Overview

The exponential distribution models the waiting time until an event that happens at a steady average rate, such as the next customer or the next failure.

Enter the time x and the rate λ, the average number of events per unit of time.

Worked examples

InputsResult
x3.5Rate Lambda (λ)0.45Exponential PDF: 0.09CDF P(X <= x): 0.79Survival P(X > x): 0.21Mean: 2.22
Show 2 more examples
x1Rate Lambda (λ)1Exponential PDF: 0.37CDF P(X <= x): 0.63Survival P(X > x): 0.37Mean: 1
x10Rate Lambda (λ)0.1Exponential PDF: 0.04CDF P(X <= x): 0.63Survival P(X > x): 0.37Mean: 10

Exponential PDF by rate lambda

With the calculator's starting values, the exponential PDF is 0.09. The table shows how it changes with the rate lambda.

Rate Lambda (λ)Exponential PDF
0.30.1
0.40.1
0.45starting value0.09
0.50.09
0.60.07

Other inputs: x: 3.5. Change them in the calculator above for your own numbers.

How it works

The CDF, 1 − e^(−λx), is the chance the event happens by time x. The survival value is the chance it hasn't happened yet. The mean waiting time is 1 ÷ λ. The distribution is memoryless: having waited a while doesn't change the chance of waiting longer.

The formula

P(X ≤ x) = 1 − e^(−λx)
λAverage events per unit of time
xWaiting time

Practical tips

  • With λ = 0.45 an hour, the chance of an event within 3.5 hours is about 0.79, and the mean wait is about 2.22 hours.
  • If calls average 2 per hour, λ = 2 and the mean gap is 30 minutes.
  • The median wait is ln(2) ÷ λ, which is shorter than the mean.

Frequently asked questions

How do I get λ from an average time?

λ = 1 ÷ average time. A 10-minute average gives λ = 0.1 per minute.

What does memoryless mean?

The chance of waiting another 5 minutes is the same however long you've already waited.

How is it related to the Poisson distribution?

If event counts follow a Poisson distribution, the gaps between events are exponential.