On Implementing Policy Improvement Algorithms in Manufacturing Flow Control Models

This paper proposes a numerical technique for the approximation of the solution of a class of piecewise deterministic control problems typically associated with manufacturing flow control models. This algorithm exploits the structure of Markov decision process with continuous state and action spaces that can be associated with piecewise deterministic control systems. The numerical method is based on an approximate policy improvement algorithm implementing a decomposition of the value functions on base functions.