This site uses encryption for transmitting your passwords. ratmilwebsolutions.com
May 2012 Editorial
Volume 8, Number 3, May 2012
KRISHNA B. MISRA
In this issue, we bring to our readers some new upcoming ideas and interesting areas of research, which it is hoped will be of interest to our readers. This issue covers two reviews of recent books and 10 regular papers covering all several areas of performability. There are two important papers from Professor Aven of Stavanger University of Norway in the area of risk including the first paper in which, the authors demonstrate how conceptual pragmatism can be used as a suitable framework for probability models which can be used in risk assessment.
In the second paper of this issue from University of Toronto, Canada, the authors provide a wavelet transformation approach to detect early failure of the gear shaft, which can be of great help in fault detection and diagonostic anlysis of gear transmission systems. This technique can be of immense value to reliability and maintenance engineers.
The third paper of the issue is from Stavanger University, Norway, and discusses an important issue of the best available technique (BAT) in fulfilling the European Union's Integrated Pollution Prevention and Control (IPPC) directive towards sustainability considerations. The paper employs the criteria of life cycle costs as the tool in arriving at the best available technique from the several technically viable alternatives of system design, particularly at the initial stage of a project. A case study related to the selection of power supply systems for offshore oil and gas installations illustrates the method.
The fourth paper of this issue is from France and presents a methodology that helps designing a monitoring system particularly in cases where a model does not exist or it is difficult to find an analytical relationship between observed features of the system to identify the system state. The methodology enables to define and extend as much as possible the regions of feature space where a trained monitoring system can operate safely and precisely according to performance requirements. If the system experiences a situation which is characterized by features outside the trained region, the monitoring system is simply disabled and data may be recorded to update the monitoring system. The proposed methodology is applied to aircraft turbine start capability monitoring system to determine the extent of the feature space in which the monitoring system can operate. Some experiments on the application are also carried out.
The fifth paper of the issue is from Brazil and the authors present a case study on verification of time to breakdown of a new insulating fluid one of the objective of the paper is to verify if times to breakdown of insulating fluid between electrodes recorded at three different voltages have an exponential distribution as predicted by theory. The other objective o the paper is if the Eyring's acceleration model of can provide shape and scale parameters for an underlying Inverse Weibull model, obtained under two accelerating conditions. The Inverse Weibull model has been used in Bayesian reliability estimation to represent the information available about the shape parameter of an underlying Weibull sampling distribution.
The sixth paper of this issue is from three authors from Sweden, U.K. and Turkey, respectively and addresses an important problem of failure of railways turnouts, which may adversely affect the system availability, safety, operating and support costs. The paper presents a diagnostics method for 'drive-rod out-of-adjustment' failure mode - one of the most frequently observed failure modes. SVM with Gaussian kernel is used for failure classification. In addition, results of the feature selection with statistical t-test and feature reduction with principal component analysis are compared in the paper.
The seventh paper of the issue is from authors from Latvia and is a review of the authors' interesting previous work related to the analysis of tensile strength of uni-directionally fiber-reinforced composite material, considered as a series of parallel systems with defects. Additionally, a specific model is studied based on an assumption that only failure of the damaged parts of a specimen can take place. A version of the Poisson distribution is used for probability mass function of the number of defects. It is claimed by the authors that the proposed models allow estimating the probability of the presence of defects in order to clarify improvements of the production technology needed to increase reliability, and to predict the scale effect of tensile strength of the composite. Strength test data of different composite materials are processed and the results analyzed. A numerical comparison of different models is provided in the paper.
The eighth paper of this issue is again from Norway and discusses an analytic approach when using event trees and fault trees in a quantitative risk assessment context. The basic question raised is when to introduce probability models and frequentist probabilities (chances) instead of using direct probability assignments for the events of the trees. The author argues that such models should only be used if the key quantities of interest of the risk assessment are frequentist probabilities and when systematic information updating is important for meeting the aim of the analysis. An example of an event tree related to the analysis of an LNG (Liquefied Natural Gas) plant illustrates the analysis and discussion.
The ninth paper of the issue is from DRDO, India and presents a case study on vehicles used in the field. The authors use Artificial Neural Networks (ANN) approach for assessing reliability/ availability of vehicles. The methodology adopted is demonstrated with the help of a case study which includes collection, sorting and grouping of vehicle failure data. Then distribution parameters are estimated and best fitting probability distribution is identified for predicting vehicle reliability. Subsequently, the trained ANN (using SLP model) is used to predict the vehicle reliability. Suitability of a RDBMS (Oracle) for training ANN and predicting vehicle reliability is also presented. The developed methodology has been able to predict reliability of vehicle very close to its observed values.
The tenth paper of this issue is again from India and presents a case study on availability analysis of a bubble gum producing system. The paper develops a model which then used to assess availability using Markov method.