Successful implementation of industry 4.0 is highly dependent on the functionality and reliability of the manufacturing equipment such as machines, robots, handling equipment, as well as equipment for measuring and quality assurance of products. It is well known that the failures of manufacturing equipment result in significant losses due to machine downtime which consequently increases manufacturing costs relating to loss of production capacity, rework and scrap. During the recent years, there has been an increase in the demand for early detection of failures and a strong need for prediction of Remaining Useful Life (RUL). Several technology enablers such as novel low-cost sensors and data acquisition systems, digitalization and cloud computing platforms have contributed to a paradigm shift from traditional Condition Monitoring to Predictive Maintenance. Awareness of current machine conditions and knowledge on the remaining time to failure are important factors for preventing machine downtime. The knowledge inferred from the monitoring of machinery is essential not only for proactive maintenance but also facilitates effective management of spare parts. This chapter presents our experience in implementing condition monitoring for predictive maintenance as one of demonstrators of the ARTC Model Factory. The chapter briefly describes topics related to sensor location optimization and sensor integration in industrial machines, data acquisition, signal processing and machine learning.