Semi-Markov control processes with non-compact action spaces and discontinuous costs
Volume 36 / 2009
Applicationes Mathematicae 36 (2009), 29-42
MSC: Primary 60K15.
DOI: 10.4064/am36-1-3
Abstract
We establish the average cost optimality equation and show the existence of an ($\varepsilon $-)optimal stationary policy for semi-Markov control processes without compactness and continuity assumptions. The only condition we impose on the model is the $V$-geometric ergodicity of the embedded Markov chain governed by a stationary policy.
