The sustainability of traditional technologies employed in energy and chemical infrastructure brings a big challenge for our society. Making decisions related with safety of industrial infrastructure, the values of accidental risk are becoming relevant points for discussion. However the challenge is the reliability of the models employed to get the risk data. Such models usually involve large number of variables, high complexity and large amounts of uncertainty. The most efficient techniques to overcome those problems are built using Artificial Intelligence (AI), and more specifically using hybrid systems and deep learning. Therefore, this paper aims to introduce a well-trained algorithm for risk assessment using the Grouping Method of Data Handling, which could be capable to deal efficiently with the complexity and uncertainty. The method of Deep Learning for risk assessment involves a regression analysis called group method of data handling (GMDH), which consists in the determination of the optimal configuration of the risk assessment model and its parameters employing polynomial theory. The Findings of this study shows that risk values could be improved using deep learning algorithms in contrast with the traditional methods by increasing the precision of risk estimation. Additional to this contribution, this study highlights the sensible and critical parameters of the learning system.