RELIABILITY PREDICTIONS IN EARLY DESIGN STAGE USING EXPERT OPINION BASED METHOD
USER'S INSTRUCTIONS: The project work you are about to view is on "reliability predictions in early design stage using expert opinion based method" Please, sit back and study the below research material carefully. This project topic "reliability predictions in early design stage using expert opinion based method" have complete 5(five) Chapters. The complete Project Material/writeup include: Abstract + Introduction + etc + Literature Review + methodology + etc + Conclusion + Recommendation + References/Bibliography.Our aim of providing this "reliability predictions in early design stage using expert opinion based method" project research material is to reduce the stress of moving from one school library to another all in the name of searching for "reliability predictions in early design stage using expert opinion based method" research materials. We are not encouraging any form of plagiarism. This service is legal because, all institutions permit their students to read previous projects, books, articles or papers while developing their own works.
TITLE PAGE
RELIABILITY PREDICTIONS IN EARLY DESIGN STAGE USING EXPERT OPINION BASED METHOD
BY
---
EE/H2013/01430
DEPARTMENT OF ---
INSTITUTE OF ---
DECEMBER,2018
APPROVAL PAGE
This is to certify that the research work, "reliability predictions in early design stage using expert opinion based method" by ---, Reg. No. EE/H2007/01430 submitted in partial fulfillment of the requirement award of a Higher National Diploma on Electrical and Electronics Engineering has been approved.
By
Dr. --- Dr. ---
Supervisor Head of Department.
Signature………………. Signature……………….
……………………………….
Dr. ---
External Invigilator
DEDICATION
This project is dedicated to Almighty God for his protection, kindness, strength over my life throughout the period and also to my --- for his financial support and moral care towards me.Also to my mentor --- for her academic advice she often gives to me. May Almighty God shield them from the peril of this world and bless their entire endeavour Amen.
ACKNOWLEDGEMENT
The successful completion of this project work could not have been a reality without the encouragement of my --- and other people. My immensely appreciation goes to my humble and able supervisor mr. --- for his kindness in supervising this project.
My warmest gratitude goes to my parents for their moral, spiritual and financial support throughout my study in this institution.
My appreciation goes to some of my lecturers among whom are Mr. ---, and Dr. ---. I also recognize the support of some of the staff of --- among whom are: The General Manager, Deputy General manager, the internal Auditor Mr. --- and the ---. Finally, my appreciation goes to my elder sister ---, my lovely friends mercy ---, ---, --- and many others who were quite helpful.
PROJECT DESCRIPTION: This work “reliability predictions in early design stage using expert opinion based method” research material is a complete and well researched project material strictly for academic purposes, which has been approved by different Lecturers from different higher institutions. We made Preliminary pages, Abstract and Chapter one of “reliability predictions in early design stage using expert opinion based method” visible for everyone, then the complete material on “reliability predictions in early design stage using expert opinion based method” is to be ordered for. Happy viewing!!!
In the past, reliability is usually quantified with sufficient information available. This is not only time-consuming and cost-expensive, but also too late for occurred failures and losses. For solving this problem, the objective of this dissertation is to predict product reliability in early design stages with limited information. The current research of early reliability prediction is far from mature. Inspired by methodologies for the detail design stage, this research uses statistics-based and physics-based methodologies by providing general models with quantitative results, which could help design for reliability and decision making during the early design stage. New methodologies which accommodate component dependence, time dependence, and limited information are developed in this research to help early accurate reliability assessment. The component dependence is considered implicitly and automatically without knowing component design details by constructing a strength-stress interference model. The time-dependent reliability analysis is converted into its time-independent counterpart with the use of the extreme value of the system load by simulation. The effect of dependent interval distribution parameters estimated from limited point and interval samples are also considered to obtain more accurate system reliability. Optimization is used to obtain narrower system reliability bounds compared to those from the traditional method with independent component assumption or independent distribution parameter assumption. With new methodologies, it is possible to obtain narrower time-dependent system reliability bounds with limited information during early design stages by considering component dependence and distribution parameter dependence. Examples are provided to demonstrate the proposed methodologies.
