The modern automobiles have become software intensive, with electronics features replacing many of the traditionally mechanical systems, and automating many of the drivers’ tasks. This transition brought new challenges to engineering design. The control system software exhibits unprecedented complexity, whose states cannot be exhaustively tested. Software does not fail like hardware due to random noise factors. Electronics and software update and change rapidly. Engineers have limited engineering experience and historical data to draw upon. Automating traditional manual tasks of the drivers may also lead to accidents. Safety regulation for automotive electronics is in its infancy, and standards do not yet provide adequate safety assurance.
Motivated by these challenges, this paper compares a number of hazard analysis methods for their ability to address the challenges posed by the modern automotive electronics systems. The System Theoretic Process and Analysis (STPA) framework developed for system safety engineering presents a paradigm shift, and is the most effective at identifying causes of hazards. As the first application on modern automotive electronic systems, STPA was applied to the Adaptive Cruise Control (ACC) feature. The outcome was compared with the ACC design standards and the actual vehicle implementation to illustrate the effectiveness of the method.