• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.7
• /
• pp.846-851
• /
• 2005

In this paper, we propose Genetic Algorithms(GAs)-based Optimized Polynomial Neural Networks(PNN). The proposed algorithm is based on Group Method of Data Handling(GMDH) method and its structure is similar to feedforward Neural Networks. But the structure of PNN is not fixed like in conventional neural networks and can be generated in a dynamic manner. As each node of PNN structure, we use several types of high-order polynomial such as linear, quadratic and modified quadratic, and it is connected as various kinds of multi-variable inputs. The conventional PNN depends on the experience of a designer that select the number of input variables, input variable and polynomial type. Therefore it is very difficult to organize optimized network. The proposed algorithm leads to identify and select the number of input variables, input variable and polynomial type by using Genetic Algorithms(GAs). The aggregate performance index with weighting factor is proposed as well. The study is illustrated with tile NOx omission process data of gas turbine power plant for application to nonlinear process. In the sequel the proposed model shows not only superb predictability but also high accuracy in comparison to the existing intelligent models.

• Journal of Korean Society of Forest Science
• /
• v.90 no.3
• /
• pp.331-337
• /
• 2001

This study was performed to develope stem analysis program(Stemwin1.0) which can be used in PC with MS-Windows operating system. Stemwin1.0 uses width of annual tree ring measured with 1/100mm unit, and calculate increments of several growth factors such as DBH, height and volume with various methods. Mean DBH can be calculated by arithmetic and quadratic mean methods. Height can be estimated by parallel line, line extending and height curve methods. Volume can be estimated by Huber, Smalian, and Spline functions. Not only Total growth, Mean Annual Increment(MAI) and Current Annual Increment(CAI) of growth factors, but also merchantable volume and height, form factor, growth rate, and merchantable volume rate are automatically calculated. Stemwin1.0 can also output accurate stem taper curve with various scale, and prepare stem taper data(diameter at different disk heights) for statistical analysis for deriving stem taper model. Stemwin1.0 can export output data and graph to Excel for more compatible use of it.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.23 no.5
• /
• pp.383-392
• /
• 2011

The basic research of the estuarine circulation at Gyeong-Gi bay has not been well studied up to now, although coastal development pressures have been continuously increased. To understand the oceanographic phenomena at the Han River estuary, it's essential to understand the propagation characteristic of tidal wave which is the strongest external forcing in this region. In this study, we investigate the tidal wave propagation characteristics along the 3 major channels using observation data and numerical model. It is found that 3 channels are all hyper-synchronous and the most important physical factor controlling the tidal wave propagation is topographical convergence of estuary shape and friction. The result of analytic solution at ideal channel considering the topographical convergence and friction show that the contribution of physical role of convergence and friction varies at 3 different channel. And the ratio of convergence and friction at Yeomha channel is four times larger than Seokmo channel. Because of this effect, the location of maximum amplitude at Yeomha channel is showed up downward than Seokmo channel. The ratio of decreasing amplitude and increasing phase per unit distance between stations is bigger than Seokmo channel. Although 3 major channel show a hyper-synchronous pattern, Yeomha shows more frictionally dominant channel and Seokmo channel is more dominantly affected by convergence effect.

