Model-Based Testing for Embedded Systems

Automated Statistical Testing for Embedded Systems 131

programming methods can be used to design testing to minimize the potential of eco-

nomic loss from field failure (Sherer 1996).

• Economic stopping criteria. Mathematically optimal rules have been developed for sup-

porting decisions to stop testing, based on the known cost of continued testing versus

the expected cost of failure in the field (Dalal and Mallows 1988). Quantitative analysis

of the usage model can assist in assessing the cost of continued testing and the risk of

failure in the field.

5.4.4 Controlling special test situations

Application of statistical science includes creating special, nonrandom test cases. Such test-

ing can remove uncertainty about how the system will perform in specific ...