9409691 Sreenivas The research deals with the study of systems whose states have logical or symbolic, rather than numerical values that change with the occurrence of events called Discrete Event Dynamic Systems (DEDS). In this proposal we outline research issues concerning modeling, analysis, control and performance evaluation of DEDS. The supervisory control problem for a broad class of infinite-state DEDS is solvable. This fact can be used to combat the complexity problem in DEDS. This issue is the focus of the modeling and analysis section. The control section addresses an archetypal problem in the control of DEDS, where we propose a novel approach to the elimination of deadlocks. Finally, in the performance section we consider an approach that will permit a system designer to modify, automatically, discrete event simulation programs that only estimate performance so as to efficiently obtain the sensitivity of the performance. This sensitivity information can be used towards system optimization using stochastic approximation techniques. ***
