• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10b
• /
• pp.635-637
• /
• 1998

시험 단계나 운용 단계에서 발견된 소프트웨어의 오류를 수정하기 위해서는 많은 비용을 투자해야 한다. 시스템 개발 초기 단계인 설계 단계에서 소프트웨어 시스템의 신뢰도에 영향을 많이 미치는 부분을 찾아 오류를 사전에 방지하는 연구가 많이 진행되고 있다. 모듈의 신뢰도를 설계 단계에서 예측할 수 있다면 프로젝트 관리자는 결함 경향이 강한 모듈 개발에 더 많은 자원을 할당함으로써 보다 신뢰성 있는 소프트웨어를 생산 할 수 있다. 본 논문에서는 실시간 소프트웨어의 설계 결과에 대한 복잡도 측정치를 토대로 신뢰도를 예측하는 모델을 제안하다. 유전자 알고리즘으로 찾아낸 이 모델을 사용하여 결함 경향이 강한(fault prone) 모듈과 그렇지 않은 모듈은 96%의 정확도로 선별해 낼 수 있다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.1
• /
• pp.566-576
• /
• 2018

As the weapon systems of the Korean Army possess massive firepower and multiple functions, the improvement of their quality through reliability prediction is becoming increasingly important. Currently, the reliability prediction of the weapon systems of the Korean Army is a difficult process, because it is conducted by naively calculating the reliability of their constituent parts. Recently, as various studies using accumulated data are undertaken across various industries, the defense industry is also attempting to analyze the Dark Data which have been accumulated but not yet used. Therefore, it is necessary to apply Post-Logistics Support (PLS) data in order to improve the reliability of the weapon systems and, for this purpose, the Korean Army needs to conduct diverse studies. Especially, the PLS data in the defense industry is very useful for reliability prediction, because the data on the defects reported after the development of the weapon systems are accumulated in this phase. This study examines the existing reliability prediction method conducted using the component parts and proposes a new reliability prediction method using PLS data. This framework can ultimately contribute to improve the prediction accuracy and quality of the weapon systems.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.5
• /
• pp.533-538
• /
• 2004

This paper proposes a systematic method to develop short-term electrical load forecasting systems using neuro-fuzzy models. The proposed system predicts the electrical loads with the lead times of 1 hour, 24 hour, and 168 hour. To do so, the load forecasting system first builds an initial structure off-line for each hour of four day types and then stores the resultant initial structures in the initial structure bank. 96 initial structures are constructed for each prediction lead time. Whenever a prediction needs to be made, the proposed system initializes the neuro-fuzzy model with the appropriate initial structure stored and trains the initialized prediction modell. To improve the performance of the prediction system in terms of accuracy and reliability at the same time, the prediction model employs only two inputs. It makes possible to interpret the fuzzy rules to be learned. In order to demonstrate the viability of the proposed method, we develop a load forecasting system by using the real load data collected during 1996 and 1997 at KEPCO. Simulation results reveal that the prediction system developed in this paper can achieve a remarkable improvement on both accuracy and reliability

• The Journal of the Acoustical Society of Korea
• /
• v.19 no.4
• /
• pp.16-22
• /
• 2000

In order to predict the performance and the usefulness of the narrow-band underwater acoustic measurement system at design stage, whose error variance is not clearly described, in this study a boundary equation to estimate the measurement accuracy is proposed based on the confidency as SNR variation. The boundary is presented as a function of SNR and the number of samples. In this paper, the measurement performance for narrow-band signal is simulated by the proposed boundary equation and the results are reviewed in the biased noise condition and separately in the background noise rejected condition.

• Journal of KIISE:Software and Applications
• /
• v.27 no.4
• /
• pp.393-401
• /
• 2000

This paper explores neuro-fuzzy system in order to improve the software reliability predictability from failure data. We perform numerical simulations for actual 10 failure count and 4 failure time data sets from different software projects with the various number of rules. Comparative results for next-step prediction problem is presented to show the prediction ability of the neuro-fuzzy system. Experimental results show that neuro-fuzzy system is adapt well across different software projects. Also, performance of neuro-fuzzy system is favorably with the other well-known neural networks and statistical SRGMs.

