Industry 4.0 means highly connected production with communication across the full value chain. It requires smart systems to provide the capabilities of sensing, data processing, data interpretation and communication. Even under harsh industry environments, these smart systems must be very reliable. Unexpected failures could lead to a loss of essential information or even a sudden stop in the highly connected automated production line. In order to prevent the substantial economic damage this could create, a new concept of prognostics health management (PHM) is introduced for adding a minimum of PHM features at all levels of the industrial electronics. It starts with adding nanoscale sensors to the individual components, which can detect delamination or interconnect damages. The information are collected and processed at board, module, and machine levels in order to trigger maintenance on demand, well before the failure occurs. At top level, schemes of artificial intelligence are employed for self-learning with continued improvement of the accuracy. Thus, the new generation of reliability methods is able to guarantee functional safety and system availability at the high level needed for Industry 4.0 applications.