Applied Statistics and Probability for Engineers

20 CHAPTER 2 PROBABILITY

depends on the objectives of the study. As specific questions occur later in the book, appro-

priate sample spaces are discussed.

EXAMPLE 2-2 If two connectors are selected and measured, the extension of the positive real line Ris to take

the sample space to be the positive quadrant of the plane:

If the objective of the analysis is to consider only whether or not the parts conform to the

manufacturing specifications, either part may or may not conform. We abbreviate yesand no

as yand n. If the ordered pair ynindicates that the first connector conforms and the second

does not, the sample space can be represented by the four outcomes:

If we are only interested in the number of conforming parts in the sample, we might sum-

marize the sample space as

As another example, consider an experiment in which the thickness is measured until a

connector fails to meet the specifications. The sample space can be represented as

In random experiments in which items are selected from a batch, we will indicate whether

or not a selected item is replaced before the next one is selected. For example, if the batch

consists of three items {a, b, c} and our experiment is to select two items without replace-

ment,the sample space can be represented as

This description of the sample space maintains the order of the items selected so that the out-

come aband baare separate elements in the sample space. A sample space with less detail

only describes the two items selected {{a, b}, {a, c}, {b, c}}. This sample space is the possi-

ble subsets of two items. Sometimes the ordered outcomes are needed, but in other cases the

simpler, unordered sample space is sufficient.

If items are replaced before the next one is selected, the sampling is referred to as with

replacement.Then the possible ordered outcomes are

The unordered description of the sample space is {{a, a}, {a, b}, {a, c}, {b, b}, {b, c}, {c, c}}.

Sampling without replacement is more common for industrial applications.

Sometimes it is not necessary to specify the exact item selected, but only a property of the

item. For example, suppose that there are 5 defective parts and 95 good parts in a batch. To

study the quality of the batch, two are selected without replacement. Let gdenote a good part

and ddenote a defective part. It might be sufficient to describe the sample space (ordered) in

terms of quality of each part selected as

S 5 gg, gd, dg, dd 6

Swith^5 aa, ab, ac, ba, bb, bc, ca, cb, cc^6

Swithout 5 ab, ac, ba, bc, ca, cb 6

S 5 n, yn, yyn, yyyn, yyyyn, and so forth 6

S 5 0, 1, 2 6

S^5 yy, yn, ny, nn^6

SRR

c 02 .qxd 5/10/02 1:06 PM Page 20 RK UL 6 RK UL 6:Desktop Folder:TEMP WORK:MONTGOMERY:REVISES UPLO D CH114 FIN L:Quark Files: