• 산업경영시스템학회지
• /
• 제28권4호
• /
• pp.48-54
• /
• 2005

Recently, a couple of process capability indices are used to evaluate that the outputs of the process satisfy the specifications. An assumption of those indices is that the specifications of the characteristics are given single constant value. The display panel is a highly precise product and all the specifications of measure points are designed by their locations in the panel. So it is very difficult to locate the measurement facility to the exact position. In this paper, we propose a new process capability index in case of multi specifications and unfixable objects.

• Asia pacific journal of information systems
• /
• 제32권2호
• /
• pp.226-248
• /
• 2022

The existing model for sentiment analysis of product reviews learned from past data and new data was labeled based on training. But new data was never used by the existing system for making a decision. The proposed Aspect-based multi-agent Deep Reinforcement learning Sentiment Analysis (ADRSA) model learned from its very first data without the help of any training dataset and labeled a sentence with aspect category and sentiment polarity. It keeps on learning from the new data and updates its knowledge for improving its intelligence. The decision of the proposed system changed over time based on the new data. So, the accuracy of the sentiment analysis using deep reinforcement learning was improved over supervised learning and unsupervised learning methods. Hence, the sentiments of premium customers on a particular site can be explored to other customers effectively. A dynamic environment with a strong knowledge base can help the system to remember the sentences and usage State Action Reward State Action (SARSA) algorithm with Bidirectional Encoder Representations from Transformers (BERT) model improved the performance of the proposed system in terms of accuracy when compared to the state of art methods.

• 한국전자거래학회지
• /
• 제27권1호
• /
• pp.15-28
• /
• 2022

최근 의류 등의 특정 쇼핑몰의 온라인 시장이 크게 확대되면서, 사용자의 리뷰를 활용하는 것이 주요한 마케팅 방안이 되었다. 이를 이용한 감성분석에 대한 연구들도 많이 진행되고 있다. 감성분석은 사용자의 리뷰를 긍정과 부정 그리고 필요에 따라서 중립으로 분류하는 방법이다. 이 방법은 크게 머신러닝 기반의 감성분석과 사전기반의 감성분석으로 나눌 수 있다. 머신러닝 기반의 감성분석은 사용자의 리뷰 데이터와 그에 대응하는 감성 라벨을 이용해서 분류 모델을 학습하는 방법이다. 감성분석 분야의 연구가 발전하면서 리뷰와 함께 제공되는 이미지나 영상 데이터 등을 함께 고려하여 학습하는 멀티모달 방식의 모델들이 연구되고 있다. 리뷰 데이터에서 제품의 카테고리와 사용자별로 사용되는 단어 등의 특징이 다르다. 따라서 본 논문에서는 리뷰데이터와 제품 정보를 동시에 고려하여 감성분석을 진행한다. 리뷰를 분류하는 모델로는 기본 순환신경망 구조에서 Gate 방식을 도입한 Gated Recurrent Unit(GRU), Long Short-Term Memory(LSTM) 그리고 Self Attention 기반의 Multi-head Attention 모델, Bidirectional Encoder Representation from Transformer(BERT)를 사용해서 각각 성능을 비교하였다. 제품 정보는 모두 동일한 Multi-Layer Perceptron(MLP) 모델을 이용하였다. 본 논문에서는 사용자 리뷰를 활용한 Baseline Classifier의 정보와 제품 정보를 활용한 MLP모델의 결과를 결합하는 방법을 제안하며 실제 데이터를 통해 성능의 우수함을 보인다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.1224-1226
• /
• 2003

Dust-storm is one of the heaviest hazardous weather which frequently affects most part of northern China in spring. Satellite multi-spectral observations can provide significant information for detecting and quantitative determining the property of dust-storm . An algorithm to monitor dust-storm automatically was developed based on satellite observation. The algorithm utilizes split widows technique and spectral classification technique and also developed a new dust remote sensing product Infra -red Difference Dust Index (IDDI) proxy dust-loading dataset using GMS-5.