• Journal of Distribution Science
• /
• v.13 no.5
• /
• pp.45-52
• /
• 2015

Purpose - Augmented reality (AR) content used in mobile media today can accommodate a wide variety of contextual information. This indicates that making people experience a sense of presence and flow is a very significant factor in augmented reality content. Flow represents a rich immersion potential as representing the progress of emotion and the means to facilitate the operation of the smart phone. Therefore, users will have friendly relational continuance behavior with products and brands that supply this experience. Based on that, the purpose of this study is to investigate the relationships among smart phone AR application traits, presence, flow experience, and relational continuance behavior. First, AR application traits are defined as three categories sensory immersion, navigation, and manipulation, based on preceding studies. This study then examines the influence of AR application traits on the presence and flow experience and looks into the relation among presence, flow experience, and relational continuance behavior. This analysis suggests more detailed and concentrated strategic implications. Research design, data, and methodology - A research model is designed to examine the relation among AR application traits, presence, flow experience, and relational continued behavior. For data collection, questionnaire surveys were composed of multi-items for each component and the direct interview method was used for the interviews. To collect the data, after running the smart phone AR applications, the consumer behaviors of the respondents were generally determined. The questionnaire surveys were conducted for one month, October 2014. A total of 300 questionnaires were distributed with 278 questionnaires used for analysis, excluding the unanswered and insincere questionnaires. The data were analyzed using SPSS ver. 20.0 and LISREL ver. 8.51. Results - The following results are found: First, AR application traits have a significantly positive effect on presence with sensory immersion, navigation, and manipulation all having a significantly positive effect. Second, sensory immersion and manipulation among the AR application traits have a significantly positive effect on flow. However, navigation did not have a significantly positive effect on flow. Third, presence has a significantly positive effect on flow and has a significantly positive effect on relational continuance behavior. Moreover, flow also has a significantly positive effect on relational continuance behavior. This behavior tends to be formed since brands want to encourage relational continuance behavior and positive emotions with the brands being used. Relational continuance behavior accompanies repeat purchasing, positive word-of-mouth and recommendation activities, and forms of trust with the brand. Conclusions - The research results showed that smart phone AR traits had significantly positive effect on presence, flow, and relational continuance behavior. Based on this, smart phone AR application providers should establish an aggressive marketing strategy to accommodate more realistic problems in order to positively influence user behavior. Additionally, the marketers should make efforts to provide fun or convenience in the AR application operation process of the user.

• Journal of Distribution Science
• /
• v.13 no.7
• /
• pp.73-83
• /
• 2015

Purpose - Recently, many large retailers have been frequently purchasing commercial brands. Not only the quality of products purchased but also the brand images are of concern in this process. Based on this rising trend, commercial brands have become an important issue in the retail business world, along with increasing general interest as well. Thus, this study focuses on the factors affecting commercial brand images and clarifies the impact of the resulting factors as well. First, store images and familiarity, price sensitiveness, and knowledge are presented as the effect factors for commercial brands. Second, the study tries to clarify the effect of commercial brand image on the reliability and loyalty of customers. Research design, data, and methodology - To conduct the study methodology, 250 questionnaires were distributed to retailers who have used large discount stores located in Busan to purchase a commercial brand a total of 234 valid questionnaires were used in the final analysis. To verify the hypotheses, a structure equation formula using Amos 20.0 was calculated. First, prior to the verification of hypotheses, the reliability and feasibility of the questions were tested, and as a result, the value of Cronbach's alpha was higher than 0.7, thereby showing reliability. Additionally, for the verification of the feasibility of the questions, a confirmation factor analysis was implemented. Results - First, variances such as store images, familiarity, price sensitivity, and knowledge were presented as the factors of effect on commercial brands. As a result of the hypotheses verification, all the effect factors presented in this study were confirmed as important variables of commercial brand images. The hypotheses were drawn based on the studies relating to existing commercial brands, and thereby, a result similar to previous studies was produced. However, the effect factors of the commercial brand image presented in this study shaped a new study model based on the previous studies. Second, it was found that commercial brand images had positive effects on the reliability and loyalty of customers. This study presented reliability and loyalty as the resulting factors of commercial brand image because they are the most important factors in relation to customer behavior and variables that can suggest marketing points for distribution businesses. Conclusions - This study focused on clarifying the factors that are important effect factors for commercial brand images. As a result, all the hypotheses were confirmed in this study, indicating a meaningful result, and thereby suggesting many points that can be presented to distribution businesses. First, the proper strategies should be developed based on the hypothesis that a store image, the familiarity of customers with brands, and the price sensitiveness and knowledge level of customers have an important effect on the choice of commercial brands. Additionally, the formation of such favorable images will have positive effects not only in terms of customer trust in the commercial brands but also in terms of their loyalty.

