Testing Installations and Evaluation of Test Results

Abstract

Parallel with the gradual mastering by testers of “orthodox methods” of manual testing in the course of the years, went the development of continuous and automated testing installations (cf. [586]). The purpose of these installations is to relieve the examiner not only of the often tiresome guiding of the probe, but also of the constantly required, and in the case of mass tests exhausting, subjective decision making by observing fleeting screen displays. A testing installation should evaluate the test result “objectively”. Naturally, this requires that the program, according to which the installation arrives at the decision “good” (“acceptable”) and “bad” (“reject”), is first worked out on test blocks or test pieces containing flaws of known size, or on the basis of an adequate fund of experience. Test blocks, as already mentioned in Chapter 18, are used for general adjustment of a testing installation in the widest sense and for checking these adjustments. Suitable test blocks will be discussed in their relationship with the various testing problems. If no, or unsuitable, test blocks are used for adjusting a given testing installation, this may put the entire subsequent test in doubt.