Course contentsShow
← Data Science
Lesson 187 of 2,1455. Continuous Probability DistributionsFree lesson

The Weibull Distribution: Shape, Scale, and Survival

Learn the Weibull PDF, parameters, and its flexibility in modeling failure rates from infant mortality to wear-out.

The Weibull Distribution: Shape, Scale, and Survival

What you'll learn: How the Weibull distribution uses shape and scale parameters to flexibly model failure rates from early "infant mortality" to late-stage wear-out.

What Is the Weibull Distribution?

The Weibull distribution is a continuous probability distribution commonly used to model time-to-failure or survival times. Unlike the exponential distribution (which has a constant failure rate), the Weibull can model failure rates that change over time—making it incredibly flexible for reliability engineering, quality control, and survival analysis.

Parameters: Shape (k) and Scale (λ)

The Weibull has two key parameters:

  • Shape parameter (k): Controls how the failure rate changes over time
  • Scale parameter (λ): Stretches or compresses the distribution along the time axis

The Weibull PDF

The probability density function is:

f(x; k, λ) = (k/λ) * (x/λ)^(k-1) * exp(-(x/λ)^k)  for x ≥ 0

Flexibility Through the Shape Parameter

What makes Weibull special is how k changes the failure pattern:

  • k < 1: Decreasing failure rate (infant mortality—defects fail early)
  • k = 1: Constant failure rate (becomes the exponential distribution—random failures)
  • k > 1: Increasing failure rate (wear-out phase—things break down over time)

Think of a light bulb: cheap bulbs might fail quickly (k < 1), quality bulbs fail randomly over their lifespan (k = 1), but mechanical parts wear out predictably as they age (k > 1).

Real-World Applications

Weibull is used for modeling:

  • Product lifetimes and warranty analysis
  • Mechanical component failures
  • Wind speed distributions (meteorology)
  • Survival times in medical studies

Key Takeaway: The Weibull distribution's shape and scale parameters let you model failure rates that decrease, stay constant, or increase over time—making it the go-to choice for reliability and survival analysis.