In this report we present some results on the formal treatment of expert opinion in reliability analysis. We discuss a procedure which uses opinion of one or two experts for undertaking the reliability assessment of a component. Then we develop the methodology for point processes broadly used in the analysis of defect and count data
TABLE OF CONTENTS
COVER PAGE
TITLE PAGE
APPROVAL PAGE
DEDICATION
ACKNOWELDGEMENT
ABSTRACT
CHAPTER ONE
1.0 INTRODUCTION
1.1 BACKGROUND OF THE STUDY
- PROBLEM STATEMENT
- OBJECTIVE OF THE STUDY
- SCOPE OF THE STUDY
- SIGNIFICANCE OF THE STUDY
- DEFINITION OF TERMS
- PROJECT ORGANISATION
CHAPTER TWO
LITERATURE REVIEW
- INTRODUCTION
- OVERVIEW OF THE STUDY
- REVIEW OF STANDARD RELIABILITY METHODS
- TIME-TO-FAILURE USING BAYESIAN ANALYSIS
- OVERVIEW OF DESIGN FOR RELIABILITY (DFR)
- IMPORTANT OF DESIGN FOR RELIABILITY (DFR)
- THE DESIGN FOR RELIABILITY PROCESS
- HISTORICAL BACKGROUND OF RELIABILITY
- RELIABILITY REQUIREMENT
- RELIABILITY PREDICTION AND IMPROVEMENT
CHAPTER THREE
3.0 METHODOLOGY
3.1 INTRODUCTION
3.2 PROPOSED BAYESIAN MODEL
CHAPTER FOUR
4.0 RESULT
4.1 BEHAVIOR OF PRIOR DISTRIBUTIONS ON SCALE PARAMETER
- APPLICATION OF THE PROPOSED BAYESIAN ESTIMATION MODEL
CHAPTER FIVE
- CONCLUSION
- REFERENCES
CHAPTER ONE
1.0 INTRODUCTION
Reliability is the probability that a product performs its intended function under specified conditions during a specified period of time [1]. In the past, reliability analysis had been primarily regarded as a study of failures and failure time data of products, meaning a product’s reliability could be quantified only after observing field failure data and/or life testing results. This is often too late due to risks and losses that have already occurred. Nowadays, reliability is viewed as an important criterion of product performance. Research indicates that the major product performance and up to 70% of the product cost are determined in early design stages [2]. With the trend of design for reliability in modern industries, reliability analysis as early as in conceptual design stages is imperative.
Progress has been made in reliability prediction during early stages, but many questions still need answers. In the conceptual design stage, reliability information is sparse or may not be available. Thus, it is hard to obtain quantitative reliability results. A series of methodologies in qualitative reliability prediction have been developed by Tumer’s and Stone’s groups based on function modeling [3-6]. Function modeling is an important stage for generating design concepts during conceptual design. The overall function is created first and is then decomposed into a number of sub-functions. Solutions are sought to realize the sub-functions. Design concepts are then generated from the solutions. The key to high reliability is to make sure that the design concepts generated have sufficient intrinsic reliability. Function modelling based methodologies, which enable the early reliability analysis mainly in a qualitative way after the product functions are determined, are more or less subjective.
Besides qualitative methods discussed above, relative reliability measures are also provided, whose objective is to rank design concepts with quantitative reliability indexes. A good attempt is the development of the Relative Reliability Risk Assessment (R3I) method [7]. In the conceptual design stage, though limit information is available, quantitative reliability prediction is usually more preferred. Traditional reliability approaches, such as Failure Modes and Effects Analysis (FMEA) and Fault Tree Analysis (FTA), often restrict the information to what is obtained from current product testing data, and they often result in unseasonable results, such as 1 or 0 for reliability. Bayesian approaches [8-11] are proposed to use in early design stages and perform better than the traditional methods because all the information available can be used, no matter if it is old or new, objective or subjective, or point or interval values. However, the application of Bayesian models is sensitive to the appropriate prior distributions.
Due to the lack of computational models during the early design stage, physical- based methods are rarely used. Recently, there was an attempt to extend one of the physical-based reliability strategies, the stress and strength interference theory, to the reliability analysis in conceptual design. The method is called the conceptual stress and conceptual strength interference theory (CSCSIT) [12]. The CSCSIT method is a good attempt to use physics-based methodologies in product early design stage; however, it did not consider the issues of component dependence and time dependence.
From the state-of-the-art, we see that the research on early design reliability methodologies has progressed in spite of the challenges and is gaining more attention.
The methodologies, however, still have their limitations, and the research in reliability prediction during early design stages is far from mature. Even though the challenges are formidable, they undoubtedly provide great opportunities of exploring new ways to deal with reliability in conceptual design.
There are different methods for reliability predictions based on product’s life cycle (Coit 2000). In the design phase even the best computer systems can’t estimate the reliability of the product but some systematic methods have been implemented to predict reliability (Fajdiga and Jurejevcic 1996). To have an accurate and credible reliability prediction, it is important to have data available early in the product life (Usher et al. 1990). Prediction can be made only based on field data that available early after production but due to the information absence at this stage, it is more likely for statistical uncertainty to happen. In other words when sufficient data is not available, using common statistical methods for predicting reliability, may provide inaccurate results (Siu and Kelly 1998). In that case Bayesian estimation method is more appropriate to obtain acceptable results (Ion et al. 2006).
