• 한국항해항만학회지
• /
• 제42권3호
• /
• pp.159-166
• /
• 2018

해양사고를 야기한 선원들의 행동오류 식별은 해양사고 예방, 저감 또는 억제에 중요한 단서가 된다. 본 연구의 목적은 SRKBB(Skill-, Rule-, and Knowledge Based Behavior) 이론을 이용하여 해양사고 종류별로 선원들의 행동오류를 식별하는데 있다. 행동오류 식별을 위하여 9년간(2008~2016)의 해양사고 재결서 1,744건에 기록된 사고내용을 수집한 후, 사고를 야기한 선원들의 행동오류를 SBBE(Skill-Based Behavioral Error), RBBE(Rule-Based Behavioral Error), KBBE(Knowledge-Based Behavioral Error) 세 가지 종류로 분류하였다. 행동오류 분류를 위하여 SRKBB 이론을 적용한 행동오류 분류용 프레임워크를 제안하고, 이 프레임워크를 이용하여 행동오류 데이터를 구축하였다. 사고종류별 행동오류의 빈도를 분석한 결과, 충돌사고는 SBBE가 가장 높은 빈도로 관측되었고, 이어서 RBBE가 두 번째로 높은 빈도로 관측되었다. 이에 반하여 좌초, 전복 및 침몰 등의 사고는 KBBE 중에서 높은 빈도로 관측되었다. 연구결과, 해양사고 종류별로 선원들의 행동오류 식별이 가능하였고, 해양사고 종류별 사고 예방에 필요한 선원들의 행동오류 보정에 관한 단서를 확보할 수 있었다.

• 한국농업기계학회:학술대회논문집
• /
• 한국농업기계학회 1993년도 Proceedings of International Conference for Agricultural Machinery and Process Engineering
• /
• pp.1222-1229
• /
• 1993

An advanced design environment , which is based on adaptive and knowledge -based finite elements (INTELMESH), has been developed. Unlike other approaches, INTEMMESH incorporates the information about the object geometry as well as the boundary and loading conditions to generate an ${\alpha}$-priori finite element mesh which is more refined around the critical regions of the problem domain. INTEMMESH is designed for planar domains and axisymmetric 3-D structures of elasticity and heat transfer subjected to mechanical and thermal loading . It intelligently identifies the critical regions/points in the problem domain and utilize the new concepts of substructuring and wave propagation to choose the proper mesh size for them. INTEMMESH generates well-shaped triangular elements by applying trangulartion and Laplacian smoothing procedures. The adaptive analysis involves the intial finite elements analyze and an efficient ${\alpha}$-posteriori error analysis involves the initial finite element anal sis and an efficient ${\alpha}$-posteriori error analysis and estimation . Once a problem is defined , the system automatically builds a finite element model and analyzes the problem though automatic iterative process until the error reaches a desired level. It has been shown that the proposed approach which initiates the process with an ${\alpha}$-priori, and near optimum mesh of the object , converges to the desired accuracy in less time and at less cost. Such an advanced design/analysis environment will provide the capability for rapid product development and reducing the design cycle time and cost.

• 지능정보연구
• /
• 제9권3호
• /
• pp.49-63
• /
• 2003

본 논문은 효율적이고 신뢰성 있는 전문가시스템을 구축하기 위한 도구를 제안하고 구현한다. 전문가 시스템에서 추론은 특정 전문분야의 지식베이스에 저장된 지식을 기반으로 행해진다. 이 때 전문가 시스템이 신뢰할 수 있는 추론의 결과를 얻기 위해서는 다양한 형태의 지식들이 이용되고, 지식의 정확성 및 일관성이 유지되어야 한다. 이러한 관점에서 본 논문은 지식이 지식베이스에 첨가되기 전에 지식의 오류를 점검함으로써 오류가 없는 지식들을 선택적으로 지식베이스에 첨가하여, 지식의 정확성 및 일관성을 유지하는 지식관리 시스템을 구축했다. 아울러 본 논문은 전문가 시스템이 추론과정에서 다양한 지식을 이용하도록 절차적 지식과 데이터베이스에 저장된 선언적 지식을 호출하여 사용할 수 있도록 했다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 19-21 Oct. 1992
• /
• pp.797-800
• /
• 1992

Much experience and knowledge is needed in welding because there are many working parameters and quantitative description is difficult. Therefore, introduction of expert system based on such data base has been required. In this study, welding heat and shape of bead was controlled by fuzzy inference with the welding condition, position error and voltage and current error of robot. For this, torch trajectory of robot was generated by modeling the working data with CAD and then welding was carried out through down loading to robot. And working error was controlled by on-line communication.

• 대한인간공학회지
• /
• 제32권1호
• /
• pp.9-16
• /
• 2013

Objective: The aim of this study is to investigate a trend of human error types observed in a series of verification and validation experiments for an Advanced Control Room(ACR) equipped with Lager Display Panel(LDP), Work Station Flat Panel Display(WS FPD), list type Alarm System(AS), Soft Control(SC) and Computerized Procedure System(CPS). Background: Operator behaviors in a fully computerized control room are quite different from those in a traditional hard-wired control room. Operators in an ACR all together monitor plant status and variables through their own interface system such as LDP and WS FPD, are notified of abnormal plant status through their own list type AS, control the plant through their own SC, and follow the structured procedure through their own CPS whereas operators in a traditional control room only separately do their duty directed by their supervisor. Especially the secondary task such as manipulating the user interface of ACR can be an extra burden to all the operators including the supervisor. Method: The Reason's human error classification method was applied to operators' behavioral data collected from a series of verification and validation experiments where operators showed their plant operational behaviors under a couple of harsh scenarios using the ACR simulator. Results: As operators accustomed to the new ACR system, knowledge or rule based mistakes appearing frequently in the early series of experiments decreased drastically in the latest stage of the series. Slip and lapse types of errors were observed throughout the series of experiments. Conclusion: Education and training can be one of the most important factors for the operators accustomed to the traditional control room to be adapted to the new system and to run the ACR successfully. Application: The results of this study implied that knowledge or rule based mistakes can be reduced by training and education but that lapse type errors might be reduced only through innovative improvement in human-system interface design or teamwork culture design including a new leadership style suitable for ACR.

