A tolerance interval says where most of the units will fall, for example a 95/99 interval that is 95 percent confident to contain 99 percent of the population. It is the mean plus or minus k times s, where k depends on sample size, coverage and confidence: for 30 samples at 95/95, k is about 2.55, against the 1.96 that would apply if the mean and sigma were known exactly. It is the right tool for setting specs from data and for validation claims about individual parts.
Don't set the spec from a confidence interval on the mean. Use a 95/99 tolerance interval on the individual parts.
Also heard as
- statistical tolerance interval
Related terms
Confidence interval
Range calculated from sample data that brackets an unknown parameter, such as a mean, at a stated confidence level.
Normal distribution
Symmetric bell-shaped distribution set by a mean and a standard deviation, the default model for measurement data.
Specification limits
Acceptance boundaries for a product characteristic set by design or customer requirements, labeled USL and LSL.
Standard deviation
Typical distance of individual values from their mean, the basic measure of spread, written as sigma or s.