An approach to model variations of a scenario: Application to Intelligent Transport Systems Fabien Bonnefoi 2 , Lom Messan Hillah 1 , Fabrice Kordon 1 , and Guy Fr´ emont 2 1 Universit´ e P. & M. Curie, LIP6-CNRS, 4 Place Jussieu, 75005 Paris, France lom-messan.hillah@lip6.fr, fabrice.kordon@lip6.fr 2 Cofiroute DSO/R&D, 6 - 10 rue Troyon, 92310 S` evres - France fabien.bonnefoi@cofiroute.fr, guy.fremont@cofiroute.fr Summary. Modern distributed systems tend to integrate more and more features and components that increase their complexity and size. This often leads to the decomposition of such systems into multiple parts to overcome the complexity of their modeling and analysis. In this paper, we present a modeling methodology for systems engineering based on a modular approach. The methodology relies on the definition of components and assembling rules to model complex systems. It is founded on formal specification formalisms and tools to enable model checking. This paper proposes an example by which we apply this methodology on a complex system from the domain of Intelligent Transport Systems. 1 Introduction Modern distributed systems tend to integrate more and more features that increase their complexity such as mobility, a variable number of components during execution or complex physical and time constrained mechanisms (i.e. braking distance or any similar complex function). An excellent example of such systems is illustrated in Intelligent Transport Systems (ITS) where road operators, the infrastructure, vehicles, their drivers and other road users must cooperate for an efficient and secure system. Such systems are even more complex to analyze than previous distributed systems and require more reliability. These distributed systems have such specific strategies that it is useless to imagine a short-term solution ”in the large” that will fit numerous applica- tions. We prefer to take into consideration the specificities of the application domain by selecting the appropriate model and designing an accurate meth- odology. Then, it is of interest to consider that these distributed systems are centered on the notion of ”case studies” where execution scenarios are elabor- ated and analyzed. This approach is practiced in ITS projects [3]. Moreover, paradigms such as client/server, that allow the reuse of Object Oriented Ap- proach, cannot scale up to the needs of such systems.