ABSTRACT
The aim of this article is to present the stochastic simulator of the AltaRica 3.0 project. Stochastic simulation is an important tool for safety and reliability analyses of systems. It provides fine results, even with complex systems, to calculate safety and reliability indicators. The principle is to run many pseudo-random simulations, corresponding to histories of the behavior of the system, and to make statistics on them.
The stochastic simulator of the AltaRica 3.0 project was designed by taking into account original features, not only by introducing concepts within the language, but also with its implementation. First, considered delay functions are a few built-in ones representing the most widely used in safety and reliability analyses (exponential, Weibull, etc.). For more specific ones, a generic mechanism allows users to define them point wisely by means of sets of points interpolated in a triangular way. In addition, observers are defined in a generic way. They are quantities to be observed during simulations in order to perform safety and reliability analyses. Finally, compilation techniques are used to reduce computation time of simulations because stochastic simulation needs an important number of simulations to stabilize the measures.
This work is a part of AltaRica 3.0 project, which aims to propose a set of authoring, simulation and assessment tools to perform Model-Based Safety Analyses. The new version of AltaRica modeling language is in the heart of the project. Its associated assessment tools already include Fault Trees or Markov Chains compilers, stochastic and stepper simulators, etc.
