An information-based metric for a human-machine control system in dynamic operating environments

This research created a metric named comprehension to help determine the likelihood that a human using an information-centric control system would be able to make a decision based on the information available. This control system affects the output of a yet larger integrated system that is composed of numerous subordinate organizations which are in turn composed of thousands of individual machines and humans joined through an integrated information network. The metric is composed of two components, called perception and expertise, and expands the concept of a human as a serial information processor. The perception component is indicative of the amount of information that the human acquires from the external environment. The expertise component is indicative of the amount of information that the human acquires from his internal memory. Three tactical-level decision making scenarios were created for the use case. These entities can be thought of as software objects that interact as finite state machines. Within the information domain, the scenario data were represented both graphically and textually in discrete ten-percent increments via a depiction of a representative battle command system. These scenarios were used to collect data from 24 recognized expert human military tactical decision makers that were determined to be likely representatives of the population of the 80 military officers projected to be in command of US Army Units of Action in 2020. The collected expert data were then analyzed to determine the statistical validity of the model and comprehension metric through the testing of an experimental hypothesis. This hypothesis was supported based on the data collected for the three domain-specific scenarios. An example application was then developed to illustrate how this metric could be used within the tactical military domain during the training of human operators using a representation of an Army Battle Command System (ABCS). This application was then extended to illustrate how this metric could be used in the design, development and testing of the proposed future military battle command system in a virtual simulation-based environment. Afterwards, the data were analyzed to discern any useful concepts for further research and application in other domains using safety-critical information-based systems.