To complete my master’s degree in statistics/mathematics at SIUE, I presented my master’s project in October 2023. There was also a paper associated with the project titled “Reliability Estimation in Series Systems: Maximum Likelihood Techniques for Right-Censored and Masked Failure Data”.
Overview
The project addresses a common challenge in reliability engineering: estimating component failure rates when:
- Masked failure data: You observe system failure but don’t know which component failed
- Right censoring: Some systems are still operational when observation ends
These data limitations are ubiquitous in real-world reliability studies, where identifying the exact failed component may be expensive or impossible.
Key Contributions
- Likelihood-based framework for handling both masking and censoring simultaneously
- Weibull distribution modeling with closed-form Fisher information for the exponential special case
- R package implementation providing accessible tools for practitioners
- Bootstrap methods for uncertainty quantification
Related Work
This project connects to several other posts and projects:
- Closed-Form Results for Masked Exponential Series Systems - The exponential distribution special case with analytical solutions
- likelihood.model R package - Software implementation
See the full project page here.
Discussion