How Poisson Distribution Models Rare Events Across Real-World Systems

The Poisson distribution is an important probability distribution with wide applications in traffic flow, service systems, insurance, and queueing theory.

Poisson Distribution

In the early 19th century, French military statisticians recorded the rare event of new recruits being kicked by horses. Although the probability for any individual recruit was tiny, the large size of the army meant the event occurred a few times each year. Simon Denis Poisson observed that the number of such events in a given time period followed a fixed probability pattern, which he later expressed with a mathematical formula now known as the Poisson distribution.

The distribution is defined by the formula P(k) = (λ^k * e^{-λ}) / k! , where λ is the average number of events per unit time (or space), k is the actual number of events observed in the interval, and P(k) is the probability of exactly k events occurring.

Key characteristic: It describes the probability distribution of the count of events occurring in a fixed interval, assuming each small sub‑interval has the same probability of an event.

Applications of Poisson Distribution

The Poisson distribution is widely used to model real‑world scenarios such as vehicle counts passing a point in a given time, customer arrivals at a service desk, insurance claim frequencies, and queue arrivals.

For example, city planners may use traffic flow data to decide whether to install a new traffic signal. The number of vehicles passing a specific point (e.g., a camera or toll booth) during a fixed interval (e.g., per hour) can be modeled with a Poisson distribution.

In this context, λ represents the average traffic flow (average vehicles per unit time) and k is the observed vehicle count.

Derivation

The Poisson distribution can be derived as a limiting case of the binomial distribution. The binomial probability mass function is P(k) = C(n, k) * p^k * (1-p)^{n-k} , where n is the number of trials and p is the probability of success in each trial.

When the number of trials n approaches infinity while the success probability p approaches zero such that the product λ = n·p remains finite, the binomial distribution converges to the Poisson distribution.

Taking the limit, we obtain P(k) = (λ^k * e^{-λ}) / k! , which is the Poisson probability mass function.

The Poisson distribution is one of the fundamental distributions in probability theory and statistics, enabling mathematical modeling across many fields to solve practical problems and improve system efficiency.