Transient analysis of superposed GSPNs

The paper considers transient analysis using randomization for superposed g eneralized stochastic Petri nets (GSPNs). Since state space explosion impli es that space is the bottleneck for numerical analysis, superposed GSPNs pr ofit from the structured representation known for its associated Markov cha in. This moves the bottleneck for analysis from space for generator matrice s to space for iteration vectors. Hence a variation of randomization is pre sented which allows to reduce space requirements for iteration vectors. An additional and welcome side effect is that during an initial phase, this al gorithm avoids useless multiplications involving states with zero probabili ty. Furthermore, it accommodates to adaptive randomization in a natural way . Although the algorithm has been developed for superposed GSPNs, it applie s to continuous time Markov chains in a more general setting.