A Restless Bandit Model for Resource Allocation, Competition, and Reservation

In “A Restless Bandit Model for Resource Allocation, Competition and Reservation,” J. Fu, B. Moran, P. G. Taylor study a resource allocation problem with varying requests resources of limited capacity shared by multiple requests. This is modeled as set heterogeneous restless multi-armed bandit problems (RMABPs) connected constraints imposed capacity. Following Whittle’s idea relaxing the Weber Weiss’s proof asymptotic optimality, authors propose an index policy establish conditions it to be asymptotically optimal in regime where both arrival rates capacities increase. particular, they provide simple sufficient condition optimality and, complete generality, method that generates candidate policies which can checked. Via numerical experiments, demonstrate effectiveness these results even pre-limit case.