Nuclear decommissioning risk management adopting a comprehensive artificial intelligence framework: An applied case in an Italian site

摘要

The increase of nuclear sites entering the decommissioning phase has drawn the attention of governments, regulatory bodies, and industries to decommissioning projects. However, despite the high number of uncertainties and risks of these projects, there is a lack of dedicated risk management frameworks beyond the guidelines established by the International Atomic Energy Agency (IAEA) and the Project Management Institute (PMI). This study aims to propose an integrated framework for managing risks in Nuclear Decommissioning Projects. For the development of the framework artificial intelligence algorithms were employed, integrating models derived from the manufacturing sector capable of assessing risks, identifying an optimal set of risk mitigation actions, and predicting the additional project delay or cost based on a quantitative risk analysis. To demonstrate its validity, the model was applied to a nuclear decommissioning project currently underway in Italy, generating a forecast of the final project delay. The results showed to be promising, as the delay prediction for the analysed phase of the project differs from the current one only by an error of 4.7. The proposed model can help project managers in managing risks and predicting delays or additional project costs during the preliminary phase. In addition, the framework is equally useful for controlling risks in the progress stages of the project by dynamically updating forecasts.