Intelligent Transportation System (ITS) with autonomic functions that are cyber-physical in nature is of rapidly increasing importance for traffic efficiency and safety. Current engineering approaches to such functions often rely on worst-case assumptions, originally used for safety engineering, due to the difficulty and cost involved in precisely modeling and analyzing the system boundaries and emergent behaviors in a highly dynamic configuration of system-of-systems. This can lead to the loss of many of the benefits in regard to traffic efficiency, but also to conditions where the transport system as a whole is prone to unacceptable high risks. We envisage a systematic approach to the development of autonomous functions in ITS resting on the basis of a formal modeling framework. This paper presents our vision for achieving such a goal on the basis of EAST-ADL, which is an ISO26262 compatible architecture modeling language and methodology for the development and management of automotive Electrics & Electronics systems. Especially, this paper elaborates on some key challenges and outlines related research issues to be regarded in a Swedish research initiative, referred to as SARMITS (Systematic Approach to Risk Management in ITS Context).
