Limits for a characteristic which include a stated franciton of the individuals in a population. Notes:
(1) If limits are set at 3o about the process average for a normal (Gaussian) distribution, they will include 99.7% of the units produced when the process is in a state of statistical control. Other limits for normal distributions may be set using tables of the distribution functions. For other., non-normal distribution limits which will include a specified percentage of the units produced should be set by other methods.
(2) In many cases several machines making the same product feed into one process. The natural process limits should then include both the natural process limits for a single machine and some measure of the difference between machine averages. In these circumstances it may not be possible to estimate reliably the percentage of units produced which will fall within the limits. (3) Natural process limits will not ordinarily be the dimensional limits shown on an engineering drawing; they ae mostly used to compae the natural capability of the process to tolerance limits. (2.3.2.4).

Value of a characteristic designated in a given design specification or drawing. Notes:
(1) This may be the target value or dimension from which variations are permitted within a specified tolerance zone.
(2) In English the term "rated value" is sometimes used with this meaning and "nominal value" is used with the meaning 'a suitable approximation'. [2: 1.4.2].

An item (unit) with one or more nonconformities. [2: 1.5.7].

A departure of a quality characteristic from its intended level or state that occurs with a severity sufficient to cause an associated product not to meet a specification requirement. [2: 1.5.6].

Statements about one or more parameters, or about a distribution, which are to be tested by means of a statistical test. Note: The null hypothesis (H0) relates to the statement being tested, whereas the alternative hypothesis (H1) relates to the statement to be accepted when the null hypothesis is rejected. Examples: A)Test of the hypothesis that the expectation of a random variable X in a population is not less than a given value, 0: H0 ( > 0) and H1 ( < 0). B)Test of the hypothesis that the proportions of nonconforming items in two lots, p1 and p2, have the same value (unspecified): H0 (p1 = p2) and H1 (p1 ? p2) C)Test of the hypothesis that a random variable X is distributed normally (with unspecified parameters). Alternative hypothesis: the distribution is not normal.[1: 2.66].

A control chart for evaluating the process in terms of the total number of items (areas of opportunity) in a sample in which a nonconformity of a given classification occurs. [2: 3.3.10].