Evaluating Human-Human Communication Protocols with Miscommunication Generation and Model Checking

摘要

Human-human communication is critical to safe operations in domains such as air transportation where airlines develop and train pilots on communication procedures with the goal to ensure that they check that verbal air traffic clearances are correctly heard and executed. Such communication protocols should be designed to be robust to miscommunication. However, they can fail in ways unanticipated by designers. In this work, we present a method for modeling human-human communication protocols using the Enhanced Operator Function Model with Communications (EOFMC), a task analytic modeling formalism that can be interpreted by a model checker. We describe how miscommunications can be generated from instantiated EOFMC models of human-human communication protocols. Using an air transportation example, we show how model checking can be used to evaluate if a given protocol will ensure successful communication. Avenues of future research are explored.