• Journal of Distribution Science
• /
• v.13 no.5
• /
• pp.103-111
• /
• 2015

Purpose - When it comes to first impressions of family restaurants, people basically think of the following factors: cozy, a refreshing space, indoor lighting, atmospheric music, the best-quality service through mutual interactions with employees, birthday events, follow-up customer management, brand image, and many others. Focusing on such family restaurants characteristics, with consumption experience and customer satisfaction determined as mediator variables, this study looks into the selection factors that affect customer loyalty when choosing a family restaurant. Subsequently, the study conducts an empirical analysis of these selection factors with the greatest influence. Research design, data, and methodology - This study analyzed 200 total surveys of family restaurant customers including college students, workers, housewives, and others. The survey was conducted over two weeks in May 2014. For statistical processing, SPSS 21 was used for factor analysis and reliability analysis as well as AMOS for structural equation analysis. In terms of a variable that would influence either consumption experience or customer satisfaction at the restaurants, the study introduced the concept of customer equity. The study designed a model based on a hypothesis developed using the following factors: customer encounter service value, customer convenience value, brand equity, relationship equity, and other factors defined as components of customer equity by Rust et al. (2000) determined as independent variables as well as with customer experience and customer satisfaction set as mediator variables. The hypothesis, after all, assumed that customer loyalty would be positively (+) influenced. Results - The analysis results from the study reported that brand value, customer encounter service value, and relationship value have positive (+) effects on consumption experience and customer satisfaction. However, as for customer convenience value, this turned out not to have a statistically significant influence on consumption experience and customer satisfaction. In regard to a relation between consumption experience and customer satisfaction, the results showed that the former affects the latter, and in terms of intervening variables, both of these appeared to have a positive (+) effect on customer loyalty. Conclusions - This study suggests that family restaurants should look for strategies that focus on how to promote customer relationships. For example, restaurants could try birthday events or discount events to achieve this goal. Since the study ended up realizing relatively low levels of suitability and statistical significance, the path of the study changed. After all, in order to increase suitability and statistical significance, the study had to eliminate factors such as service quality and others and for this reason the study was unable to conduct the analysis as initially planned. In addition, the data samples of the research were mainly gathered from students, which made it hard to say that the analysis was carried out on samples from every different group. Future research should analyze the characteristics of each group of customers for factors affecting customer satisfaction and customer loyalty in the family restaurant chain.

• Journal of Distribution Science
• /
• v.10 no.11
• /
• pp.61-70
• /
• 2012

Purpose - The entry of Super Supermarkets (SSM), operated by large corporations, into the retail distribution market weakens the competitive power of local supermarkets (SM) and their families leading to a rapid collapse of the local-SM market. Accordingly, this is a very sensitive and urgent issue in Korean society and politics. Therefore, the government is required to take steps to activate a local-SM market and regain the loyalty of their customers. However, in spite of the urgent need of a comparative study on the advantages and competitive power of local-SM in relation to the SSM, little research has been done on this issue to date. One purpose of this study is to analyze the differences in store characteristics of the local SM and SSM, and to explore the relationships amongst various store characteristics, such as consumption emotion, relation quality, and store loyalty between store types to provide an efficient solution to activate a local-SM market. Research design, data, methodology - To analyze the differences in the evaluation of store characteristics between store types, and to test the moderator and mediator effects, data were collected from 488 customers in Seoul, Incheon,and Gyeonggi Province. The theoretical model consists of four hypotheses, and data was processed to test these using factor analysis, t-test, regression analysis, hierarchical regression, and the three step regression proposed by Baron and Kenny. Results - First, the analysis shows that local-SM customers evaluate location characteristic factors significantly higher compared to SSM customers. However, SSM customers evaluate quality of merchandise, sales promotion activities, and store environment significantly higher. Yet, there are no significant differences in the evaluation of consumption emotion, relation quality,and store loyalty between the store types. Second, the analysis of moderator effects of the store type show that sales promotion and store atmosphere factors have a significantly negative moderating effecton positive consumption emotion, while positive consumption emotion has a significantly positive moderating effect on absorption of relation quality, while there are no significant moderating effects between consumption emotion and store loyalty. Third, while both trust and absorption of relation quality are partially mediated between consumption emotion and store loyalty, only absorption is partially mediated between positive consumption emotion and store loyalty. There are no significant mediating effects of absorption between negative consumption emotion and store loyalty. Conclusions - We strongly recommend that location restrictions on entrance of SSM into local-SM market should either be further strengthened or sustained for the activation of the local store market. Government should strengthen financial support to improve the quality of merchandise, sales promotion, employee education, store environment of the local-SM by providing appropriate business consulting. Moreover, the result implies that the owners of alocal-SM should take measures to increase customers' absorption and store loyalty, such as store cleanliness, maintaining a delightful atmosphere at the store, and training employees on attitudes towards customers. As this study is restricted within Seoul, Incheon, and Gyeonggi Province, care needs to be taken in generalizing the interpretation and application to the national level.

