Theory for a Control Architecture of Fuzzy Discrete Event Systems for Decision Making

Since we introduced a method for control of fuzzy discrete event systems (FDES) in 2001, we have been focusing our attention on medical applications. In this paper, we propose a new control architecture of FDES that includes a fuzzy objective generator for generating optimal control objectives online and an online optimal control scheme using both disablement and enforcement. The optimal control problem is nontrivial because its performance index is state dependent and hence not monotonic. Furthermore, the state space of a FDES is infinite in general. We show that our online approach can solve this problem efficiently. The architecture is general and can be used for decision making in many complex systems. We demonstrate the usefulness of the architecture by applying it to HIV/AIDS treatment planning, because it poses some of the most difficult treatment challenges in medicine. We build a FDES decision model from expert's knowledge, treatment guidelines, clinic trials, patient database statistics, and other information available in the medical literature. The system generates optimal control objectives for real patients from our database and applies our online approach to decide a regimen for each patient.