Design and implementation of a reduced base model construction technique for stochastic activity networks

摘要

Performance/dependability evaluation via modeling results in fast and economic system development. Traditionally, simulations have been used to evaluate realistic systems but do not yield exact results. Analytic solution methods can yield exact results, but can suffer from the state-space explosion problem. A reduced base model construction technique for stochastic activity networks, which exploits the net structure to generate a considerably reduced state-space for systems with high degree of replication, has been developed. An implementation of this technique shows that it can be successfully used for efficiently evaluating realistic systems, previously considered unsolvable by traditional stochastic Petri net methods.