A tolerance interval is an interval that contains $p%$ of a population with $(1-\alpha)$ confidence. A TI is distinct from (but sometimes confused with) prediction intervals & confidence intervals.
A Tolerance Interval (TI) is an interval that contains $p\%$ of a population with $(1-\alpha)$ confidence. A tolerance interval can be contrasted with a prediction interval that has a $p\%$ probability of containing a future value (realization) drawn from the population, and a confidence interval that contains the true value of a parameter (e.g., the mean) with $(1-\alpha)$ confidence. Note that a tolerance interval is specified by two values (the desired proportion, and the desired confidence), whereas prediction intervals and confidence intervals are specified by only one.
