The Reversed Compound Agent Theorem (RCAT) is a compositional result that uses Markovian process algebra (MPA) to derive the reversed process of a cooperation between two agents. The equilibrium state probability distribution follows directly from a reversed process, resulting in a product-form solution for the joint state probabilities. This paper generalizes RCAT to multiple (more than two) cooperating agents. This greatly reduces the complexity of applying the original RCAT in multiple agent cooperations, removing the need for multiple applications and inductive proofs. The principle advantage is the potential of mechanically applying Multiple Agents RCAT to complicated networks, such as G-networks and BCMP networks, giving automated proofs of product-forms.
Information from pubs.doc.ic.ac.uk/marcat.