Services
• /
• v.11 no.4
• /
• pp.233-250
• /
• 2012

In this paper, credit card delinquency means the possibility of occurring bad debt within the certain near future from the normal accounts that have no debt and the problem is to predict, on the monthly basis, the occurrence of delinquency 3 months in advance. This prediction is typical binary classification problem but suffers from the issue of data imbalance that means the instances of target class is very few. For the effective prediction of bad debt occurrence, Support Vector Machine (SVM) with kernel trick is adopted using credit card usage and payment patterns as its inputs. SVM is widely accepted in the data mining society because of its prediction accuracy and no fear of overfitting. However, it is known that SVM has the limitation in its ability to processing the large-scale data. To resolve the difficulties in applying SVM to bad debt occurrence prediction, two stage clustering is suggested as an effective data reduction method and ensembles of SVM models are also adopted to mitigate the difficulty due to data imbalance intrinsic to the target problem of this paper. In the experiments with the real world data from one of the major domestic credit card companies, the suggested approach reveals the superior prediction accuracy to the traditional data mining approaches that use neural networks, decision trees or logistics regressions. SVM ensemble model learned from T2 training set shows the best prediction results among the alternatives considered and it is noteworthy that the performance of neural networks with T2 is better than that of SVM with T1. These results prove that the suggested approach is very effective for both SVM training and the classification problem of data imbalance.

• Journal of Convergence for Information Technology
• /
• v.9 no.1
• /
• pp.28-37
• /
• 2019

Various fields are being combined and developed for the development of the country and industry. The purpose of this study is to examine the effects of interdisciplinary research on the technology commercialization of researchers in the national R&D project. For the purpose of empirical analysis, major research and development have been reviewed. Major research and development projects conducted by government-funded research institution in which various technological fields are integrated in order to commercialize technology. The research data is based on the major R&D project data for railway/public transportation/logistics fields technology in 2015-17, and 149 project were selected for analysis. As a result, it was found that diversity of participants' experts had significant positive effect on technology commercialization when R&D expense, The number of research participants, The number of participating organizations, TRL before R&D were controlled. It is expected that the results of this study will be used meaningfully when the R&D program plan, analysing the academic convergence factor of research topic.

• Journal of the Economic Geographical Society of Korea
• /
• v.21 no.4
• /
• pp.409-424
• /
• 2018

Since the 1950s, many of the traditional industrial cities of advanced economies in Europe and North America were affected by a series of de-industrialization. The de-industrialization process, characterized by company shut-downs and massive lay-offs, has resulted in high unemployment rates and massive redundancies in physical infrastructure. Since the 1980s, many of the old industrial cities have attempted to overcome such problems. However, it has been found that not many of the cities are found to be successful. The City of Dortmund, one of the core cities of the large German industrial conurbation of the past, the Ruhr, is found to be an exceptional case demonstrating a clear success in overcoming deindustrialization problems. The City in fact strategically pursued transforming backbone of its economy from steel-making, coal-mining and beer-brewery to high-technology and future-oriented industries, based on microsystems, biomedical, electronic logistics and information technology. This paper attempts to analyse the processes and outcomes of transforming Dortmund beginning from the 1980s to articulate the roles of the agencies contributing to the success.

• Journal of the Korea Society for Simulation
• /
• v.28 no.1
• /
• pp.1-9
• /
• 2019

Recently, as the importance of warehouses has increased, much efforts are being made to improve the picking process in order to cope with a small amount of high frequency and fast delivery. This study proposes an algorithm to assign orders to pickers in the situation where Single Order Picking policy is used. This algorithm utilizes five attributes related to picking such as picking processing time, elapsed time after receipt of order, inspection/packing workstation situation, picker error, customer importance. A measure of urgency is introduced so that the units of measure for each attribute are the same. The higher the urgency, the higher the allocation priority. In the proposed algorithm, the allocation policy can be flexibly adjusted according to the operational goal of the picking system by changing the weight of each attribute. Simulation experiments were performed on a hypothetical small logistics warehouse. The results showed excellent performance in terms of system throughput and flow time.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.2
• /
• pp.140-152
• /
• 2021