The main objective of this paper is to construct a methodology to estimate reliability of evolutionary product early after introduction to market. For this reason a novel model is introduced to merge limited failure information available for new product with historical data of previous generation of that product to have more realistic prediction of failure. Bayesian statistic is used to join prior information with sparse few field data on current product’s performance Bayesian method is one of the expert based methods. Different from other studies, our model provides more accurate estimation that cover existing uncertainty.
1.1 BACKGROUND OF THE STUDY
Reliability prediction has become an important area of research for many decades. Many studies have been done to develop quantitative approaches for reliability estimation (Blanks 1998; Johannesson et al. 2013). When a new product has a long life time or it quickly moves from design phase to production line, accelerated life test is used to predict reliability (Nelson 1980). Another way for reliability prediction is physics of failure models such as block diagram or fault tree analysis (Denson 1998). With today’s computers’ development using computational methods in the field of reliability prediction increase. Monte Carlo simulation (Ormon et al. 2002), Neural Networks (Lolas and Olatunbosun 2008; Tong and Liang 2005), Support vector machine (SVM) (Lins et al. 2012) and hierarchical methods like Genetic Algorithms (Liang 2008) develop for this purpose. None of these methods consider uncertainty exist in the development phase of product’s life cycle. Also information exist from previous product generation does not take in to account.
Bayesian models have been used widely in different area of research studies (Singh et al. 2001). The motivation for almost any statistical analyses is that some target population is not well understood or some aspects of it are unknown or unsure. So the idea of Bayesian methods is to say that any uncertainty can be modeled in a probabilistic way. Yadav et al. (2003) suggest a general framework for reliability prediction in the development phase of product based on Bayesian methods. Pan (2009) presents a new model for reliability prediction, combined accelerated life test information and new product field data using the Bayesian approach. Houben et al. (2009) model reliability prediction of product change through Bayesian inference. Wang et al. (2013) develop a Bayesian model for condition monitoring of units with heterogeneous degradation rates and imperfect inspections. Ye et al. (2013) propose a Bayesian evaluation method to integrate the Accelerated degradation testing (ADT) data from laboratory with the field failure data. The Recent study by Alzbutas and Iešmantas (2014) present an application of Bayesian methods for age- dependent reliability analysis. Although these approaches try to solve uncertainty problem exist in early life reliability prediction by using Bayesian methods but none of them consider past generation reliability information for model building. Stochastic modelling for degradation of highly reliable products has been proposed (Ye and Xie 2015).
Bayesian method has been developed for product reliability prediction using Weibull distribution (Canavos et al. 1973). Ion et al. (2006) suggests a Bayesian model for product that time to failure has Weibull distribution while scale and shape parameters has normal distribution. Even though this method overcome some of the problems exist in previous model but it is not ideal. The reason is in Weibull distribution scale and shape parameters both are greater than zero while normal distribution covers all real numbers both negative and positive. In this paper we develop new Bayesian model to predict reliability of evolutionary product when time to failure has Weibull distribution and scale parameter of it follow different prior distribution. To best of our knowledge this study is first attempt to build a Bayesian model for reliability prediction in early life of evolutionary product using both historical information and available failure data of new product.
1.2 PROBLEM STATEMENT
In engineering field specifically in manufacturing sector, every component of device produce have a designed period of time to expire which is also called failure. This period of failure is given by the producer by prediction which means that this prediction can be under-prediction or over-prediction, and this can lead to unexpected components failure which is said to affect the system design reliability. In order to improve in this area, a Design for Reliability (DfR) study was carried out which is a process that encompasses tools and procedures to ensure that a product meets its reliability requirements, under its use environment, for the duration of its lifetime. DfR is implemented in the design stage of a product to proactively improve product reliability.
1.3 RESEARCH OBJECTIVE
The objective of this research is to predict product reliability in early design stages using Bayesian model which is one of expert opinion based methods. With the predicted reliability, the research results can help engineers reduce the likelihood of failures to an acceptable level before the test of manufactured products or field deployment.
1.4 SCOPE OF THE STUDY
The application of the Bayesian model which is one of expert opinion based methods has approach to problem identification and reliability evaluation of structural engineering systems is described in this paper. The paper explains potential application of the method to structural system reliability evaluation and presents a specific case study on the reliability analysis of a utility network system. The case study used in this work is the study from automotive industry shows a battery failure data for a specific vehicle’s model
1.5 SIGNIFICANCE OF THE STUDY
This study enables engineers to understand how dependence affects the reliability prediction in early design stages and how to predict system reliability efficiently with good accuracy. With the accurate system reliability prediction in early design stages, this dissertation will enhance system designs in decision making with respect to system configurations, optimization, lifecycle cost, maintenance, and warranty.
