On the stability of a dynamic stochastic capacity pricing scheme for resource allocation in a multi-agent environment

Abstract

Following the view point of Evolutionary Dynamics, we have built a multi-agent system to study resource allocation in a heterogeneous network of resources. Resources are modeled as strategies, and agents distribute processing requirements onto them using imperfect information and local decision making. Agents are endowed with bounded rationality in an imperfect information structure environment. Our intent is to achieve cooperative equilibrium using competitive dynamics by controlling congestion through capacity pricing. A distributed differentiated pricing scheme is proposed to improve loose coupling between agents and resources through a loosely coupled interaction model. However, the benefits of greater decentralization and increased local decision-making come at the expense of greater stochastic dynamics which can have unpredictable effects on the stability of the system. We had to come up with an appropriate approach for its stability analysis. This poster outlines the system built and some aspects of our stability analysis approach. © Springer-Verlag Berlin Heidelberg 2004.