INDE 6336Reliability Engineering

Instructor: Dr. Qianmei (May) Feng

E217-D3, (713) 743-2870qmfeng@uh.edu

Handouts

1. Course Syllabus

2. Guideline to Course Project

3. Academic Honesty Statement

4. Student Information

Agenda

• Introduction to the course– What are the course grading rules?– What do you expect to learn from this course?

• Introduction to reliability engineering– Why is reliability important?– Why study reliability?– What is the objective of reliability engineering?– What is reliability?– What is the relationship between quality and reliability?– Where can we apply reliability engineering?

4

My Background

• Ph.D. in Industrial Engineering, University of Washington, Seattle, WA

• B.S. and M.S. in Manufacturing and Industrial Engineering, Tsinghua University, Beijing, China

• Teaching: Reliability Engineering, Quality Control, Engineering Statistics

• Research: Reliability and Quality Engineering (Supported by NSF and State of Texas)

• Past-President for QCRE Division in IIE

Introduce Yourself• Name:

• Dept & Program: (e.g. INDE, MSIE):

• Probability/Statistics classes (level and time):

• Relevant information: (e.g. experience in reliability/quality, certificates)

• Favorite memory during the holiday:

Course Overview

• Text: Ebeling, C.E. (2010)• Prerequisites

– Basic probability and statistics• Grading Policy:

Homework: 20%Exam 1: 30%Course Project/Research Paper: 15% Exam 2: 35% Total = 100%

Homework are due at the beginning of class on the due date

No late homework is acceptable

Course Objectives

• To understand basic concepts of reliability, maintainability, and availability

• To learn failure distributions and their properties

• To learn system reliability models, including series, parallel and complex systems

• To consider engineering design from the viewpoint of reliability

• To learn how to analyze reliability data

Course Topics• Introduction to reliability Ch. 1• Reliability measures Ch. 2• Probability models

– Exponential distribution Ch. 3– Weibull, Normal, Lognormal, Gamma Ch. 4

• Reliability of state-independent systems Ch. 5• State-dependent systems Ch. 6• Stress-strength model Ch. 7• Design for reliability Ch. 8• Reliability estimation

– Nonparametric estimation Ch.12– Parametric estimation Ch.15, Ch.16

• Reliability testing and reliability growth testing Ch.13, Ch.14

Review Questions

How many “Yes” do you get?1. Do you understand the difference between mean and median?2. Do you know what is skewness and what is Kurtosis?3. Are you familiar with Bayes' Theorem?4. If X follows normal distribution with mean and standard deviation

, do you know what is the probability of -3 < X < +3?5. Do you understand the relationship between exponential

distribution and Poisson distribution?6. Do you know the unique feature of exponential distribution?7. Have you heard of Weibull distribution?8. Are you familiar with Markov process?9. Can you derive the CDF (cumulative distribution function) given

the PDF (probability density function) for a continuous variable?10. Can you calculate the integral and derivative of e-x?

Why is Reliability Important? – Things Fail

• Have you experienced things fail in daily life?• More significantly:

• 1986 - Space Shuttle Challenger • The failure of the rubber O-rings which were used to seal the four

sections of the booster rockets.

• 2003 - Space Shuttle Columbia• Loss was a result of damage sustained during launch when a piece

of foam insulation the size of a small briefcase broke off the Space Shuttle external tank.

• 2007 - The Interstate 35W bridge collapsed• Failure of undersized steel gusset plates was reason for collapse.

Engineers who designed the bridge in the 1960s either failed to calculate or improperly calculated the thickness needed for the plates that were to hold the bridge together.

Attribute Average ScorePerformance 9.5Lasts long time (reliability) 9.0Service 8.9Easily Repaired (maintainability) 8.8Warranty 8.4Easy to Use 8.3Appearance 7.7Brand Name 6.3Packaging/Display 5.8Latest Model 5.4

Why is Reliability Important? – Gallup Survey

Why Study Reliability?

• Intense global competition and increased customer expectations Reduce product development cycle time Reduce product life-cycle costs Improve quality and reliability

• Rapid advances in technology More complex and sophisticated systems

More challenges for reliability

• Increasing pressure on manufacturers to produce products with high quality and reliability

What is the Objective of Reliability Engineering?

