Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Reliability›Degradation Models
Regression modelReliability & risk

Degradation Models

Degradation Models (Accelerated Degradation) · Also known as: Accelerated Degradation Testing, Degradation Path Models, Performance Degradation Analysis, Bozunma Modelleri

Degradation models estimate product lifetime by tracking measurable performance characteristics—such as crack length, light output, or insulation resistance—over time rather than waiting for outright failure. Introduced in rigorous form by Meeker, Escobar, and Lu (1998), these models fit a stochastic degradation path to repeated measurements and define failure as the first time the characteristic crosses a predetermined threshold, enabling reliable lifetime inference from accelerated test data with very few or no observed failures.

ScholarGate
  1. Regression model
  2. v1
  3. 1 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

Degradation Models
Maintenance OptimizationReliability AnalysisWeibull Regression

When to use it

Use degradation models when units rarely or never fail during a feasible test period yet exhibit a measurable, monotone performance decline. Key assumptions are that the degradation path follows a parametric form, the critical failure threshold is physically meaningful and pre-specified, and degradation measurements are obtainable without destroying the unit. The method is less appropriate when degradation is non-monotone, when multiple competing failure modes exist, or when no suitable measurable characteristic can be identified. Alternatives include accelerated life testing (when failures occur) and Wiener-process or gamma-process models for non-linear stochastic degradation.

Strengths & limitations

Strengths
  • Enables lifetime inference from tests with few or zero observed failures
  • Makes efficient use of continuous degradation measurement data rather than only pass/fail outcomes
  • Allows extrapolation from accelerated stress levels to normal use conditions through well-defined acceleration factors
  • Provides physically interpretable model parameters tied to the actual degradation mechanism
Limitations
  • Requires a measurable degradation characteristic that correlates reliably with the failure mode of interest
  • Parametric path assumptions (linearity, power-law) can be misspecified, leading to biased lifetime extrapolations
  • Extrapolation to use conditions assumes the acceleration model holds across the entire stress range tested
  • Threshold d_f must be defined a priori; wrong threshold choice directly distorts all lifetime estimates

Frequently asked

How many test units are needed for a reliable degradation model?

There is no universal rule, but Meeker and Escobar recommend at least 10–20 units per stress level to estimate both fixed path parameters and unit-to-unit random effects with adequate precision. Fewer units lead to wide confidence intervals on lifetime quantiles, especially in the tails. Sample size planning software or simulation studies are advisable before committing to a test plan.

Can degradation models handle non-linear degradation paths?

Yes. The mean path μ(t; β) can be any parametric function—power-law, exponential, logarithmic, or Arrhenius-based—as long as it is monotone and the parameters are identifiable from the data. Alternatively, Wiener-process and gamma-process models provide fully stochastic non-linear frameworks. The key requirement is that the chosen functional form be physically motivated, not selected purely for empirical fit.

What is the difference between accelerated degradation testing and accelerated life testing?

Accelerated life testing observes actual failures at elevated stress and models time-to-failure directly; it requires enough units to fail. Accelerated degradation testing measures a performance characteristic over time without requiring failure, then extrapolates to a failure threshold. Degradation testing is preferred when failures are extremely rare, but it requires a reliable measurable surrogate for the failure mode.

Sources

  1. Meeker, W. Q., Escobar, L. A., & Lu, C. J. (1998). Accelerated degradation tests: modeling and analysis. Technometrics, 40(2), 89–99. DOI: 10.1080/00401706.1998.10485191 ↗

How to cite this page

ScholarGate. (2026, June 2). Degradation Models (Accelerated Degradation). ScholarGate. https://scholargate.app/en/reliability/degradation-models

Related methods

Maintenance OptimizationReliability AnalysisWeibull Regression

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • Maintenance OptimizationReliability↔ compare
  • Reliability AnalysisReliability↔ compare
  • Weibull RegressionSurvival↔ compare
Compare side by side →

Referenced by

Maintenance OptimizationReliability Analysis

Similar methods

Reliability AnalysisBayesian Reliability AnalysisPrognostics and Remaining Useful LifeAccelerated Failure Time ModelHighly Accelerated Life TestingRobust Reliability AnalysisAccelerated Shelf-Life TestingSoftware Reliability Model

Related reference concepts

Survival Analysis and Time-to-Event MethodsCox Regression ModelsItem Response TheoryStructural and Latent Variable ModelsKaplan-Meier Survival CurvesCopula Models

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — Degradation Models (Degradation Models (Accelerated Degradation)). Retrieved 2026-07-21 from https://scholargate.app/en/reliability/degradation-models · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
Meeker, Escobar & Lu
Year
1998
Type
Stochastic degradation path model
Subfamily
Reliability & risk
Data Requirement
Repeated measurements over time
Failure Criterion
Threshold-crossing of a measurable characteristic
Related methods
Maintenance OptimizationReliability AnalysisWeibull Regression
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account