Page 319 - Applied Statistics And Probability For Engineers
P. 319

c08.qxd  5/15/02  6:13 PM  Page 271 RK UL 6 RK UL 6:Desktop Folder:TEMP WORK:MONTGOMERY:REVISES UPLO D CH114 FIN L:Quark Files:






                                                        8-7 TOLERABLE INTERVALS FOR A NORMAL DISTRIBUTION  271


                                   One-sided tolerance bounds can also be computed. The tolerance factors for these bounds are
                                   also given in Appendix Table XI.


                 EXAMPLE 8-9       Let’s reconsider the tensile adhesion tests originally described in Example 8-4. The load
                                                                                         x
                                   at failure for n   22 specimens was observed, and we found that    31.71 and s   3.55.
                                   We want to find a tolerance interval for the load at failure that includes 90% of the
                                   values in the population with 95% confidence. From Appendix Table XI the tolerance
                                   factor k for n   22,    0.90, and 95% confidence is k   2.264. The desired tolerance
                                   interval is

                                             1x   ks, x   ks2  or  331.71   12.26423.55, 31.71   12.26423.554

                                   which reduces to (23.67, 39.75). We can be 95% confident that at least 90% of the values of
                                   load at failure for this particular alloy lie between 23.67 and 39.75 megapascals.

                                   From Appendix Table XI, we note that as n S 
 , the value of k goes to the z-value associated
                                   with the desired level of containment for the normal distribution. For example, if we want
                                   90% of the population to fall in the two-sided tolerance interval, k approaches z 0.05   1.645 as
                                   n S 
 . Note that as n S 
 , a 100(1   )% prediction interval on a future value approaches a
                                   tolerance interval that contains 100(1   )% of the distribution.



                 EXERCISES FOR SECTION 8-7
                 8-60.  Compute a 95% tolerance interval on the life of the  8-64.  Compute a 90% tolerance interval on the compres-
                 tires described in Exercise 8-22, that has confidence level  sive strength of the concrete described in Exercise 8-26 that
                 95%. Compare the length of the tolerance interval with the  has 90% confidence.
                 length of the 95% CI on the population mean. Which interval  8-65.  Compute a 95% tolerance interval on the diameter of
                 is shorter? Discuss the difference in interpretation of these  the rods described in Exercise 8-27 that has 90% confidence.
                 two intervals.                                  Compare the length of the prediction interval with the length
                 8-61.  Consider the Izod impact test described in Exercise  of the 95% CI on the population mean. Which interval is
                 8-23. Compute a 99% tolerance interval on the impact  shorter? Discuss the difference in interpretation of these two
                 strength of PVC pipe that has confidence level 90%.  intervals.
                 Compare the length of the tolerance interval with the length  8-66.  Consider the bottle wall thickness measurements
                 of the 99% CI on the population mean. Which interval is  described in Exercise 8-29. Compute a 90% tolerance interval
                 shorter? Discuss the difference in interpretation of these two  on bottle wall thickness that has confidence level 90%.
                 intervals.                                      8-67.  Consider the bottle wall thickness measurements
                 8-62.  Compute a 99% tolerance interval on the brightness  described in Exercise 8-29. Compute a 90% lower tolerance
                 of the television tubes in Exercise 8-24 that has confidence  bound on bottle wall thickness that has confidence level
                 level 95%. Compare the length of the prediction interval with  90%.  Why would a lower tolerance bound likely be of
                 the length of the 99% CI on the population mean. Which  interest here?
                 interval is shorter? Discuss the difference in interpretation of  8-68.  Consider the fuel rod enrichment data described in
                 these two intervals.                            Exercise 8-30. Compute a 99% tolerance interval on rod
                 8-63.  Consider the margarine test described in Exercise 8-25.  enrichment that has confidence level 95%. Compare the
                 Compute a 99% tolerance interval on the polyunsaturated  length of the prediction interval with the length of the 95%
                 fatty acid in this particular type of margarine that has confi-  CI on the population mean.
                 dence level 95%. Compare the length of the prediction in-  8-69.  Compute a 95% tolerance interval on the syrup vol-
                 terval with the length of the 99% CI on the population mean.  ume described in Exercise 8-31 that has confidence level 90%.
                 Which interval is shorter? Discuss the difference in inter-  Compare the length of the prediction interval with the length
                 pretation of these two intervals.               of the 95% CI on the population mean.
   314   315   316   317   318   319   320   321   322   323   324