Predictive maintenance is the act of realistically monitoring the condition and performance of assets through specific condition-based maintenance algorithms. It prevents assets breakdowns in advance and saves property damages. Predictive maintenance employs sensors, machine learning algorithms, and data analytics of predictive processes. Predictive maintenance also plays a significant role in the resource management of IIoT.
   By applying predictive maintenance, the IIoT system speculates the resource consumption level of specific operations and saves the network resources efficiently. It schedules the maintenance process through specific algorithms. Predictive maintenance improves system efficiency and productivity through efficient predictions. The unexpected system breakdowns are significantly minimized through predictive maintenance. The predictive maintenance identifies the fault occurrence in a system quickly and prevents the device from failures. Notable advantages of predictive maintenance in IIoT are efficient resource management, system lifetime expansion, minimizing sudden system breakdowns, minimizing maintenance costs, and increased revenues. The predictive maintenance technologies are acoustic monitoring, infrared technology, vibration analysis, oil analysis, and motor circuit analysis are examples of predictive maintenance technologies.
   Predictive maintenance is a recent research topic in the context of IIoT. The failure prediction considers the concepts as predictive maintenance decisions as well as design support systems. Recently developed works explore the advantages of machine learning and reasoning for predictive maintenance over IIoT. Moreover, it is essential to consider the challenges in effectively applying the machine learning techniques and ontologies in the context of predictive maintenance.