Hierarchical team coordination in a distributed dynamic environment

Date of Completion

January 1993


Psychology, Behavioral|Psychology, Industrial|Engineering, System Science




The development of information technology brings a variety of new possibilities in structuring today's decisionmaking systems with different organizational structures, functional distributions and communication capacities. There arises the need to understand how a team of distributed decisionmakers coordinate in a dynamic environment. This thesis presents a normative-descriptive study of this issue by adopting new developments in team psychology and systems science.^ A conceptual framework is developed where the coordination and decision processes of team members are represented in two channels of information flow: situation information from the environment and communication among team members. The framework reveals mechanisms of how a team can coordinate with or without communication. Based on the framework, a distributed mathematical model is developed for a generic task coordination problem, where a team of one coordinator and three subordinates is required to process a mixture of randomly arriving individual and multi-operation tasks. The multi-operation tasks require sequential processing by multiple decisionmakers and necessitate team coordination. Team communication is explicitly captured in the model, and the mathematical solution procedures are examined in relation to the conceptual framework to give an intuitive "feel" for the model.^ A laboratory experiment is presented with which to examine how a hierarchical team of human decisionmakers coordinate their decisions via a distributed computer system. The experiment yields significant findings on the effects of coordinator's role, team communication structure, and the team's adaptation to time pressure. By applying the mathematical model to solve a subset of the experiment cases, two team coordination trends (over-communication and over-cooperation), and a team adaptation feature (planning horizon reduction) are identified in the model-data comparison. Incorporating these descriptive factors in the model, a normative-descriptive model is developed which demonstrates similar decisionmaking strategies as observed in human teams. The resulting model, used in simulation, quantitatively examines the joint effects of communication and mutual prediction capability among team members. ^