1.6 DEFINITION OF TERMS
Many of the terms used in this paper are defined differently depending on their application and area of research. To reduce ambiguity and any uncertainty that goes along with this issue, a list of definitions/descriptions of relevant terms is provided here.
Failureof a system or design is described using functionality. The relationship between failure and function is natural because function is derived using customer requirements and failure is defined by the customer. For this reason, a failure is a partial or complete loss of functionality in a design during a specified life [6].
- Failure rate λ is a commonly used and well accepted variable for risk and reliability calculations. In general, λ is recorded in units of Failure/Million Hours or Failure/Million Miles. This depends on the source and how the data was collected. Failure rate is the number of failures divided by a specified time increment where the design was in operation.
- Reliabilityis defined at the probability that a failure will not occur. There exist several formulations to calculate reliability, some of which are shown in later sections. The inclusion of failure rates and time as variables is a common characteristic among all different reliability formulations. Reliability is always reduced as the failure rate and time increase.
- Early design processcan be defined as any step up to the selection of a final concept. Detailed design is not considered to be part of early design. Specifically, problem definition, requirements formulation, functional modeling, concept generation, and concept selection are considered as elements of early design.
- Component or candidate solution is used interchangeably in this research to represent the solution to a function. These have physical structure and represent the source of the data (evidence) collected.
- The terms design concept and design alternativeare used interchangeably to refer to the combination of components (or candidate solutions) which form a design. These synonymous terms do not indicate any feasibility to show that the components in the design are capable of being physically connected.
1.7 PROJECT ORGANISATION
The work is organized as follows: chapter one discuses the introductory part of the work, chapter two presents the literature review of the study, chapter three describes the methods applied, chapter four discusses the results of the work, chapter five concludes the research outcomes.
CHAPTER ONE: The complete chapter one of “reliability predictions in early design stage using expert opinion based method” is available. Order full work to download. Chapter one of “reliability predictions in early design stage using expert opinion based method”consists of the literature review. In this chapter all the related work on“reliability predictions in early design stage using expert opinion based method” was reviewed.
CHAPTER TWO: The complete chapter two of “reliability predictions in early design stage using expert opinion based method” is available. Order full work to download. Chapter two of “reliability predictions in early design stage using expert opinion based method” consists of the literature review. In this chapter all the related work on “reliability predictions in early design stage using expert opinion based method” was reviewed.
CHAPTER THREE: The complete chapter three of “reliability predictions in early design stage using expert opinion based method” is available. Order full work to download. Chapter three of “reliability predictions in early design stage using expert opinion based method” consists of the methodology. In this chapter all the method used in carrying out this work was discussed.
CHAPTER FOUR: The complete chapter four of “reliability predictions in early design stage using expert opinion based method” is available. Order full work to download. Chapter four of “reliability predictions in early design stage using expert opinion based method” consists of all the test conducted during the work and the result gotten after the whole work
CHAPTER FIVE: The complete chapter five of “reliability predictions in early design stage using expert opinion based method” is available. Order full work to download. Chapter five of “reliability predictions in early design stage using expert opinion based method” consist of conclusion, recommendation and references.
To "DOWNLOAD" the complete material on this particular topic above click "HERE"
Do you want our Bank Accounts? please click HERE
To view other related topics click HERE
To "SUMMIT" new topic(s), develop a new topic OR you did not see your topic on our site but want to confirm the availiability of your topic click HERE
Do you want us to research your new topic? if yes, click "HERE"
Do you have any question concerning our post/services? click HERE for answers to your questions
You can also visit our facebook Page at fb.me/hyclas to view our related construction (or design) picture.
For more information contact us through Any of the following means:
Mobile No :+2348146561114 or +2347015391124 [Mr. Innocent]
Email address :engr4project@gmail.com
Watsapp No :+2348146561114
To View Our Design Pix: You can also visit our facebook Page at fb.me/hyclas for our design photos/pics.
COUNTRIES THAT FOUND OUR SERVICES USEFUL
Australia, Botswana, Canada, Europe, Ghana, Ireland, India, Kenya, Liberia, Malaysia, Namibia, New Zealand, Nigeria, Pakistan, Philippines, Singapore, Sierra Leone, South Africa, Uganda, United States, United Kindom, Zambia, Zimbabwe, etc
Support: +234 8146561114 or +2347015391124
Watsapp No :+2348146561114
Email Address :engr4project@gmail.com
IF YOU ARE SATISFIED WITH OUR SERVICES, PLEASE DO NOT FORGET TO INVITE YOUR FRIENDS AND COURSEMATES TO OUR PAGE.