问题
I've been conducting black box testing for a software program that does engineering analysis on different types of concentrated solar power (CSP) models.
The quality assurance testing I've been conducting has involved only changing one default parameters of the model at a time. It's been very simple where I've just been verifying whether or not the output matches the expected result.
However, now I want to start exploring combinations of configurations that differ from the default set up to investigate if any of these combinations will output an incorrect value or even cause the user interface (UI) to fail.
The first instance I am working with, the UI has (6) different types of parameters I am expected to explore with respect to the default configuration. On top of this, I want to explore combinations of the (6) configuration options by varying them by either increasing (+) or decreasing (-) them, with respect to the default configuration.
As you can imagine, by wanting to explore these (6) parameters in combinations of (+) & (-), I'll easily create many MANY different types of black box tests.
Is there any suggestions on how to conduct these black box tests in a more efficient way? I am trying to avoid conducting all different types of configurations while still exploring the computing code in all aspects.
来源：https://stackoverflow.com/questions/57047777/are-there-efficient-methods-of-black-box-testing-that-can-avoid-redundancy