This project attempts to advance theoretical understanding of problem solving in a multiple agent environment, as applied to a resource allocation problem. The research is cast in a decision theoretic framework. It explores the process of decision making where: a) information is obtained about the preference profiles of each agent; b) the information acquired is then used as a basis for finding socially optimal resource allocation that maximize total net welfare, and; c) the costs involved in acquiring information are considered an integral part of the process. The objective of the research is to deal with the following specific issues: 1) the study of various communication patterns and strategies for information gathering and various system topologies for team decision making; 2) the investigation of the incorporation of computational costs, and the tradeoff between the relative cost and time of computation versus communication for effective management of resources; 3) the study of the problems concerning incentive compatibility; 4) the investigation of alternative goals to finding a socially optimal resource allocation; 5) the examination of the static resource allocation problem, versus the dynamic one, with incremental addition and deletion of resources. Results of this investigation should have significance, not only for the practical use of decision theoretic models, but also in providing insights into research fundamental to the design and use of computer and communication systems to coordinate work and decision making.
