Distributed, Stochastic Resource Allocation In Teams

Abstract
This paper studies distributed resource allocation of two person teams in dynamic and stochastic environments. The problem presented herein consists of a team of two decisionmakers (DMs), who, share, finite-amounts of renewable resources, have different but overlapping responsibilities, and centralized information. The team's goal, is to cooperatively process stochastically arriving tasks within an opportunity window. The approach taken is to develop a normative/descriptive model by including well-known cognitive limitations and biases into a normative model. The model is then solved as a constrained optimization problem. Driven by the model, the DREAM paradigm (Distributed REsource Allocation and Management) is developed at the Cyberlab of the University of Connecticut. A sequence of experiments are performed under different investigation conditions. Model results and experimental data are compared to validate the predictive power of the model.

This publication has 3 references indexed in Scilit: