• Journal of the Korea Safety Management & Science
• /
• v.17 no.2
• /
• pp.241-247
• /
• 2015

The purpose of the warranty data analysis can be classified into two categories. Two goals is a failure cause analysis and life prediction analysis. In this paper first, we applied multivariate analysis method that can be estimated in consideration of various factors on the failure cause warranty data. In particular, we apply the Tree model and Cox model. The advantage of the Tree is easy to interpret this result as compared to other models. In addition Cox model can quantitatively express the risk. Second, this paper proposed a multivariate life prediction model (AFT) considering a variety of factors. By applying the actual warranty data confirmed the usability.

• Journal of Industrial Technology
• /
• v.27 no.B
• /
• pp.255-261
• /
• 2007

Product reliability plays a significantly important role these days. This study has been performed to examine the widely being used specifications, MIL-HDBK-217 and SR-332 for electronic parts. We specifically selected an electronic ballast of the low wattage fluorescent lamp for the study. We briefly reviewed the reliability specifications with the basic concepts of the ballast. We then valuated failure rates of the parts using MIL-HDBK-217 and SR-332 specifications. Since the quality and environment factor values are subjectively determined for failure rate evaluations, we excluded them for comparison.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.5 no.3
• /
• pp.141-152
• /
• 2001

This paper presents the evaluation of behavior and the prediction of tensile capacity of anchors that can cause a failure of the concrete on the basis of the design for anchorage. Tests of cast-in-place headed anchors, domestically manufactured and installed in uncracked and unreinforced concrete member are conducted to test the effected of embedment length and edge distance. The failure modes and the load-deformation responses of the anchors are discussed and then the concrete failure data are compared with capacities by the two present methods : the 45 degree cone method of ACI 349, 318 and the concrete capacity design (COD) method. Differences between the results by test and by two prediction methods are analyzed Finite Element Method (FEM).

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 2008.05a
• /
• pp.556-559
• /
• 2008

In this study, the failure in circular cup deep drawing simulation at warm temperature is predicted using forming limit diagram (FLD). The FLD is used in sheet metal forming analysis to determine the criterion for fracture prediction. The simulation with heat transfer of circular cup deep drawing at warm temperature was conducted. To predict the failure, the simulation with heat transfer used FLD at temperature in the vicinity of maximum thinning. The result of the simulation with heat transfer shows that the drawn depth increases with increasing temperature and is in accord with the experimental results above $150^{\circ}C$. The FLD provides a good guide for the failure prediction of warm forming simulation with heat transfer.

• Journal of the Korean Society for Precision Engineering
• /
• v.15 no.11
• /
• pp.218-226
• /
• 1998

In this study a neural network based expert system designed to diagnose operating status of a rotating spindle system supported by ball bearings was introduced. In order to facilitate practical failure situations, five exemplary abnormal status was fabricated. Out of several possible data source locations, seven most effective spots were chosen and proven to be the most successful in predicting single and multiple abnormalities. Increased signal strength was measured around where abnormality was embedded. Signal mea-surement locations producing high prediction rate were also classified. Even though multiple abnormalities were hard to be decoupled into their individual causes, proposed diagnostic system was somewhat effective in predicting such cases under certain combination of sensor locations. Among several abnormal operating conditions, highest prediction rate can be expected when signal is spoiled by the failure or damage in outer race. Proposed diagnostic system was again proven to be the most effective system in analyzing and ranking the importance of data sources.

• Communications for Statistical Applications and Methods
• /
• v.17 no.1
• /
• pp.99-106
• /
• 2010

In this paper, we present two methods for obtaining prediction intervals for the times to failure of units censored in multiple stages in a progressively censored sample from proportional hazard rate models. A numerical example and a Monte Carlo simulation study are presented to illustrate the prediction methods.

• Proceedings of the KSR Conference
• /
• 2007.05a
• /
• pp.255-261
• /
• 2007

In this paper we introduce the method, procedure and result of RAM prediction for interlocking system which is applied for AREX signaling system. For RAM prediction, we breakdown the interlocking system to the LRU level and select the LRUs of which failure can cause the service delay. The prediction of reliability is based on the Reliability Block Diagram which is the functional diagram composed of selected LRUs and finally, the availability of interlocking system is estimated from the combination of reliability and maintainability.