• Design & Manufacturing
• /
• 제17권4호
• /
• pp.57-62
• /
• 2023

As a representative method for mass production, a multi-cavity type mold capable of simultaneously molding products of the same shape can be applied. It has the advantage of improving the productivity from several times to several tens of times, but it may cause disadvantages which is the quality deviation with each cavity. This study, therefore, has tried to increase the cavity filling balance by using a melt flipper and a flow distance control part in the runner part of the mold. Along with this, the design and manufacturing of air vents during injection molding have been verified through experimental methods to achieve a higher level of multi-cavity filling balance and dimensional accuracy.

• Industrial Engineering and Management Systems
• /
• 제12권3호
• /
• pp.254-263
• /
• 2013

Disassembly of discarded products takes place in the process of remanufacturing, recycling, and disposal. The disassembly lines have been taken as available choice for automated disassembly; therefore, it has become essential that it be designed and balanced to work efficiently. The multi-objective disassembly line balancing problem seeks to find a disassembly sequence which provides a feasible disassembly sequence, minimizes the number of workstations and idle time, and balances the line for the disassembly of post consumed product by considering the environment effects. This paper proposes a multi-criteria decision making technique based heuristic for assigning the disassembly tasks to the workstations. In the proposed heuristic, the PROMETHEE method is used for prioritizing the tasks to be assigned. The tasks are assigned to the disassembly workstations according to their priority rank and precedence relations. The proposed heuristic is illustrated with an example, and the results show that substantial improvement in the performance is achieved compared with other heuristics.

• Tribology and Lubricants
• /
• 제26권1호
• /
• pp.52-55
• /
• 2010

This study investigates the friction of the roll stands in the high performance multi- holes slate fin machine which pin is using in heat transfer of radiator, oil cooler, inter cooler, condenser and evaporator. The roll stand part is very important to maintain the high performance fin machine. This multi-holes form roll type is the first time in our country so it will be helpful to increases the export and product of heat transfer. It was include the technical of form rolling system which was self development. Then it will be improve the electric home appliances, future motor vehicle device and industrial machine.

• 한국생활과학회지
• /
• 제17권4호
• /
• pp.745-758
• /
• 2008

In this paper, we examined the trend information collected by apparel industries to develop fashionable products. We also investigated the product development process of a swimwear industry, ELLE Swimwear, through in-depth interviews. The results are as follows; 1. The apparel industries were collecting information about a fashion trend, international exhibitions and fairs, world collections, and a market trend to develop fashionable products. 2. As the swimwear consumers are spreaded in all ages, the launching of more specialized swimwear brand with ages is required. 3. These days, swimwear is not only an item of sportswear but also being a fashionable product on street. Therefore, the industries should try to develop a various assortment of swimwear capable of coordinating with multi-mix instead of developing a separate item. 4. To design and produce swimwear satisfying functional, sensual and cultural performance, the development of new materials and patterns are required. 5. Nurturing of swimwear pattern specialists dealing with CAD is urgently required. 6. Turning of understanding about the importance of VMD and more attractive marketing strategies by a specialized team are also required.

• 대한용접접합학회:학술대회논문집
• /
• 대한용접접합학회 2008년도 춘계학술발표대회 개요집
• /
• pp.36-36
• /
• 2008

• 한국전자거래학회지
• /
• 제24권3호
• /
• pp.53-71
• /
• 2019

소프트웨어 시스템이 복잡하고 대규모화 되어감에 따라 새로운 기능을 제공하기 위해 소프트웨어 컴포넌트 또는 모듈을 재사용하는 방법을 요구하고 있다. 본 논문에서는 서비스 제공자와 서비스 사용자에게 SOA와 MSPL을 결합하여 가변성 서비스를 제공하고 재사용하여 웹 시스템 개발 방법을 제시하였다. 제시한 방법은 서비스 제공자 관점에서, 재사용 가능한 가변성 서비스를 휘처 식별 가이드라인을 적용하여 신택스 기반, 기능기반과 행위기반 방법으로 휘처로 식별하고 구현하여 재사용 자산으로 관리하였다. 그리고 사용자 관점에서 서비스를 구조적으로 조합하고 재구성하는 방법으로서 워크플로우를 모델링하여 서비스를 구성하여 웹 시스템을 구현하였다. 본 논문에서 구축한 웹 시스템의 재사용을 기능점수로 측정한 결과 재사용의 증가와 유사프로젝트에 적용해 비용 절감 효과를 검증하였다.