The development and validation of an exercise countermeasure protocol management expert system based on authentic methods of reasoning

Experiments conducted during NASA's Skylab missions using treadmills and bicycles clearly demonstrated that inflight exercise provides some degree of protection against deconditioning. Exercises will be used as a primary countermeasure against deconditioning on extended Space Station flights to prevent muscle atrophy, cardiovascular deconditioning and bone demineralization resulting from prolonged exposures to a microgravity environment. An expert system has been developed to tailor exercise prescriptions to each crew member based on the crew member's past performances with a prescribed regime and expected Intravehicular/Extravehicular activity (IVA/EVA). This expert system provides initial exercise prescriptions based on a subject's preflight stress test data and his/her response to a questionnaire that extracts information regarding the subject's current fitness level and exercise modality preference. Additionally, the expert system also adjusts the exercise prescription based on complaints of the prescription being either too difficult or too easy.;This dissertation develops a methodology for validating the performance of authentic reasoning expert systems that is based on comparing the expert system's reasoning processes and use of reasoning factors to that of the experts (i.e., using several experts to solve the same problem that the authentic reasoning expert system solved). We also established a mathematical process that includes the use of rough set theory as a means of capturing and quantifying the reasoning factors and reasoning processes of the experts. Additionally, a generalized entropy criterion for measuring consensus effectiveness based on Dempster-Shafer's theory of mathematical evidence is used in conjunction with rough set theory to quantify the problem-solving performance of the experts. This is used for ascertaining whether or not the the expert system's way of solving a given set of test cases is evident in the way the experts solve the same test cases, which is an essential task in validating authentic reasoning expert systems.