• Proceedings of the Korean Association for Survey Research Conference
• /
• 2000.06a
• /
• pp.15-34
• /
• 2000

Since the results of the election forecasting survey that was executed jointly by T.V. stations and survey research companies in the 16th Korea General Election(April 13, 2000) had many errors, the reliability of the election forecasting survey was greatly damaged. Therefore, in order to recover the reliability and to increase the accuracy of the election forecasting survey I the future, we figure out the sources of the survey\\`s errors and suggest methods of reducing them through deeply analyzing the forecasting data from many angles. In addition, we discuss some problems and an improvable direction on exit poll executed for the first time.

• Survey Research
• /
• v.1 no.2
• /
• pp.15-34
• /
• 2000

Since the results of the election forcasting survey that was executed jointly by T.V. stations and survey research companies in the 16th Korea General Election(April 13, 2000) had many errors, the reliability of the election forecating survey was greatly damaged. Therefore, in order to recover the reliability and to increase the accuracy of the election forecasting survey in the future, we figure out the sources of the survey's errors and suggest methods of reducing them through deeply analyzing the forecasting data from many angles. In addition, we discuss some problems and an improvable direction on exit poll executed for the first time.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.01a
• /
• pp.347-348
• /
• 2014

최근 급증하는 교통 혼잡으로 인해 시간적/물질적 손실이 크게 발생하고 있다. 이러한 교통난 해소는 시설투자만으로는 근본적인 해결책이 될 수 없다는 판단 하에 지난 수년간 보다 정확한 교통량을 예측하기 위해 시계열 기반의 다양한 교통량 예측 모델들이 개발 되어 왔다. 그러나 시계열 기반의 모델들은 회귀분석을 통해 과거 교통량을 분석하고 과거의 교통패턴이 미래에도 지속적으로 연장된다는 가정 하에 연구되었기 때문에 실시간으로 급변하는 불규칙한 교통 패턴에 대한 예측의 신뢰성을 떨어트린다. 또한 시계열 기반의 예측 기법은 어떠한 회귀분석 모델을 사용하는지에 따라 성능의 차이가 많이 나타나기 때문에 회귀분석 모델 선택이 중요하다. 이러한 제약을 극복하기 위해 본 논문에서는 은닉 마르코프 모델(Hidden Markov model)을 이용해 동적인 교통 패턴에 따라 현재 상황에 맞는 회귀분석 모델을 선택하는 신뢰도 높은 교통량 예측 시스템을 제안한다.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.1621-1623
• /
• 2011

발전소 수명연장의 필요성과 더불어 발전기 수명예측기법이 발전기 제작사를 중심으로 연구되어 왔으며 국내에도 수명예측과 관련기술들이 주로 일본의 영향을 받아 도입되어 발전소 현장에 활용되고 있고 평가의 신뢰성을 높이기 위해 노력하고 있다. 본 연구에서는 국내에 적용되고 있는 발전기 수명예측방법과 적용기준을 살펴보고 실제 발전소를 대상으로 수명평가한 사례중에서 평가결과의 신뢰성에 문제가 될 수 있는 몇가지 사례를 예시하고 분석하였다. 그 결과로 수명예측의 신뢰성을 높이기 위해서는 발전기의 냉각방식에 따라 수명평가 기준이 달리 적용되어야 한다는 것과, 최대부분방전량 측정오차를 줄여야 한다는 것, 고정자권선 단말부 흡습에 대한 고려가 필요하다는 것 등을 제시하였다.

• Journal of Korean Society of Steel Construction
• /
• v.14 no.2
• /
• pp.365-373
• /
• 2002

In this paper, the system reliability concept was presented to predict the lifespan of bridges. Lifetime distribution functions (survivor functions) were used to model real bridges to predict their remaining life. Using the system reliability concept and lifetime distribution functions (survivor functions), a program called LIFETIME was developed. The survivor functions give the reliability of component at time t. The program was applied to an existing Colorado state highway bridge to predict the failure probability of the time-dependent system. The bridge was modeled as a system, with failure probability computed using time-dependent deteriorating models.