• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2001년도 The Pacific Aisan Confrence On Intelligent Systems 2001
• /
• pp.224-229
• /
• 2001

When a semiconductor package is assembled, various materials such as die attach adhesive, lead frame, EMC (Epoxy Molding Compound), and gold wire are used. For better preconditioning performance, the combination between the packaging materials by studying the compatibility of their properties as well as superior packaging material selection is important. But it is not an easy task to find proper packaging material sets, since a variety of factors like package design, substrate design, substrate size, substrate treatment, die size, die thickness, die passivation, and customer requirements should be considered. This research applies case-based reasoning(CBR) technique to solve this problem, utilizing prior cases that have been experienced. Our particular interests lie in building decision support model to aid the selection of proper die attach adhesive. The preliminary results show that this approach is promising.

• 국제학술발표논문집
• /
• The 6th International Conference on Construction Engineering and Project Management
• /
• pp.213-216
• /
• 2015

Despite the significant increase of Korean contractors in the international construction market, many SMCCs (Small & Medium Construction Companies) have suffered in the global financial crisis, and some of them have been kicked out of the international market after experiencing huge losses on projects. SMCCs face obstacles in the international market, such as an insufficient ability to gather information and inappropriate management of associated risks, which lead to difficulties in establishing effective business strategies. In other words, making immature decisions without an effective business strategy may cause not only the failure of one project but also the bankruptcy of the SMCC. To overcome this, the research presented herein aims to propose a decision support system for SMCCs, which would screen projects and make a go/no-go decision at the early stages of international projects. The proposed system comprises a double axis: (1) a profit prediction model, which evaluates 10 project properties using an objective methodology based on a historical project performance database and roughly suggests expected profit rate, and (2) a feasibility assessment model, which evaluates 17 project environment factors in a subjective and quantitative methodology based on experience and supervision. Finally, a web-based system is established to enhance the practical usability, which is expected to be a good reference for inexperienced SMCCs to make proper decisions and establish effective business strategies.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.1032-1035
• /
• 2003

The field of handwriting recognition has been researched for many years. A hybrid classifier has been proven to be able to increase the recognition rate compared with a single classifier. In this paper, we combine support vector machine (SVM) and hidden Markov model (HMM) for offline handwritten numeral recognition. To improve the performance, we extract features adapted for each classifier and propose the modified SVM decision structure. The experimental results show that the proposed method can achieve improved recognition rate for handwritten numeral recognition.

• 한국건설관리학회논문집
• /
• 제20권4호
• /
• pp.56-68
• /
• 2019

거푸집 공사는 전체 건축공사의 공사기간 및 공사비 측면에서 상당 부분을 차지하고 있으며, 구조물공사의 품질, 마감 및 설비 등 후속공사에도 직접적인 영향을 미치는 매우 중요한 공정이다. 그러나 국내 건설현장에서의 거푸집 공법선정은 객관적인 데이터를 바탕으로 하지 않고, 현장 실무자의 경험과 직관에 의존하여 이루어지는 경우가 대부분이다. 그 결과, 다수의 현장에서 공기단축, 비용절감 등의 효과가 큰 신공법을 적용하고자 할 경우, 객관적이고 합리적인 의사결정 절차의 부재로 인해 최적 공법선정 시 어려움을 겪을 뿐만 아니라, 현장의 특성을 고려하지 못한 의사결정으로 인해 공기지연, 비용증대 등의 문제가 발생하고 있는 실정이다. 이러한 문제를 해결하기 위해 거푸집 공법선정과 관련한 다양한 선행연구가 이루어졌으나 현장조건이라는 중요한 요인을 고려하지 못해 실제 현장에서 적용하여 활용하기에 다소 한계성이 있었다. 따라서 본 연구는 현장 유형별 조건을 고려한 정량적 거푸집 선정 의사결정 지원 모델 개발을 그 최종 목적으로 한다. 개발된 의사결정 지원 모델은 공사현장에서 거푸집 선정 의사결정시 정량적인 평가 데이터를 제공함으로써, 기존의 경험에 의존하여 결정하던 의사결정 방법보다 신속하고 객관적으로 최적의 거푸집 공사를 선정 할 수 있도록 도울 것으로 사료된다. 또한, 경험이 풍부하지 않은 엔지니어도 본 연구의 의사결정 지원모델을 활용하여 현장의 조건을 고려한 보다 객관적이고 합리적인 의사결정을 내릴 수 있을 것으로 기대된다.

• 한국경영과학회지
• /
• 제19권3호
• /
• pp.15-40
• /
• 1994

We designed a knowledge-based decision support system for structuring semi-or unstructured problems. Problem structuring involves extraction of the relevant factors from the identified problem, and model construction that represents the relationships among those factors. In this research, we employed a directed graph called Influence Deiagram as a tool for problem structuring. In particular, our proposed system is designed as a shell. Therefore, a decision maker can change the content of the knowledge base to suit his/her own interested domain.

