Follow-Up & Survival Analysis Flashcards
7 cards from real CET practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 7 Follow-Up & Survival Analysis flashcards as text
Accelerated Life Testing (ALT) shortens test duration by:
Answer: Applying elevated stress levels to speed up failure mechanisms
ALT exposes units to stresses above normal operating levels (temperature, voltage, vibration) to cause failures faster while remaining in the same failure mode.
The Arrhenius model relates acceleration factor to which environmental stress?
Answer: Absolute temperature
The Arrhenius equation models thermally activated failure mechanisms; acceleration factor = exp[(Ea/k)(1/T1 − 1/T2)].
Reliability Growth Testing (Duane model) shows that system reliability improves as:
Answer: Failure modes are identified and design corrections are implemented
In the Duane model, cumulative MTBF grows as a power function of cumulative test hours because discovered failure modes are corrected.
In sequential (time-terminated) life testing, the test ends when:
Answer: A predetermined number of test hours has accumulated
A time-terminated test runs until a fixed total test time is reached, after which the observed failures are used to estimate MTBF.
A Failure Mode, Effects, and Criticality Analysis (FMECA) assigns a Risk Priority Number (RPN) based on:
Answer: Severity, occurrence probability, and detectability
RPN = Severity × Occurrence × Detection; higher RPN flags failure modes requiring immediate design attention.
Chi-squared (χ²) distribution tables are used in life testing to:
Answer: Construct confidence intervals on estimated MTBF
Two-sided χ² critical values at the chosen confidence level convert observed test hours and failures into MTBF confidence bounds.
Demonstrated MTBF in field data often differs from predicted MTBF because:
Answer: Predicted MTBF uses handbook failure rates that may not match actual application stresses
MIL-HDBK-217 and similar handbooks provide generic part failure rates that rarely match specific operating environments, causing discrepancy.