Model-driven engineering is a promising approach used to develop and analyze complex systems from different domains. In this paper, we focus on the safety aspect and introduce a methodology and associated framework for modeldriven safety analysis (SA) of large critical systems. The methodology is meant to cope with design complexity and reduce time of SA process. The framework, called Sophia, supports proposed methodology and includes facilities (i) to automatically perform various SA methods, (ii) to make semantic connections with formal SA tools, (iii) to represent SA results in the system modeling environment. We illustrate our approach using a case study from transport domain.