Infrastructure systems represent a very important aspect of life on Earth. Existing Infrastructure is subjected to degradation while the demands are growing for a better infrastructure system in response to the high standards of safety, health, population growth, and environmental protection. Bridges play a crucial role in the urban development by providing access for people to services such as health care units, schools, markets, etc. Bridges are vulnerable to high levels of deterioration because of some factors such as deferred maintenance actions, extreme weather conditions, variable traffic loading, etc. A reliable deterioration model is required for the successful development of Bridge Management Systems (BMSs) which helps in performing accurate maintenance, repair, and rehabilitation activities. This paper presents a hybrid Bayesian model that is capable of predicting the condition ratings of the concrete bridge decks along its service life. Bayesian belief networks (BNs) are utilized to model the factors that affect the condition rating of the bridge decks. BNs are used to calculate the transition probabilities based on the severity of five bridge defects which are: Corrosion, delamination, cracking, spalling and pop-out. Finally, a Markovian model is used to predict the future performance of the concrete bridge decks. A case study of the concrete bridges in Quebec is presented to demonstrate the capabilities of the proposed model.