Contents I. INTRODUCTION
Recent advances in wireless technology have given rise to the Internet of Things (IoT), which is convenient for use in both current and future real-world applications. IoT has gained attention due to the ongoing development of electronic devices, cloud computing services, and data analytics [1]. In addition to the many advantages it has already brought about in the areas of healthcare, transportation, the environment, and smart homes, it is also significantly impacting the industry by lowering the cost of more efficiently managed, organized, and effective monitoring. Industrial IoT represents a major shift in how industrial operations are carried out by combining advanced sensor technologies, networked devices, and sophisticated data analytics in industrial processes. The use of IoT technology, particularly in industrial settings including manufacturing facilities, energy grids, and transportation systems, is referred to as "IIoT". By implementing this paradigm shift, smart factories with improved automation, connectivity, and operational efficiency are intended to be created. Businesses benefit from the IIoT by operating more profitably, which lowers operating and capital expenditure costs. The objective of the Industrial Revolution is to develop smart industries that are endowed with a variety of technologies that facilitate interaction between machines and humans (M2H) and between machines and machines (M2M), such as improved robotics, networking, and high-power computing. One of the most promising applications of the Industrial Revolution is predictive maintenance which is employed to identify anomalies in manufacturing equipment, products, and production processes and is capable of predicting the future of the abnormal state of failure [2]. In industrial operations, maintenance has always been essential to guaranteeing the dependability, security, reliability and effectiveness of machinery and equipment. Traditional production plans and programmes, including maintenance, have been computerized by engineers due to the shift towards autonomy. A predictive maintenance system is an automated fault-finding network of interconnected parts with the self-awareness to foresee abnormalities or problems and their root causes. It has been shown that using data-driven Machine Learning (ML) algorithms to identify equipment faults and maintenance requirements in advance has a significant return on investment because it lengthens the cycle time and manufacturing capacity.
