Statistical Methods for the Routine Testing of Bottles

JSGT 1932 V16 T360-T375

The paper describes the use of statistical methods in comparing results from various tests on glass bottles. The several necessary and important statistical relationships are defined, their meaning is explained. And their relative usefulness pointed out. Illustrations are provided from data on impact, internal pressure and thermal shock tests. The results indicate that the size of the sample which should be taken for test varies within fairly large limits depending on the permissible accuracy of the variability of the product, and on the accuracy of the testing method used.

F. C. Flint & A. K. Lyle