| Summary: | Thesis (Ph.D.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 1998. === Includes bibliographical references (p. 229-234). === The introduction of aeronautical datalinks as an alternate communication modality will enable direct connectivity between airborne and ground-based computers. The resulting ability of automation to accomplish communication tasks without human intervention will make feasible a variety of decision-support systems that require the communication of time-varying information. Historically, the exchange of information via voice communication has been managed through heuristically developed manual procedures. This thesis proposes a generalized approach to designing processes, both automatic systems and manual procedures, for efficiently managing the communication of time-varying information, when communication resources are constrained. To answer questions about how frequently or under what conditions aging information should be updated, a novel model of time-dependent information value, which combines elements of classic information value theory with estimation techniques, is developed for a class of "proceduralized" decision problems. The value of information is measured by its effect on the ability of the decision maker to estimate relevant state variables in the context of the proceduralized decision problem. The periodic rate at which information must be updated is shown to depend on the rate at which the expected error in the decision maker's state estimate becomes significant in the context of the proceduralized decision problem, as well as the acceptable latency in the decision maker detecting unexpected events. Derivative or intent information is shown to effect the required update rate by changing the rate at which the expected error grows. A cooperative information management environment is shown to reduce the required update rate by providing efficient detection of unexpected events. Application of the model is demonstrated through two case studies. The first examines the measurement and dissemination of airport surface observations, to support an early decision whether or not to divert to an alternate airport. The second examines the rate at which aircraft state information should be transmitted to maintain separation between proximate aircraft in an Automatic Dependent Surveillance (ADS) environment. Results provide insight into the time-dependence of information and its implications on information management processes. === by Stephen Carroll Atkins. === Ph.D.
|
|---|
