Strategy for scalable scenarios modeling and calculation in early software reliability engineering

Abstract

System scenarios derived from requirements specification play an important role in the early software reliability engineering. A great deal of research effort has been devoted to predict reliability of a system at early design stages. The existing approaches are unable to handle scalability and calculation of scenarios reliability for large systems. This paper proposes modeling of scenarios in a scalable way by using a scenario language that describes system scenarios in a compact and concise manner which can results in a reduced number of scenarios. Furthermore, it proposes a calculation strategy to achieve better traceability of scenarios, and avoid computational complexity. The scenarios are pragmatically modeled and translated to finite state machines, where each state machine represents the behaviour of component instance within the scenario. The probability of failure of each component exhibited in the scenario is calculated separately based on the finite state machines. Finally, the reliability of the whole scenario is calculated based on the components’ behaviour models and their failure information using modified mathematical formula. In this paper, an example related to a case study of an automated railcar system is used to verify and validate the proposed strategy for scalability of system modeling.