• 한국환경복원기술학회지
• /
• 제14권3호
• /
• pp.101-113
• /
• 2011

The purpose of this study are to develop a SDSS (Spatial Decision Support System) that can incorporate diverse opinions of stakeholders related the designation of protected areas (PA), and to employ the model for the readjustment of the boundary line of the Jirisan National Park of Korea. The SDSS would lead to more rational and less controversial decision-making during the expansion or removal of PA in Korea. Research methods are as follows. Firstly, to select evaluation criteria for SDSS for PA designation by using expert interview and literature survey. Secondly, to measure their preferences on the designation of additional PA or the removal of a part of PA based on the opinions of various stakeholders such as local residents, environmental groups, or public officials. Thirdly, to produce conservation priority maps based on a multi-criteria decision making technique. The SDSS would be used to rational decision making for the expansion of PA or the release of a certain part of PA by reflecting diverse preferences on biodiversity conservation and economic interest of residents. The visualization of conservation priority maps would also increase the efficiency of such decision making processes. The evaluation criteria for the expansion of PA for biodiversity conservation includes vegetation conservation value, wildlife conservation value, and the habitats of key species. The evaluation criteria for the removal of PA includes the proximity to roads and the boundary of PA, land use types, and conservation zoning of the PA. Preference weights are based on data collected from the Jirisan National Park. Both the conservation priority and removal priority maps are based on land parcels so that property rights of all parcels would be correctly represented.

• 품질경영학회지
• /
• 제28권2호
• /
• pp.89-102
• /
• 2000

Many organizations have developed their own traditional quality information systems. But, they think of it as one of the functional information systems not as a company-wide decision support information systems. A study on traditional quality information systems(QIS) has been conducted and a new conceptual framework of quality information system is proposed in this paper. In order to support enterprise wide total quality management aggressively, a new conceptual framework, named quality management support information system(QMSIS) is developed and proposed. This framework is based upon Malcolm Baldrige National Quality Award(MBNQA) model integrates management information system approach and traditional quality information system concept. In this model, organizational performance and process performance can be monitored to support managers , decision making about organizational quality management activities.

• 대한토목학회논문집
• /
• 제41권4호
• /
• pp.449-459
• /
• 2021

농업용 저수지의 80 % 이상이 50년 이상 된 노후 시설물로 안전성 및 기능 저하가 발생하고 있다. 이로 인해 저수지의 붕괴 등 안전사고가 발생하고 있는 실정이다. 이에 따라서 저수지의 붕괴 등의 안전사고를 미연에 방지하고자 정밀안전진단을 실시하고 우선순위에 따라 수리시설개보수사업을 시행하고 있다. 하지만 사업 우선순위 선정의 대부분은 시설물 관리자의 주관적인 판단을 통해 이루어지고 있다. 이에 본 연구에서는 정밀안전진단 결과 및 기존의 수리시설 개보수사업의 의사결정 사례를 D/B화하여 80개의 가설을 설정하고 상관분석 및 유의성검정을 통해 45개의 변수를 선정하였다. 선정된 변수들을 로지스틱회귀분석을 이용하여 의사결정지원모델을 제시하였다. 의사결정지원모델의 변수는 총 21개가 채택되었으며 모델의 분류 정확도는 86.8 %로 나타났다. 본 연구는 수리시설개보수사업 선정을 위한 의사결정의 정량적인 지표를 제시 한 부분에 중요한 의의를 가진다.

• 한국물환경학회지
• /
• 제20권2호
• /
• pp.151-156
• /
• 2004

A decision support system, Watershed Analysis Risk Management Framework(WARMF), was applied to the Kyungan Stream watershed, a tributary of Lake Paldang, for calculation of total maximum daily loads(TMDL). The WARMF system was developed by Systech Engineering, USA, and has been successfully used in several watersheds, for TMDL studies. The study area was divided into 14 sub-basins, based on digital elevation model(DEM). The integrated watershed and stream model of WARMF was validated by flow and BOD data measured during the year of 1999. There were reasonable agreements between model results and field data, both in water flow and BOD. The validated Kyungan WARMF was extensively utilized to study the quantitative relationship between waste loads and receiving water quality. Based on TMDL guideline at Paldang Lake and Kyungan Stream, the water quality criterion were set to be 3.0mg/L, 3.5mg/L, and 4.0mg/L at the watershed outlet. The allowable waste loads of BOD, both from point and non-point sources, were determined at each water quality criterion. From this study, it was concluded that the WARMF provided several advantages over the conventional application of watershed and stream models for TMDL study, such as time variable simulations, multiple possible soutions, and reduction loads for goal water quality, etc.

• 농촌계획
• /
• 제4권2호
• /
• pp.117-127
• /
• 1998

In this study, a model was developed to analyze the capacity and the total price of the agricultural products marketing between nations through the estimation of the production and consumption amount of the agricultural products in each nation and the analysis of the price and transport cost to each nation. The method which can contribute to the agricultural policy decision support was devised. The main concept of the method is to compute the potential import-export amount and total cost among the nations. In the application, wheat was selected to evaluate the model. The application results showed that the model could analyzed the unit consumption and storage amount per capital of each nation and the price and transport cost per unit weight from each export nation, provided the policy decision maker with the basic data analyzed by GIS.