Historically I have employed computer simulation as a
pedagogical vehicle and as a managerial
analysis methodology, especially for risk management and
actuarial problem solving. More recently
this research focus has expanded to include fuzzy systems.
I am especially interested in ways in which
imprecise expert knowledge can be better represented and
processed using the formalism of fuzzy
logic. Our ongoing research in this area has produced
several research projects and publications.
''Fuzzy uncertainty in imperfect competition,''
and
''Numerical construction of fuzzy profit,''
represent an evolving direction within my fuzzy sets
research. We are also working on a theoretical
extension to the underlying mathematical model which
explicitly models uncertainty from both chance
and incomplete knowledge.
We have begun a program of research which
addresses the
manner in which humans process and
represent fuzzy information, especially from the
perspective of vocabulary. Fuzzy models require
such vocabulary as input and produce such terms as output.
Considerable research needs to be
performed to permit fuzzy models to be properly calibrated
to human users' needs.
Fuzzy controllers are being used widely in
industry to
yield better control of consumer and
commercial products. The problems being addressed are
becoming more and more complex. These
complex problems require even more understanding of
controller learning and adaptation, for
example. My current research is providing answers to these
complex problems especially to the
meaning and robustness of the linguistic terms that define
expert knowledge.
Return to Brian Schott @ GSU