AbstractHeating, ventilation, and air conditioning components are among the significant component groups in building services. Around 50% of a buildings’ energy consumption is related to HVAC systems. Published research has indicated there is a strong relationship between the deterioration rate and energy consumption of HVAC systems. High energy consumption due to aging can significantly impact the total life cycle operating cost of HVAC systems. The deterioration process depends on various factors such as exposure condition, utilization, types of unit, maintenance regime, and others. A comprehensive literature review indicated that existing models do not consider the aforementioned parameters in predicting the degradation and, therefore, the accuracy of the current models may be low. Existing modeling practices do not consider runtime as an influencing parameter on deterioration forecasting for HVAC components. The primary knowledge contribution of this article is addressing the aforementioned limitations in existing modeling practices. Deterioration prediction models based on the Markov process are developed to identify the effect of different runtime cluster groups on the degradation of selected HVAC component groups in 2019 condition inspection data of buildings in Melbourne, Australia. In the first part of the article, a comprehensive literature review is carried out to identify the knowledge gap. Data were analyzed for 20 critical HVAC components with three main cluster groups based on runtime in the second part. Deterioration prediction models are derived based on the Markov-chain Monte Carlo (MCMC) method and the nonlinear optimization method. In the end, a detailed analysis of the results is carried out with a three-way comparison in order to demonstrate the effect of runtime on deterioration forecasting for HVAC components.