During the shift from gasoline vehicles to electric ones, auto parts manufacturing companies have realized the importance of improvement in the manufacturing process that does not require any layout changes nor extra investments, while maintaining their current production rate. Due to these reasons, for the auto part manufacturing company, I-company, this study has developed the simulation model of the PUSH system to conduct a process analysis in terms of production rate, WIP level, and logistics work's utilization rate. In addition, this study compares the PUSH system with other three manufacturing systems -KANBAN, DBR, and CONWIP- to compare the performance of these production systems, while satisfying the company's target production rate. With respect to lead-time, the simulation results show that the improvement of 77.90% for the KANBAN system, 40.39% for the CONWIP system, and 69.81% for the DBR system compared to the PUSH system. In addition, with respect to WIP level, the experimental results demonstrate that the improvement of 77.91% for the KANBAN system, 40.41% for the CONWIP system, and 69.82% for the DBR system compared to the PUSH system. Since the KANBAN system has the largest impacts on the reduction of the lead-time and WIP level compared to other production systems, this study recommends the KANBAN system as the proper manufacturing system of the target company. This study also shows that the proper size of moving units is four and the priority allocation of bottleneck process methods improves the target company's WIP and lead-time. Based on the results of this study, the adoption of the KANBAN system will significantly improve the production process of the target company in terms of lead-time and WIP level.

• Journal of Korea Port Economic Association
• /
• v.37 no.3
• /
• pp.75-91
• /
• 2021

Due to increased competition in supply side to reduce operational costs, port professionals have experienced extreme pressure, which demanded academicians to develop the model for efficient port operations from the industry perspective. Among many ports in the world, U.S. ports are our primary interest to analyze in our study for its high volume of cargoes transacted in the U.S. ports. We primarily employed DEA (Data Envelopment Analysis) technique to research the productivity of U.S. ports and applied the algorithm of double bootstrapped DEA proposed by Simar & Wilson (2007) to further investigate the driving forces of the performance of U.S. port operations. The external variables employed in our study comprise onDock Rail, Channel Depth, Location, Area, Acres, ForeignCargoRatio, and TEUChange, out of which onDock Rail, Acres, ForeignCargoRatio, and TEUChange were significant. In order to evaluate the effects of methodology selection, we conducted the same analysis applying the Censored model (Tobit) and contrasted the outcomes drawn from the two different techniques. Based on the findings from this work we proposed managerial implications and concluded.

• Journal of Advanced Navigation Technology
• /
• v.28 no.4
• /
• pp.518-523
• /
• 2024

The explosive increase in demand for drones poses a major threat to the safety of existing aircraft operations and important national facilities operating in low-altitude airspace. In order to determine the type and degree of safety threats for low-altitude airspace drones, the types and types of threats from drones are evaluated through analysis of AHP(analysis hierarchy process) for aviation workers in each field. The composition of the threat factor hierarchy from drones was designed using a specific operation risk assessment (SORA) technique previously studied by the European Aviation Safety Agency (EASA), an advanced aviation country. Based on this, it will be possible to secure the low-altitude safety operation of existing aircraft by identifying and removing prior hazards between each aircraft operation and mission performance.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.5
• /
• pp.35-44
• /
• 2021

In recent times, several studies have been conducted focusing on crashes occurring on the main segment of the highway. However, there is a dearth of research dealing with traffic safety relating to other highway facilities, especially ramp areas. According to the Korea Expressway Corporation's Expressway Information Service, 6,717 crashes have occurred on ramps in the five years from 2015~2019, which accounts for about 15% of all highway accidents. In this study, the simple and full safety performance functions (SPFs) were evaluated and explored using different statistical distributions (i.e., Poisson Gamma (PG) and Poisson Inverse Gaussian (PIG)) and techniques (i.e., fixed effects (FE) and random effects (RE)) to provide more accurate crash prediction models for highway ramp sections. Data on the geometric characteristics of traffic and roadways were collected from various systems and with extensive efforts using a street-view application. The results showed that the PIG models present more accurate crash predictions in general. The results also indicated that the RE models performed better than FE models for simple and full SPFs. The findings from this study offer transportation practitioners using the Korea Expressway Corporation's Expressway a dependable reference to enhance and understand traffic safety in ramp areas based on accurate crash prediction models and empirical evidence.

• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.23-45
• /
• 2020

Big data is creating in a wide variety of fields such as medical care, manufacturing, logistics, sales site, SNS, and the dataset characteristics are also diverse. In order to secure the competitiveness of companies, it is necessary to improve decision-making capacity using a classification algorithm. However, most of them do not have sufficient knowledge on what kind of classification algorithm is appropriate for a specific problem area. In other words, determining which classification algorithm is appropriate depending on the characteristics of the dataset was has been a task that required expertise and effort. This is because the relationship between the characteristics of datasets (called meta-features) and the performance of classification algorithms has not been fully understood. Moreover, there has been little research on meta-features reflecting the characteristics of multi-class. Therefore, the purpose of this study is to empirically analyze whether meta-features of multi-class datasets have a significant effect on the performance of classification algorithms. In this study, meta-features of multi-class datasets were identified into two factors, (the data structure and the data complexity,) and seven representative meta-features were selected. Among those, we included the Herfindahl-Hirschman Index (HHI), originally a market concentration measurement index, in the meta-features to replace IR(Imbalanced Ratio). Also, we developed a new index called Reverse ReLU Silhouette Score into the meta-feature set. Among the UCI Machine Learning Repository data, six representative datasets (Balance Scale, PageBlocks, Car Evaluation, User Knowledge-Modeling, Wine Quality(red), Contraceptive Method Choice) were selected. The class of each dataset was classified by using the classification algorithms (KNN, Logistic Regression, Nave Bayes, Random Forest, and SVM) selected in the study. For each dataset, we applied 10-fold cross validation method. 10% to 100% oversampling method is applied for each fold and meta-features of the dataset is measured. The meta-features selected are HHI, Number of Classes, Number of Features, Entropy, Reverse ReLU Silhouette Score, Nonlinearity of Linear Classifier, Hub Score. F1-score was selected as the dependent variable. As a result, the results of this study showed that the six meta-features including Reverse ReLU Silhouette Score and HHI proposed in this study have a significant effect on the classification performance. (1) The meta-features HHI proposed in this study was significant in the classification performance. (2) The number of variables has a significant effect on the classification performance, unlike the number of classes, but it has a positive effect. (3) The number of classes has a negative effect on the performance of classification. (4) Entropy has a significant effect on the performance of classification. (5) The Reverse ReLU Silhouette Score also significantly affects the classification performance at a significant level of 0.01. (6) The nonlinearity of linear classifiers has a significant negative effect on classification performance. In addition, the results of the analysis by the classification algorithms were also consistent. In the regression analysis by classification algorithm, Naïve Bayes algorithm does not have a significant effect on the number of variables unlike other classification algorithms. This study has two theoretical contributions: (1) two new meta-features (HHI, Reverse ReLU Silhouette score) was proved to be significant. (2) The effects of data characteristics on the performance of classification were investigated using meta-features. The practical contribution points (1) can be utilized in the development of classification algorithm recommendation system according to the characteristics of datasets. (2) Many data scientists are often testing by adjusting the parameters of the algorithm to find the optimal algorithm for the situation because the characteristics of the data are different. In this process, excessive waste of resources occurs due to hardware, cost, time, and manpower. This study is expected to be useful for machine learning, data mining researchers, practitioners, and machine learning-based system developers. The composition of this study consists of introduction, related research, research model, experiment, conclusion and discussion.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.5
• /
• pp.1-9
• /
• 2017

Recently, the active research into wireless sensor networks has led to the development of sensor nodes with improved performance, including their mobility and location awareness. One of the most important goals of such sensor networks is to transmit the data generated by mobile sensors nodes. Since these sensor nodes move in the mobile wireless sensor networks (MWSNs), the energy consumption required for them to transmit the sensed data to the fixed sink is increased. In order to solve this problem, the use of mobile sinks to collect the data while moving inside the network is studied herein. The important issues are the mobility and energy consumption in MWSNs. Because of the sensor nodes' limited energy, their energy consumption for data transmission affects the lifetime of the network. In this paper, a mobile-sink based energy-efficient clustering scheme is proposed for use in mobile wireless sensor networks (MECMs). The proposed scheme improves the energy efficiency when selecting a new cluster head according to the mobility of the mobile sensor nodes. In order to take into consideration the mobility problem, this method divides the entire network into several cluster groups based on mobile sinks, thereby decreasing the overall energy consumption. Through both analysis and simulation, it was shown that the proposed MECM is better than previous clustering methods in mobile sensor networks from the viewpoint of the network energy efficiency.