Type II error

concluding H0 when H1 really true.

Type I error

concluding H1 (or rejecting H0) when H0 is really true.

Type Error

The error in rejecting Ho when it is in fact true, or in saying there is no difference.

two-level deign

also known as a two-sided test, it is a hypothesis test with a two-sided alternative hypothesis.  That is, one could possibly err on either side of the center.

t-tests

t-tsets neddd to be based on normal or close to normal data. Their variancesmust be close and their quantity must be similar.

t-test

a hypothesis test of population means when small samples are involved.

t-Test

A statistical test applied to one or two mean values.

Trial

One of the facttor combinations in an experiment

trend

a gradual, systematic change with time or some other variable.

Transportation

Cost of bringing material and components to the production facility and cost of bringing product to market.

total quality management(TQM)

a management philosophy of integrated controls, including engineering, purchasing, financial administration, marketing and manufacturing, to ensure customer satisfaction and economical cost of quality.

Tolerance

Revenues which really express customer satisfaction with the company.

Thermal Cycling

The inverse power law is used to model fatigue failure of metals that are subjected to thermal cycling.

The Hidden factory

The notion that much of the endeavour of the company that is not quality minded is directed inadvertently to creating waste and performing wasteful tasks - examples of wasteful activities are the production of non-conforming products and the holding of excessive stock. The hidden factory is the extra useful, positive output that would theoretically be possible if the energy directed at creating waste were released and directed instead at making good quality items. In 1977, the quality guru Armand Feigenbaum estimated the endeavour within the hidden factory might be 15% to 40% of total company effort. The notion of the hidden factory is bound up with the metric COPQ (cost of poor quality). The COPQ may be estimated by multiplying the number of defects per period of time by the average unit cost to fix a defect (labour and materials). Such a calculation however omits such costs as loss of goodwill and loss of competitiveness, and such other matters as warranty costs and even legal damages.

test statistic

a single value which combines the evidence obtained from sample data.  The P-value in a hypothesis test is directly related to this value.

Technical Requirement

A quantitative requirement set on a dimension or system of parts or dimensions which, when satisfied, will meet a consumer's expectations.

t-distriution

a sysmmetric, bell-shaped distiribution that resembles the standardized normal(or Z) distribution, but it typically has more area in its tails than does the Z distribution.  That is, it has greater variability than the Z distribution.