• Asia pacific journal of information systems
• /
• v.8 no.3
• /
• pp.51-73
• /
• 1998

This paper is to evaluate the application potentials of data mining in the areas of Supply Chain Management (SCM) and to suggest the architectures of Decision Support Systems (DSS) that support data mining activities. We first briefly introduce data mining and review the recent literatures on SCM and then evaluate data mining applications to SCM in three aspects: marketing, operations management and information systems. By analyzing the cases about pricing models in distribution channels, demand forecasting and quality control, it is shown that artificial intelligence techniques such as artificial neural networks, case-based reasoning and expert systems, combined with traditional analysis models, effectively mine the useful knowledge from the large volume of SCM data. Agent-based information system is addressed as an important architecture that enables the pursuit of global optimization of SCM through communication and information sharing among supply chain constituents without loss of their characteristics and independence. We expect that the suggested architectures of intelligent DSS provide the basis in developing information systems for SCM to improve the quality of organizational decisions.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.05a
• /
• pp.303-306
• /
• 2000

본 논문에서는 선박에서 화재탐지를 위해서 규칙 기반 추론과 사례 기반 추론을 통합하는 방법에 대해서 논의하였다. 규칙은 어떤 영역에서 광범위한 경향을 표현하는데 적합하며 사례는 규칙에서 예외적인 상황을 다루는데 적합하다는 점에서 규칙과 사례는 상호 보완적이라 할 수 있다. 즉 어떤 행동이 충분히 반복되면 자연스럽게 규칙이 되며, 잘 확립된 규칙이 있다면 사례를 먼저 추론할 필요가 없다. 그러나 규칙이 실패하게 되면 실패를 만회하기 위해서 사례를 생성하는 것이 하나의 대안이 될 수 있다. 본 논문에서는 일반적인 화재탐지 지식은 규칙으로 표현하고, 예외적인 화재탐지 지식은 사례로 표현함으로써 규칙과 사례가 서로 보완적인 역할을 할 수 있는 통합 방법을 제안하였다. 또한 기존의 규칙 기반 FFES(Fire Fighting Expert System)와 사례기반 추론에 의해 확장된 C-FFES(Combined-Fire Fighting Expert System)를 비교를 통해, 제안한 접근 방법이 화재 탐지율을 향상시킴을 보였다.

• Journal of Korean Institute of Industrial Engineers
• /
• v.26 no.4
• /
• pp.306-314
• /
• 2000

In this paper we present a new approach to estimate link travel speed based on the hybrid neuro-fuzzy network. It combines the fuzzy ART algorithm for structure learning and the backpropagation algorithm for parameter adaptation. At first, the fuzzy ART algorithm partitions the input/output space using the training data set in order to construct initial neuro-fuzzy inference network. After the initial network topology is completed, a backpropagation learning scheme is applied to optimize parameters of fuzzy membership functions. An initial neuro-fuzzy network can be applicable to any other link where the probe car data are available. This can be realized by the network adaptation and add/modify module. In the network adaptation module, a CBR(Case-Based Reasoning) approach is used. Various experiments show that proposed methodology has better performance for estimating link travel speed comparing to the existing method.

• International conference on construction engineering and project management
• /
• 2007.03a
• /
• pp.239-248
• /
• 2007

In the highly competitive construction industry, a slight inaccuracy of estimation can easily cause the loss of a project. Erroneous experience-based cost estimates or allocations of on-site supervisory manpower often offset the profit gained from the project and may jeopardize the management processes. To counter these types of problems, we develop a model using mathematical analysis and case-based reasoning to automate the allocation of on-site supervisory manpower and estimate construction site costs. The method is founded upon laborious data collection processes and analysis by matching statistical assumptions, and is applicable to construction projects. In the modeling the costs and allocation of on-site supervisory manpower are quantified for both owners and contractors before initiating or bidding on the projects. The findings confirm that the degree of variation of the model predictions has an accuracy rate at 88.47%. Single-site construction projects can be accurately predicted and the assignment of supervisory manpower feasibly automated.

• International conference on construction engineering and project management
• /
• 2005.10a
• /
• pp.830-835
• /
• 2005

4D models help construction planners to develop and evaluate construction plans. However, current analyses using 4D models are mainly visual and limit the quantitative comparison of construction alternatives. This paper explores the usefulness of extracting quantitative information from 4D models to support time-space analyses. We use two 4D models of an industry test case to illustrate how to analyze 4D content quantitatively (i.e., work space areas and distances between concurrent activities). This paper shows how these two types of 4D content can be extracted from 4D models to support 4D-based-analysis and novel presentation of construction planning information. We suggest further research to formalize the content of 4D models to enable comparative quantitative analyses of construction planning alternatives. Formalized 4D content will enable the development of reasoning mechanisms that automate 4D-model-based analyses and provide the information content for informative presentations of construction planning information.

