Significant energy initiatives within the context of the “Belt and Road” are characterized by substantial financial transactions, intricate technological complexities, and prolonged construction timelines. Furthermore, the array of risk factors involved is diverse and continuously in flux. To identify pivotal risk indicators in investment projects, we employ the ISM‐MICMAC (interpretive structural modeling‐matriced impacts corise‐multiplication appliance classement) method. We integrate the Leaky Noisy‐OR gate model and dynamic Bayesian network (DBN) to formulate a dynamic model depicting the evolution of investment risk. This model enables us to trace the path of risk evolution at various temporal nodes through forward and backward reasoning. In a practical application of our methodology, we conduct an empirical analysis focusing on the M5 thermal power station in Belarus. The study findings reveal the establishment of a multidimensional and all‐encompassing index system. We identify primary factors contributing to project risk, notably including political instability and policy alterations. Additionally, we unveil the intricate interplay among risk factors within major energy projects and highlight specific risks at pivotal junctures in their development. Through the creation of a dynamic risk evolution model and the subsequent analysis of risk evolution paths, our work offers a foundational framework for conducting risk assessments and understanding the evolution of investments in energy projects within the Belt and Road initiative (BRI). The study’s findings have important policy implications for improving the management of risks associated with BRI major energy projects, including enhancing government macroeconomic policies, strengthening industry associations, and optimizing enterprise risk management.