• Reliability engineering attempts to study, characterize, measure, and analyze the failure of systems in order to improve upon their operational use by increasing their design life, eliminating or reducing the likelihood of failures and safety risks, and reducing downtime thereby increasingavailable operating time.

Deterministic Failures

• Traditional approach to safety in engineering is to design into a product a high safety margin or safety factor• a deterministic method• Safety factors often result in

• Overdesign thus increasing costs or • Underdesign when an unanticipated load or a

material weakness results in a failure

Random Failures

• Approach taken in reliability engineering is to treat failures as random or probabilistic occurrences. • In theory, if we were able to comprehend the exact physics and

chemistry of a failure process, many internal failures of a component could be predicted with certainty.

• With limited data on the physical state of a component, and an incomplete knowledge of the physical, chemical (and perhaps biological) processes which cause failures, failures will appear to occur at random over time.

• This random process may exhibit a pattern which can be modeled by some probability distribution.

What is Reliability?

• A typical definition of reliability:

Reliability is the probability that a component or system will perform a required function for a given period of time when used under stated operating conditions

Quality and Reliability

• Quality: An inherent characteristic or property (Webster)

• Quality: State of acceptance of products or services that reflects customer satisfaction (Thomas and Richard, 2006)

• Reliability: State of being reliable– Time-oriented quality– Quality over time (Condra, 1993)– A vital element of quality

Quality Customer Satisfaction

Discussion of Reliability Definition

Reliability is a relative measure, which is relative to• Definition of failure from the viewpoint of customer

– Tail light out – Air brake chamber cracked– Rubber O-ring brittle

• Definition of time scale– Hours, cycles, miles

• Definition of operating conditions– Environment– Design loads

An Example of Reliability Definition

• Definition of failure– Windshield wiper arms do not work properly

• Definition of time scale– Cycles

• Definition of operating conditions– Frequency of your wipers– Number of rain days in your area

The reliability of windshield wiper arms at 100,000 cycles is 0.8.

A Measure of Reliability

Reliability is a probability that a system performs for a given period of time, and it is a function of “time”

• T = “time”-to-failure random variable

• Reliability at time t=100:

R(100) = P(T >= 100)

• More generally, reliability at time t:

R(t) = P(T >= t)

Reliability function

Reliability Data

• Life data

• Failure data

• Time-to-failure (TTF) data “time”

• Time-between-failure data

• Survival data

• Event-time data

Reliability Applications

1. Assess characteristics of materials (strength, fatigue crack, failure modes, …)

2. Assess the effect of a proposed design change3. Compare several different

manufacturers/suppliers/designs, …4. Predict product reliability in design stage5. Assess product reliability in field6. Predict product warranty costs7. Risk assessment for safety-critical systems8. Determine maintenance/repair/replacement schedules

Illustrative Applications

Case 1

In a large electronics company that manufactures small electrical parts such as resistors, capacitors, transistors, and inductors, a new component is experiencing a high failure rate. In order to meet government contract specifications, the function being performed by this component must have a 90 percent or better reliability over a 4-hour mission in a high-stress environment. Design engineers can not redesign the component because of the cost and time required. What should the company do?

Illustrative Applications (cont.)

Case 2The Notso Reliable Manufacturing Company has been experiencing a high number of failures with its five-year-old industrial robot used for arc welding. Once the unit fails, it is often down for what is considered to be an excessive length of time for repair. Downtime costs the company $750 an hour in lost production and repair costs. A replacement unit will cost $21,000. The company wishes to determine whether it is economical to replace the unit. The unit was advertised as having a 10-year design life. Should the company replace the unit?

• MaintainabilityThe probability that a failed component or system will be repaired to a specified condition within a period of time.

• AvailabilityThe probability that a component or system will perform its required function at a given point in timewhen used under stated operating conditions.

Reliability, Availability, and Maintainability

Relationships among RAM

Definition Purpose of study

Reliability Probability of a nonfailure over time

• Increasing design life• Eliminating or reducing the likelihood of failures

Maintainability Probability of repair in a given time

• Reducing downtime

Availability Percentage of operating time over a specified time interval (combined effect of both the failure and the repair process)

• Increasing available operating time

Course Overview

• Discuss reliability concepts and how to answer engineering questions if we know the underlying distributions (Probability models)

In practice, reliability distributions unknown

• How to estimate the distributions based on reliability data and then answer engineering questions of interest? (Statistical estimation)