The 'common slope assumption' shouldn't be an assumption. It is a rule that should be checked. If you are doing what we call 'accelerated life testing' or ALT, then you are exposing your system to different stresses in order to speed up time to failure to (for example) replicate 10 years worth of use with two weeks of testing. But if the 'slope' changes, then something bad is happening. Need to learn more? Listen to this podcast!