Reliable and efficient risk assessments are essential to deal effectively with potential risks in international construction projects. However, most conventional risk modeling methods are based on the hypothesis that risk factors are independent, which does not account adequately for the causal relationships among risk factors. In this study, a risk assessment model for international construction projects was developed to improve the efficacy of risk management by integrating fault tree analysis and fuzzy set theory with a Bayesian belief network. The risk rating of each risk factor, expressed as the product of risk occurrence probability and impact, was incorporated into the risk assessment model to evaluate degrees of risk. Therefore, risk factors were categorized into different risk levels taking into account their inherent causal relationships, which allowed the identification of critical risk factors. The applicability of the fuzzy Bayesian belief network-based risk assessment model was verified using a case study through a comparative analysis with the results from a fuzzy synthetic evaluation method. The comparison shows that the proposed risk assessment model is able to provide guidelines for an effective risk management process and ultimately to increase project performance in a complex environment such as international construction projects.