What is Normal Distribution?

A fundamental frequency distribution that produces a symmetrical bell-shaped diagram based on the Gaussian distribution to form a normal law of errors.

Why use Normal Distribution?

The distribution is easily described with two statistics, the mean (X-bar, which is a location parameter) and the standard distribution (sigma, which is a shape parameter carrying units of the location parameter) as these are parameters of the population.

When to use Normal Distribution?

The distribution is widely used for quality issues where errors are frequently symmetrically distributed and for a few cases of reliability problems where life data is also symmetrically distributed. For symmetrical life data, the normal data makes a good Weibull plot whereas a Weibull data usually makes a poor normal plot-thus Weibull plots have almost displaced normal plots for reliability data.

Where to use Normal Distribution?

The distribution is used where the statistics simplify descriptions of the distribution so it is easy to describe and explain.

These definitions are written by H. Paul Barringer

Return to Reliability Tools