Why Exponential & Weibull Distributions Matter: Key Concepts and Applications

1 Exponential Distribution

1.1 Basic Concept

The exponential distribution is a continuous probability distribution commonly used to model time intervals between events. Its probability density function (pdf) and cumulative distribution function (cdf) are defined as ... where λ > 0 is the rate parameter representing the average number of events per unit time.

Image showing pdf curves for different λ values.

When λ increases, the peak of the distribution becomes higher, indicating a faster event rate.

Larger λ values concentrate the distribution, while smaller λ values spread it out.

1.2 Properties and Proof

Property: Memorylessness. For all non‑negative t and s, the exponential distribution satisfies P(T>t+s \| T>t) = P(T>s). This means that if no event occurs in the first t units of time, the probability of an event occurring in the next s units is the same as it would be at the start.

Proof sketch: Using the definition of conditional probability and the cdf, we derive the above equality, confirming the memoryless property.

1.3 Example Calculation

Assume a customer arrives at a service desk on average every 10 minutes (λ = 0.1 per minute). What is the probability that the arrival time exceeds 15 minutes?

Solution: Using the exponential cdf, P(T>15) = e^{-λ·15} = e^{-1.5} ≈ 0.223.

2 Weibull Distribution

2.1 Basic Concept

The Weibull distribution is a more flexible model that can describe increasing or decreasing failure rates over time. It is a continuous probability distribution frequently used in survival analysis and reliability engineering. Its pdf and cdf are defined as ... where k > 0 is the shape parameter and λ > 0 is the scale parameter.

When k = 1, the Weibull distribution reduces to the exponential distribution.

When k < 1, the failure rate decreases over time (the “infant mortality” effect).

When k > 1, the failure rate increases over time (the “wear‑out” effect).

2.2 Properties

Generality: The Weibull distribution can model risk that increases, decreases, or remains constant over time. When k = 1 it simplifies to the exponential distribution.

Reliability function: R(t) = exp[-(t/λ)^k], the probability that an item survives up to time t.

Hazard function: h(t) = (k/λ)(t/λ)^{k-1}, describing the instantaneous failure rate at time t.

2.3 Example Calculation

Assume a mechanical component’s failure time follows a Weibull distribution with given scale λ and shape k. What is the probability that the component does not fail within 10 hours?

Solution: Using the reliability function R(10) = exp[-(10/λ)^k]. Substituting the parameters yields R(10) ≈ 0.0423.

The flexibility of the Weibull distribution makes it a valuable tool for modeling diverse risk patterns.

Exponential vs. Weibull

Although the two distributions differ in form, the exponential distribution is a special case of the Weibull distribution when the shape parameter k equals 1.

Both distributions are essential for describing random events and lifetimes, with applications across engineering, biology, economics, and other fields.