On optimal distributed decision architectures in a hypothesis testing environment

The authors consider the distributed detection problem, in which decision-makers (DMs) receive observations of the environment and transmit finite-valued messages to other DMs according to prespecified communication protocols. A designated primary DM makes the final decision on one out of two alternative hypotheses. All DMs make decisions, in order to maximize a measure of organizational performance. Given the DMs and the communication resources, the problem is to find an architecture for the organization which remains optimal for a variety of operating conditions (if it exists). It is shown that even for very small organizations this problem is quite complex, because the optimal architecture depends on variables external to the team such as the prior probabilities of the hypotheses and the misclassification costs, so that global conclusions on optimal organizational structures cannot be drawn. The authors thus also consider suboptimal solutions and obtain bounds on their performance