FAA Research Project on System Complexity Effects on Aircraft Safety: Estimating Complexity of a Safety Argument
• White Paper
Software Engineering Institute
Growth of avionics complexity threatens the ability of the FAA to certify the safety of an aircraft using a new avionics set. This project aims to determine when such safety certification will become impossible because of avionics complexity. This report presents a formula for estimating the complexity of an avionics system, in a way that selects a specific kind of complexity from the many variations, and directly connects the complexity to the size of its safety argument. We measure the complexity of the safety argument. Namely, what subclaims will have to be made and how they will be structured to argue successfully that the system is safe. Our goal is to develop a formula for complexity and its relationship to safety that distinguishes order-of-magnitude differences and can be estimated early enough in aircraft development that action can be taken to make the system less complex if there will be a problem certifying with the initial complexity. We have developed a means to estimate of the size of the safety case, which can be used to compare future avionics systems to today’s avionics systems, and then determine how difficult it will be to assure the safety of a future aircraft compared to today’s aircraft. The order of magnitude of the result can give an early indication of whether the avionics will be usable on a safety-critical aircraft.