• Journal of the Korea Concrete Institute
• /
• v.21 no.5
• /
• pp.541-548
• /
• 2009

A theoretical prediction model of fiber bridging curve was established based on the assumption that fibers are uniformly distributed on the crack surface. However, the distance between fibers and their orientation with respect to crack surface can greatly affect the prediction of fiber bridging curve. Since, the shape of fiber bridging curve is a critical factor for predicting the tensile stress-strain relationship of ECC, it is expected that the assumption of uniform distribution of fiber may cause a significant error when predicting the tensile behavior of ECC. To overcome this shortcoming, a new prediction method of stress-strain relation of ECC is proposed based on the modified fiber bridging curve. Only effective fibers are taken into account considering the effects of their orientation and distance between them. Moreover, the approach for formulating the tensile stress-strain relation is discussed, where a procedure is presented for obtaining important parameters, such as the first crack strength, the peak stress, the displacement at peak stress, tensile strain capacity, and the crack spacing. Subsequent uniaxial tensile tests were performed to validate the proposed method. It was found that the predicted stress-strain relations obtained based on the proposed modified fiber bridging curve exhibited a good agreement with experimental results.

• Journal of Korean Society of Environmental Engineers
• /
• v.35 no.12
• /
• pp.867-876
• /
• 2013

To promote the carbon emission trading scheme and reduce greenhouse gas (GHG) emission as following 'Korean GHG & Energy Target Management System', GHG emissions should be accurately determined in each industrial sector. For the estimation method of GHG emission from waste landfill, there are several error parameters, therefore we reviewed the estimation method and proposed a revised method. Methane generation from landfill must be calculated by the selected method based on methane recovery rate, 0.75. However, this methodology is not considered about uncertainty factor. So it is desirable that $CH_4$ generation is estimated using first order decay model and methane recovery should use field monitoring data. If not, $CH_4$ recovery could be applied from other study results; 0.60 of operational landfill with gas vent and flaring system, 0.65 of operational site with landfill gas recovery system, 0.90 of closed landfill with final cover. Other parameters such as degradable organic carbon (DOC) and fraction of DOC decompose ($DOC_f$) need to derive the default value from studies to reflect a Korean waste status. Proper application of MCF that is selected by operation and management of landfill requires more precise criteria.

• Journal of Korean Society of Disaster and Security
• /
• v.12 no.4
• /
• pp.27-41
• /
• 2019

To analyze the effectiveness of the safety e-reporting system, the present study carried out analysis of safety e-reporting data reported between September 2014 and July 2019, and selected items for measuring the effectiveness of safety e-reporting. Using these items, the effects of adopting the reporting system for the four major parking violations was analyzed, alongside an analysis of effects in terms of traffic accidents using the unit model. When we count the securement of the tax revenue through measures such as charging fines as the beneficial factor per case, the estimation of the benefit is around 62,000 KRW per case. Summing the two factors up, the total value of citizen's reports pertaining to the big four parking violations is about 275,000 KRW per case. Most of the reports made through the Safety e-Report system are about traffic and facilities. When we calculate the total annual benefit with the representative reporting value defined with traffic and facilities, the system received a total of 1,164,439 cases from 2014 to 2019, while citizens reported 52,721 cases for the big four parking violations from April to July 2019. As the value of a safety report is around the net benefit for last five years is around 27,340,000,000 KRW.