With the advent of the fourth industrial revolution in recent years, engineering systems require prognostics and health management techniques that can accurately diagnose and prognose the state of the system. In this study, we developed a state-of-the-art fault diagnostics model using real-time vibration signals in the engineering systems. We established a degradation model that follows the Weibull hazard function and performed Bayesian estimation based on Markov chain Monte Carlo simulation to develop a fault diagnostics model. From the fault diagnostics model, we developed a data-driven fault prognostics model by monitoring the condition of the system continuously. In the experiment, this model is applied in the actual Intelligent Maintenance System bearing data provided by the University of Cincinnati. Compared with the previous methods, the results showed better performance.