• Journal of Information Technology Services
• /
• v.17 no.1
• /
• pp.91-103
• /
• 2018

This study developed a prediction model using machine learning technology and predicted the success of health consulting by using life log data generated through u-Health service. The model index of the Random Forest model was the highest using. As a result of analyzing the Random Forest model, blood pressure was the most influential factor in the success or failure of metabolic syndrome in the subjects of u-Health service, followed by triglycerides, body weight, blood sugar, high cholesterol, and medication appear. muscular, basal metabolic rate and high-density lipoprotein cholesterol were increased; waist circumference, Blood sugar and triglyceride were decreased. Further, biometrics and health behavior improved. After nine months of u-health services, the number of subjects with four or more factors for metabolic syndrome decreased by 28.6%; 3.7% of regular drinkers stopped drinking; 23.2% of subjects who rarely exercised began to exercise twice a week or more; and 20.0% of smokers stopped smoking. If the predictive model developed in this study is linked with CBR, it can be used as case study data of CBR with high probability of success in the prediction model to improve the compliance of the subject and to improve the qualitative effect of counseling for the improvement of the metabolic syndrome.

• Journal of the Korea Society of Computer and Information
• /
• v.8 no.1
• /
• pp.20-29
• /
• 2003

According as quantity of information is augmented rapidly in World Wide Web, users are investing more times finding correct information to on. Search function that a search agent is personalized according to user's preference degree or search objective to solve these problem should be offered. Therefore, a search agent accumulates experienced knowledge connected with user's past search in this research. When new query was given, search agent offered learning function of intelligence that decides category group through estimation method of similarity using this knowledge. So this paper showed that case based search can bring superior result in the correctness rate than other search method.

• International Journal of Quality Innovation
• /
• v.3 no.2
• /
• pp.138-158
• /
• 2002

Nowadays, new product development (NPD) is one of the most crucial factors for business success. The manufacturing firms cannot afford the resources in the long development cycle and the costly redesigns. Good product planning is crucial to ensure the success of NPD, while the Quality Function deployment (QFD) is an effective tool to help the decision makers to determine appropriate product specifications in the product planning stage. Traditionally, in the QFD, the product specifications are determined by a rather subjective evaluation, which is based on the knowledge and experience of the decision makers. In this paper, the traditional QFD methodology is firstly reviewed. An improved Hybrid Quality Function Deployment (HQFD) [MSOfficel] then presented to tackle the shortcomings of traditional QFD methodologies in determining the engineering characteristics. A structured questionnaire to collect and analyze the customer requirements, a methodology to establish a QFD record base and effective case retrieval, and a model to more objectively determine the target values of engineering characteristics are also described.

• International conference on construction engineering and project management
• /
• 2015.10a
• /
• pp.597-598
• /
• 2015

In order to improve the reliability of cost estimation results using CBR, there has been a continuous issue on similarity measurement to accurately compute the distance among attributes and cases to retrieve the most similar singular or plural cases. However, these existing similarity measures have limitations in taking the covariance among attributes into consideration and reflecting the effects of covariance in computation of distances among attributes. To deal with this challenging issue, this research examines the weighted Mahalanobis distance based similarity measure applied to CBR cost estimation and carries out the comparative study on the existing distance measurement methods of CBR. To validate the suggest CBR cost model, leave-one-out cross validation (LOOCV) using two different sets of simulation data are carried out. Consequently, this research is expected to provide an analysis of covariance effects in similarity measurement and a basis for further research on the fundamentals of case retrieval.

• Proceedings of the Korean Society for the Gifted Conference
• /
• 2003.05a
• /
• pp.137-138
• /
• 2003

Understanding how and why people differ is a fundamental, if distant, goal of research efforts to bridge psychological and biological levels of analysis. General fluid intelligence (gF) is a major dimension of individual differences and refers to reasoning and novel problemsolving ability. A conceptual integration of evidence from cognitive (behavioral) and anatomical studies suggeststhat gF should covary with both task performance and neural activity in specific brain systems when specific cognitive demands are present, with the neural activity mediating the relation between gF and performance. Direct investigation of this possibility will be a critical step toward a mechanistic model of human intelligence. In turn, a mechanistic model might suggest ways to enhance gF through targeted behavioral or neurobiological intervent ions, We formed two different groups as subjects based on their scholarly attainments. Each group consists of 20 volunteers(aged 16-17 years, right-handed males) from the National Gifted School and a local high school respectively. To test whether individual differences in general intelligence are mediated at a neural level, we first assessed intellectual characteristics in 40 subjects using standard intelligence tests (Raven's Advanced Progressive Matrices, Wechsler Adult Intelligence Scale, Torrance Tests of Creative Thinking) administered outside of the MR scanner. We then used functional magnetic resonance imaging (fMRl) to measure task-related brain activity as participants performed three different kinds of computerized reasoning tasks that were intended to activate the relevant neural systems. To examine the difference of neural activity according to discrepancy in general intelligence, we compared the brain activity of both extreme groups (each, n=10) of the participants based on the standard intelligence test scores. In contrast to the common expectation, there was no significant difference of brain region involved in high-g tasks between both groups. Random effect analysis exhibited that lateral prefrontal, anterior cingulate and parietal cortex are associated with gF. Despite very different task contents in the three high-g-low-g contrasts, recruitment of multiple regions is markedly similar in each case, However, on the task with high 9F correlations, the Prodigy group, (intelligence rank: >99%) showed higher task-related neural activity in several brain regions. These results suggest that the relationship between gF and brain activity should be stronger under high-g conditions than low-g conditions.