Introduction to Reliability Engineering & Warranty Analysis
Basic topics in reliability engineering and warranty analysis are presented. After an introduction to basic reliability concepts and reliability statistics the course shifts to methods for demonstrating reliability using success-based and test-to-failure strategies. Life data analysis (i.e., Weibull analysis) is covered along with a brief tutorial of Weibull++ software for performing these analyses. Reliability confidence bounds are reviewed along with their impact on test plans and customer reliability statements. With this footing, warranty analysis is introduced. After a brief review of the practical aspects of interpreting raw customer returns data, attendees will be introduced to four formats for inputting and analyzing this data using Weibull++ software. These include Nevada, times to failure, dates of failure, and the usage format. Methods for estimating usage of suspended units are covered. Finally, methods for modeling the data and making projections of future returns (actuals or percentages) is covered, including the impact of future sales volumes and warranty terms.
Topics Addressed:
- Bathtub curve, theoretical and actual
- Stress vs strength interference
- Product design life
- Mean time to failure (MTTF)
- Reliability functions, failure rate, cumulative distribution function (CDF), probability density function (PDF)
- Weibull, lognormal and exponential probability distributions
- Reliability block diagrams (RBD), serial and parallel systems
- Success testing strategies and sample size trade-offs
- Life data analysis and parameter estimation (Weibull analysis)
- Weibull++ software primer (ReliaSoft)
- One and two-sided Confidence bounds
- Interpreting reliability specifications
- Making sense of returns data
- Warranty analysis using Weibull++
- Nevada, times to failure, dates of failure and usage formats
- Modeling and estimating future returns
- Accounting for future sales
- Usage estimation methods for suspended parts: average and distributed use
- Projection estimates that account for warranty period (time and/or use)
Target Audience:
Warranty analysts, reliability engineers, validation engineers, and managers of these disciplines.
Delivery Method:
The training is designed to be interactive and conversational and delivered at the client’s location or remotely. A number of in-class exercises are included to practice and reinforce the concepts. Bound training materials are provided to each class participant. Weibull++ software will be introduced and used for many of the exercises; demo licenses are provided if needed.
Expected Outcomes:
At the conclusion of this training, class participants will be able to:
- Understand the life cycle of a product family in the field by way of the bathtub curve
- Understand the factors that influence the selection of warranty period and product design life
- Make trade-offs in sample size vs test duration for a success-based test
- Conduct a basic life data analysis and make inferences using Weibull++
- Select an appropriate confidence bound for your application
- Create and interpret customer and internal reliability specifications and identify ambiguities and shortcomings if they exist
- Conduct a basic warranty analysis: from raw returns data to modeling and estimating future returns