Analysis on steady state behavior of DEVS models

B. P. Zeigler´s (1984) DEVS formalism supports a unified modeling and simulation framework for discrete event systems, but it lacks an analytic means for reasoning about system behavior. To provide both analytic and simulation means for the formalism, an approach to analyzing the steady state behavior of DEVS models without simulation experiments is proposed. By establishing correspondence between a DEVS model and a continuous-time Markov chain in steady state, the approach transforms the given DEVS model into an equivalent continuous time Markov chain. By analyzing the Markov chain, various steady state probabilities, such as mean sojourn time and mean waiting time, are obtained. The proposed approach is validated by comparing the results from the approach with those from simulation experiments