Byeong Min Mun

Change-point detection in failure intensity: A case study with repairable artillery systems

Repairable systems can experience unexpected environmental changes over long operational periods. Such changes affect the incidence of failures, causing different system failure patterns before and after the changes. In this article, we propose an informational change-point approach for the pattern of recurrent failures in repairable artillery systems. Unlike other trend tests, this approach provides additional information about the locations of change-points over rates of occurrence of failures (ROCOFs) as well as failure trends. We adopt the modified information criterion (MIC) proposed by Pan and Chen (2006) to detect the locations of the changes and propose sequential procedures for determining the number of change-points in independent exponential sequences. The change-point approach is applied to unscheduled maintenance data from eight artillery system exercises performed by the Republic of Korea Army. The change-point test along with a graphical presentation of estimated ROCOF lines can provide easy interpretation of changes in failure trends/intensities in a homogeneous Poisson process.

A Superposed Log-Linear Failure Intensity Model for Repairable Artillery Systems

This article investigates complex repairable artillery systems that include several failure modes. We derive a superposed process based on a mixture of nonhomogeneous Poisson processes in a minimal repair model. This allows for a bathtub-shaped failure intensity that models artillery data better than currently used methods. The method of maximum likelihood is used to estimate model parameters and construct confidence intervals for the cumulative intensity of the superposed process. Finally, we propose an optimal maintenance policy for repairable systems with bathtub-shaped intensity and apply it to the artillery-failure data.

Optimal maintenance policy of repairable system with bathtub shaped intensity

The power law process (PLP) and log linear process (LLP) are able to describe the processes with monotonic trend during the operating time. However, these models are not able to describe the processes with a non-monotonic trend such as bathtub shaped intensity. In this paper we proposed the superposed log linear process (S-LLP) for modeling bathtub shaped intensity function. The S-LLP is a nonhomogeneous Poisson process (NHPP) that results from the superposition of two LLP with different parameters. To get the maximum likelihood (ML) estimates, we proposed the three-parameter log-likelihood function evaluated using a genetic algorithm. Also, to obtain approximated confidence intervals, the information matrix and the variance-covariance matrix have been evaluated. The suggested methods are applied to field data from a repairable system.

Data depth based support vector machines for predicting corporate bankruptcy

In financial distress analysis, the diagnosis of firms at risk for bankruptcy is crucial in preparing to hedge against any financial damage the at-risk firms stand to inflict. Some pre-alarm signals that indicate a potential financial crisis exist when a firm faces a default risk. Early studies on corporate bankruptcy prediction include parametric and nonparametric approaches, such as artificial intelligence (AI), for detecting pre-alarm signals. Among nonparametric techniques, the methods involving support vector machine (SVM) have shown potential in predicting corporate bankruptcy. We propose a hybrid method that combines data depths and nonlinear SVM for the prediction of corporate bankruptcy. We employed data depth functions to condense multivariate financial data with nonlinear and non-normal characteristics into one-dimensional space. The SVM method was introduced to classify the data points on a depth versus depth plot (DD-plot). Based on data set that records failed and non-failed manufacturing firms in Korea over 10 years, the empirical results demonstrated that the proposed method offers a higher level of accuracy in corporate bankruptcy prediction than existing methods. The proposed method is expected to provide a guidance in corporate investing for investors or other interested parties.

Condition monitoring of a steam turbine generator using wavelet spectrum based control chart

Abstract Condition-based maintenance (CBM) is designed to take maintenance actions only when there is an imminent evidence of failure for a monitoring system. The parameters indicating health status of the system are continuously monitored in CBM. This article proposes a condition monitoring scheme based on energy profiles generated from wavelet spectrum analysis. The energy of time series is represented by a wavelet spectrum in scale representations of signals. After deriving wavelet spectrums using a discrete wavelet transform at pre-specified windows, we aim to monitor the system based on multivariate T 2 chart for the parameters in the linear energy profiles. The monitoring scheme is applied to temperature signals measured from a steam turbine generator. The proposed T 2 chart based on the energy profiles shows a potential in early detecting the abnormality of a monitoring system which is not clearly detectable in original time scales.

Informal Quality Data Analysis via Sentimental analysis and Word2vec method

Purpose: This study analyzes automobile quality review data to develop alternative analytical method of informal data. Existing methods to analyze informal data are based mainly on the frequency of informal data, however, this research tries to use correlation information of each informal data. Method: After sentimental analysis to acquire the user information for automobile products, three classification methods, that is, naïve Bayes, random forest, and support vector machine, were employed to accurately classify the informal user opinions with respect to automobile qualities. Additionally, Word2vec was applied to discover correlated information about informal data. Result: As applicative results of three classification methods, random forest method shows most effective results compared to the other classification methods. Word2vec method manages to discover closest relevant data with automobile components. Conclusion: The proposed method shows its effectiveness in terms of accuracy and sensitivity on the analysis of informal quality data, however, only two sentiments (positive or negative) can be categorized due to human errors. Further studies are required to derive more sentiments to accurately classify informal quality data. Word2vec method also shows comparative results to discover the relevance of components precisely.

Literature Review on the Reliability in KSQM for 50 Years

Purpose: This paper reviews the papers on reliability issues which are published in the Journal of the Korean Society for Quality Management (KSQM) since 1965. The literature review is purposed to survey a variety of reliability issues for several categories Methods: We divide all of reliability iss...