• Structural Engineering and Mechanics
• /
• v.35 no.5
• /
• pp.571-592
• /
• 2010

An energy-based fatigue life prediction framework was previously developed by the authors for prediction of axial, bending and shear fatigue life at various stress ratios. The framework for the prediction of fatigue life via energy analysis was based on a new constitutive law, which states the following: the amount of energy required to fracture a material is constant. In the first part of this study, energy expressions that construct the constitutive law are equated in the form of total strain energy and the distortion energy dissipated in a fatigue cycle. The resulting equation is further evaluated to acquire the equivalent stress per cycle using energy based methodologies. The equivalent stress expressions are developed both for biaxial and multiaxial fatigue loads and are used to predict the number of cycles to failure based on previously developed prediction criterion. The equivalent stress expressions developed in this study are further used in a new finite element procedure to predict the fatigue life for two and three dimensional structures. In the second part of this study, a new Quadrilateral fatigue finite element is developed through integration of constitutive law into minimum potential energy formulation. This new QUAD-4 element is capable of simulating biaxial fatigue problems. The final output of this finite element analysis both using equivalent stress approach and using the new QUAD-4 fatigue element, is in the form of number of cycles to failure for each element on a scale in ascending or descending order. Therefore, the new finite element framework can provide the number of cycles to failure at each location in gas turbine engine structural components. In order to obtain experimental data for comparison, an Al6061-T6 plate is tested using a previously developed vibration based testing framework. The finite element analysis is performed for Al6061-T6 aluminum and the results are compared with experimental results.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.38 no.4
• /
• pp.142-148
• /
• 2015

Due to advancements in technology and manufacturing capability, it is not uncommon that life tests yield no or few failures at low stress levels. In these situations it is difficult to analyse lifetime data and make meaningful inferences about product or system reliability. For some products or systems whose performance characteristics degrade over time, a failure is said to have occurred when a performance characteristic crosses a critical threshold. The measurements of the degradation characteristic contain much useful and credible information about product or system reliability. Degradation measurements of the performance characteristics of an unfailed unit at different times can directly relate reliability measures to physical characteristics. Reliability prediction based on physical performance measures can be an efficient and alternative method to estimate for some highly reliable parts or systems. If the degradation process and the distance between the last measurement and a specified threshold can be established, the remaining useful life is predicted in advance. In turn, this prediction leads to just in time maintenance decision to protect systems. In this paper, we describe techniques for mapping product or system which has degrading performance parameter to the associated classical reliability measures in the performance domain. This paper described a general modeling and analysis procedure for reliability prediction based on one dominant degradation performance characteristic considering pseudo degradation performance life trend model. This pseudo degradation trend model is based on probability modeling of a failure mechanism degradation trend and comparison of a projected distribution to pre-defined critical soft failure point in time or cycle.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.4
• /
• pp.227-233
• /
• 2021

Predictive maintenance has been one of important applications of data science technology that creates a predictive model by collecting numerous data related to management targeted equipment. It does not predict equipment failure with just one or two signs, but quantifies and models numerous symptoms and historical data of actual failure. Statistical methods were used a lot in the past as this predictive maintenance method, but recently, many machine learning-based methods have been proposed. Such proposed machine learning-based methods are preferable in that they show more accurate prediction performance. However, with the exception of some learning models such as decision tree-based models, it is very difficult to explicitly know the structure of learning models (Black-Box Model) and to explain to what extent certain attributes (features or variables) of the learning model affected the prediction results. To overcome this problem, a recently proposed study is an explainable artificial intelligence (AI). It is a methodology that makes it easy for users to understand and trust the results of machine learning-based learning models. In this paper, we propose an explainable AI method to further enhance the explanatory power of the existing learning model by targeting the previously proposedpredictive model [5] that learned data from a core facility (Hyper Compressor) of a domestic chemical plant that produces polyethylene. The ensemble prediction model, which is a black box model, wasconverted to a white box model using the Explainable AI. The proposed methodology explains the direction of control for the major features in the failure prediction results through the Explainable AI. Through this methodology, it is possible to flexibly replace the timing of maintenance of the machine and supply and demand of parts, and to improve the efficiency of the facility operation through proper pre-control.