We agree with the reviewers’ comments. Experimentally, the study of fires problems in full-scale remains quite difficult, and there is a little study in this context. Moreover, the small-scale experimentally studies are simpler to implement and allow to perform many necessary steps that are difficult to perform in full-scale. To simulate the problem in the laboratory, many experiments were conducted using a similarity, based on the Froude number preservation between the real situation and the model. This theory remains valid and can be applied for the numerical CFD studies on a reduced scale. In addition, the small-scales numerical studies allow to save the computation time. However, the full-scale studies need to use excessively fine meshes, withere significant constraints, in terms of time and computing power. In this study, a numerical sexaminatudyion is performed through the FDS software. This software is internationally recognized in the fire modeling. For tunnel fires, FDS has been used by several researchers in situations with large and/or small scale and these studies show the ability of this software to provide a results of sufficient accuracy for post-fire investigation. In the present work, numerical simulations on a reduced scale have been carried out on the basis of the small-scale experimental studies of Li, et ,al. while ensuring the Froude number preservation between the full-scale and the model. The FDS Results compared with those experimentally obtained by Li, et, al. show that there is an acceptable agreement. Moreover, dimensional analysis of the results is performed to find some correlations that we hope is that they provide the necessary information in the fire safety science and supprovidely the framework for the experimentally continuation of this experimental work.