Gabriel Dengler,Laura Carnevali,Carlos E. Budde,Enrico Vicario
Abstract:Non-Markovian models have great expressive power, at the cost of complex analysis of the stochastic process. The method of Stochastic State Classes (SSCs) derives closed-form analytical expressions for the joint Probability Density Functions (PDFs) of the active timers with marginal expolynomial PDF, though being hindered by the number of concurrent non-exponential timers and of discrete events between regenerations. Simulation is an alternative capable of handling the large class of PDFs samplable via inverse transform, which however suffers from rare events. We combine these approaches to analyze time-bounded transient properties of non-Markovian models. We enumerate SSCs near the root of the state-space tree and then rely on simulation to reach the target, affording transient evaluation of models for which the method of SSCs is not viable while reducing computational time and variance of the estimator of transient probabilities with respect to simulation. Promising results are observed in the estimation of rare event probabilities.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the evaluation of time - bounded transient properties of non - Markovian models. Specifically, these models have multiple concurrent timers, and their distributions are general distributions (GEN) that are non - exponential, which enables them to capture the characteristics of real - time systems, cyber - physical systems, and software systems affected by aging. Such models define quantitative safety and liveness properties in safety - critical systems in the aerospace, railway, and nuclear industries, and are used to describe the semantics of RAMS standards (reliability, availability, maintainability, and safety), including fault - tree analysis and reliability block diagrams.
However, the complexity of non - Markovian models lies in the challenges faced when evaluating properties of the underlying stochastic processes. For example, evaluating system reliability (i.e., the probability of observing an undesired event within the mission time) is a time - bounded safety property. For the evaluation of such properties, numerical algorithms such as Value Iteration (VI) need to handle GEN transitions through approximation methods (such as phase - type distributions), but this exacerbates the state - space explosion problem, thereby limiting the research ability of numerical algorithms for non - Markovian models.
The paper proposes a method that combines state - space analysis (via stochastic state classes SSCs) with simulation, aiming to solve the above problems. This method can capture rare events without causing state - space explosion, while reducing the computation time and the variance of the probability estimator. Experimental results show that this method can also effectively evaluate models for which the SSCs - only method is not feasible, especially achieving promising results in estimating the probabilities of rare events.