Resumen: The multifaceted nature of cyber-physical systems needs holistic study methods to detect essential aspects and interrelations among physical and cyber components. Like the systems themselves, security threats feature both cyber and physical elements. Although to apply divide et impera approaches helps handling system complexity, to consider just one aspect at a time does not provide adequate risk awareness and hence does not allow to design the most appropriate countermeasures. To support this claim, in this paper we provide a joint application of two model-driven techniques for physical and cyber-security evaluation. We apply two UML profiles, namely SecAM (for cyber-security) and CIP\_VAM (for physical security), in combination. In such a way, we demonstrate the synergy between both profiles and the need for their tighter integration in the context of a reference case study from the railway domain. Idioma: Inglés DOI: 10.1016/j.compeleceng.2015.07.011 Año: 2015 Publicado en: COMPUTERS & ELECTRICAL ENGINEERING 47 (2015), 275-285 ISSN: 0045-7906 Factor impacto JCR: 1.084 (2015) Categ. JCR: COMPUTER SCIENCE, HARDWARE & ARCHITECTURE rank: 24 / 51 = 0.471 (2015) - Q2 - T2 Categ. JCR: ENGINEERING, ELECTRICAL & ELECTRONIC rank: 148 / 257 = 0.576 (2015) - Q3 - T2 Categ. JCR: COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS rank: 73 / 104 = 0.702 (2015) - Q3 - T3 Factor impacto SCIMAGO: 0.436 - Computer Science (miscellaneous) (Q1) - Electrical and Electronic Engineering (Q2) - Control and Systems Engineering (Q2)