An Approximation Algorithm for Solving Multistage, Stochastic, Dynamic Resource Allocation Problems with Single Period Travel Times

We consider a stochastic version of a dynamic resource allocation problem. In this setting, reusable resources must be assigned to demands that arise randomly over time. Both resources and demands are heterogeneous, and satisfying a demand generally will change the attributes of a resource. These problems arise in eet management, personnel training and assignment, and the remnant inventory problem. We formulate the problem as a multistage dynamic program, and suggest an approximation strategy that naturally yields integer solutions, and based on experimental evidence, provides near-optimal results.