The Randomization Technique as a Modeling Tool and Solution Procedure for Transient Markov Processes

摘要: We present a randomization procedure for computing transient solutions to discrete state space, continuous time Markov processes. This computes probabilities. It is based on construction relating process chain. Modifications and extensions of the method allow computation distributions first passage times sojourn in processes, also expected cumulative occupancy number events occurring during interval. Several implementations are discussed. In particular we an implementation general class processes that can be described terms space S, event set E, rate vectors R, target T-abbreviated as SERT. approach handle systems whose spaces quite large, if they have sparse generators.