Statistical Methods for Reliability Studies

Index:

4001

Year:

2003

Thesis Title:

Statistical Methods for Reliability Studies

Student Name:

El Siddig Ahmed Ishag

Supervisor:

Dr. Mirghani Fath ElRahman

Department:

Electrical

Abstract

 

In a modern society, professional engineers and technical managers are responsible for the planning, design, manufacture tine/ operation of products and systems ranging from simple products to complex systems. the failure of these can often cause effects that range from inconvenience and irritation to severe impact on society and on environment. Users', customers and society in general expect that products and systems are reliable and safe. The question that arises is, How reliable or how safe will the system be during its future operating life? This Question can be answered, in part, by I the use of quantitative reliability evaluation. In consequence a considerable awareness has developed in the application of such techniques to the design and operation of simple and complex systems together with an increasing amount of legal requirements, including product liability aspects and statutory directives. Roy Billinton & Roland N. Allan This project is intended, beside the purpose of fulfillment of the master degree, to throw some light on an important subject that had had growing importance in recent days. Customers became aware by the quality of products and they expect to buy products that last longer. Besides competition in the free market made competitors try to stay in business by producing more reliable products that maintain customer satisfaction, hence reliability evaluation techniques have grown very rapidly in the last decades.

Since the occurrence of failure events in components and systems is of probabilistic nature, then using of probability and statistical methods found their way to become the most appropriate and favorable methods, because they enable many complex situations to be dealt with and analyzed. Also prediction of the future performance of components and systems became possible.

This Thesis is composed of seven chapters, through which (lie most essential and fundamental concepts of reliability studies were explored. Chapter one is an introductory chapter, in which basic ideas about the topic were highlighted and the importance of reliability studies and techniques was explained.

Chapter two dealt with the mathematical background, where basic probability and statistic concepts were introduced will) some detailing in the subject of probability distributions. The fundamental modeling techniques for the engineering systems were discussed in chapter three combined with network reduction techniques and also the types of systems were mentioned and their classification to series, parallel or complex was discussed.

Chapter four was devoted to the discussion of the usage of probability distributions in reliability evaluation. Some relevant topics such as maintenance, wear out, mean-time-lo-failure and their impact on reliability of systems were displayed. In chapter five, reliability indices that measure the systems' performance and quality of supply such as SAFI, SADI, CAIDI, ASAI, and ASUI were introduced, also the failure mode and effect analysis was discussed.

Economical considerations and the impact of cost upon reliability and vice versa, were presented in chapter six, in which the conflicting relation between reliability and cost was illustrated. Chapter seven constituted the application chapter, where all the preceding knowledge and techniques were utilized to evaluate the reliability of a selected substation of the Khartoum town distribution network. A five step procedure was followed lo evaluate the performance of this substation and some indices were calculated which showed the existing level of NEC quality of supply. Many lessons were learned from this application, and first and most important was the significance of data in any reliability study, secondly, data collection policies that should be established by companies, and thirdly, the significance of the analysis techniques.