• 말소리와 음성과학
• /
• 제10권1호
• /
• pp.49-55
• /
• 2018

This paper aims to describe a knowledge-based pronunciation generation system for French. It has been reported that a rule-based pronunciation generation system outperforms most of the data-driven ones for French; however, only a few related studies are available due to existing language barriers. We provide basic information about the French language from the point of view of the relationship between orthography and pronunciation, and then describe our knowledge-based pronunciation generation system, which consists of morphological analysis, Part-of-Speech (POS) tagging, grapheme-to-phoneme generation, and phone-to-phone generation. The evaluation results show that the word error rate of POS tagging, based on a sample of 1,000 sentences, is 10.70% and that of phoneme generation, using 130,883 entries, is 2.70%. This study is expected to contribute to the development and evaluation of speech synthesis or speech recognition systems for French.

• 대한인간공학회지
• /
• 제30권1호
• /
• pp.169-177
• /
• 2011

The aim of this study is to review previous studies on human errors in the service delivery processes. Service industry is sharply growing in the advanced countries. Many people are looking for something to contribute to the service industry. Although there are many research topics related to service domain that human factors and ergonomics specialists can do contribute, a few researchers are studying such topics. This paper indicated how previous researches on human factors and human errors have addressed the service domain, in order to prompt human factor study on the service domain. A variety of sources were inspected for literature reviews, including books and journals of managements, medicine, psychology, consumer behavior as well as human factor and ergonomics. The characteristics of human errors in the service domain were investigated. Human error studies in several service sectors were summarized such as medical service, automotive service operation, travel agent service and call center service. Until now, human factors community was not much interested in human errors in service domain. However, there is much space to contribute to service domain; human error identification, human error analysis and control of human error. The research of human error in service domain can provide clues to improve service quality. This paper helps to guide to identify human error of service domain and to design service systems.

• 응용통계연구
• /
• 제6권1호
• /
• pp.29-40
• /
• 1993

모델을 이용한 표본론에서는 오차에 대한 함수식이 불확실한 경우가 종종 발생되는데 이러 한 오차에 대한 지식이 결여 되었을 때 발생되는 잘못된 효과를 줄일 수 있는 방법이 연구 되었다. 제시된 표본방법론은 모든 가능한 오차함수식에 대한 비효율성에 대한 평균을 최소 화하는데 그 목적이 있다. 컴퓨터를 이용한 알고리즘이 제시되었고 회계감사에 관련된 특수 한 경우의 예를 들어 이러한 방법의 효율성을 알아 보았다.

• Journal of Mechanical Science and Technology
• /
• 제20권11호
• /
• pp.1863-1872
• /
• 2006

In FMS (Flexible Manufacturing System) and CIM (Computer Integrated Manufacturing), machine-tools have been the target of integration in the last three decades. The conventional concept of integration is being changed into the autonomous manufacturing device based on the knowledge evolution by applying advanced information technology in which an open architecture controller, high-speed network and internet technology are included. In the advanced environment, the machine-tools is not the target of integration anymore, but has been the key subject of cooperation. In the near future, machine-tools will be more improved in the form of a knowledge-evolutionary intelligent device. The final goal of this study is to develop an intelligent machine having knowledge-evolution capability and a management system based on internet operability. The knowledge-evolutionary intelligent machine-tools is expected to gather knowledge autonomically, by producing knowledge, understanding knowledge, reasoning knowledge, making a new decision, dialoguing with other machines, etc. The concept of the knowledge-evolutionary intelligent machine is originated from the machine control being operated by human experts' sense, dialogue and decision. The structure of knowledge evolution in M2M (Machine to Machine) and the scheme for a dialogue agent among agent-based modules such as a sensory agent, a dialogue agent and an expert system (decision support agent) are presented in this paper, with intent to develop the knowledge-evolutionary machine-tools. The dialogue agent functions as an interface for inter-machine cooperation. To design the dialogue agent module in an M2M environment, FIPA (Foundation of Intelligent Physical Agent) standard platform and the ping agent based on FIPA are analyzed in this study. In addition, the dialogue agent is designed and applied to recommend cutting conditions and thermal error compensation in a tapping machine. The knowledge-evolutionary machine-tools are expected easily implemented on the basis of this study and shows a good assistance to sensory and decision support agents.

• 대한임베디드공학회논문지
• /
• 제17권4호
• /
• pp.209-215
• /
• 2022

This paper proposes a novel approach for training an unsupervised depth estimation algorithm. The objective of unsupervised depth estimation is to estimate pixel-wise distances from camera without external supervision. While most previous works focus on model architectures, loss functions, and masking methods for considering dynamic objects, this paper focuses on the training framework to effectively use depth cue. The main loss function of unsupervised depth estimation algorithms is known as the photometric error. In this paper, we claim that direct depth cue is more effective than the photometric error. To obtain the direct depth cue, we adopt the technique of knowledge distillation which is a teacher-student learning framework. We train a teacher network based on a previous unsupervised method, and its depth predictions are utilized as pseudo labels. The pseudo labels are employed to train a student network. In experiments, our proposed algorithm shows a comparable performance with the state-of-the-art algorithm, and we demonstrate that our teacher-student framework is effective in the problem of unsupervised depth estimation.