Why is the Markov method not used as a standard technique in PSA?

This paper reviews the current applications in PSA, and discusses the perspectives in the various subdomains of PSA, like Markov vs. fault trees in system analysis and Markov vs. event trees in event sequence quantification, in both of which it is deemed that combined use of Markov and fault trees or event trees, respectively, would produce more effective (and also more efficient) modelling; or the capability of the method to deal with special topics, f.i., human reliability analysis and modelling of certain time-dependent phenomena. Finally, the paper attempts at defining under what specifications can the Markov method be made a standard guideline in PSA. (orig./DG)