ARA Module 4 – Estimation of Reliability Metrics

of the Applied Reliability Analysis Course

Using Reliasoft Weibull++

This module is the core of life data analysis.  We start by providing an overview of the common methods for estimating the distribution parameters (e.g. Maximum Likelihood Estimation) and provide guidance for choosing a method.  Since quantifying uncertainty in our estimates is critical, we review the use of confidence intervals (or bounds) for this purpose.  Applying the concepts introduced previously, we learn how to utilize selected distributions to estimate reliability statistics of interest.  Treatment of censored data and multiple failure modes follow.  We also look at comparing multiple groups with respect to overall reliability performance.  Finally, we briefly look at nonparametric estimation which does not require any distribution assumption.  

A little background, motivation, full course overview, and a welcome from the instructor, Steven Wachs

Training Objectives

The key training full course objectives are summarized below (highlighted objectives relate to this module)

• Understand reliability concepts and unique aspects of reliability data
• Understand underlying probability and statistical concepts for reliability analysis
• Develop competency in the modeling and analysis of time-to-failure data
• Use and interpret probability plots for distribution fitting
• Understand reliability metrics and how to estimate and report them
• Handle multiple failure modes in reliability estimation
• Utilize degradation Data and models to predict failure times for life data analysis
• Use nonparametric estimation methods when appropriate
• Determine if reliability specifications are met (at specified confidence level) or whether design improvements are required
• Estimate estimate uncertainty using confidence intervals and bounds
• Estimate reliability of subsystems and systems
• Handle basic series and parallel systems
• Become aware of system reliability activities such as Reliability Block diagramming, Reliability importance, and Reliability allocation
• Develop competency in the planning of reliability tests (sample sizes)
• Use simulation to support estimation test planning
• Develop reliability demonstration test plans (testing time vs. number of units tradeoffs)
• Analyze existing warranty data to predict future returns
• Analyze data from Accelerated Life Tests to estimate reliability at normal use conditions (single stress, multiple stress, time-varying stress models)
• Plan Accelerated Life tests (sample sizes, determination of test stress combinations, allocation of units to stress levels)
• Incorporate degradation data modeling into Accelerated Life Test analysis
• Perform stress-strength analysis using analytical methods, software, and simulation
• Model reliability from binary response data (i.e. pass/fail)
• Analyze repair data from repairable systems to predict the number of failures over time, conditional reliability, and failure rates
• Develop proficiency in the use of Reliasoft Weibull++ for Reliability analyses

Steven Wachs, Course Instructor

Steven Wachs has 25 years of wide-ranging industry experience in both technical and management positions. Steve has worked as a statistician at Ford Motor Company where he has extensive experience in the development of statistical models, reliability analysis, designed experimentation, and statistical process control.

Steve is currently a Principal Statistician at Integral Concepts, Inc. where he assists manufacturers in the application of statistical methods to reduce variation and improve quality and productivity. He also possesses expertise in the application of reliability methods to achieve robust and reliable products as well as estimate and reduce warranty. Steve regularly speaks at industry conferences and provides workshops in industrial statistical methods worldwide.

 

Thorough knowledge of subject matter and ability to communicate it – exceptional instructor.

— Steve Fleishman, Magna Powertrain

---

Why is This Course Important?

Achieving high product reliability has become increasingly vital for manufacturers in order to meet customer expectations amid the threat of strong global competition.  Poor reliability can doom a product and jeopardize the reputation of a brand or company.  Inadequate online reliability training courses also present financial risks from warranty, product recalls, and potential litigation.  When developing new products, it is imperative that manufacturers develop reliability specifications and utilize methods to predict and verify that those reliability specifications will be met.  This presents a difficult challenge in many industries with short product cycles and compressed product development timeframes.  This course provides an in-depth treatment of quantitative methods for predicting and demonstrating product reliability from data gathered from physical testing or from field data.

Typical Attendees

• Product Engineers
• Design Engineers
• Quality Engineers
• Reliability Engineering Training
• Project / Program Managers
• Manufacturing Personnel
• Six Sigma Professionals

Enroll in the ARA Module 4 – Estimation of Reliability Metrics course

Steve is an excellent instructor who is very good on a dry topic. His application ability is great.

Bruce Schubert, BAE Systems