Source

Adapted from my own handwritten and hand-typed notes from Year 1 to Year 3 of my EEE degree

The Poisson distribution models the number of random arrivals of events in a fixed period of time or region of space, when events happen independently at a steady average rate.

Typical examples:

  • telephone calls to a switchboard in an hour
  • customers entering a bank at lunch time
  • machine breakdowns in a day
  • flaws on a piece of marble

Formulas

, where is the average number of events in the period:

So is both the mean and the variance.

Scaling the rate

must match the period in the question. If power failures happen at 4.2 per year:

  • for a 4 month period,
  • for a 6 month period,

Worked example

Customers arrive at a shop at an average of 2 per hour, .

Solution:

  • no customers:
  • at most 1:
  • more than 3:

Use the complement for "more than" questions, since the Poisson values go on forever. For the count of successes in a fixed number of trials, use the Binomial Distribution instead.