Development of a real-time learning scheduler using adaptive critics concepts
Mehmet Murat Sahinoglu · OhioLink ETD Center (Ohio Library and Information Network) · 1993
Manufacturing systems involve human beings, machinery, non-linear dynamics, and a fusion of hierarchical and distributed organizational schemes.The cosr-efterti~~e control and scheduling of these mariufacturing sys:ems require adaptability.Therefore, it is very i~~psrtan?to implement competent on-line learning mechanisms that may accomplish balanced and adequate operation of PCCCCSS~S wwil..rl unknown dynamics in the manufacturing environment.The integration of Dynamic Prograrrzning 2nd Artif ir,~i~l Neural Networks has very important c h a ~ acter istj.c::sr,iiii.c.h may provide efficient real-time ie2rninq mechanisms for manufacturing.This thesis presents a system t h h t a c h ~e v ~s real-time learning using the mentioned integrati.?~ for manufacturing scheduling.?he system is capable of operational mappings.In addition, it utilizes reinforcercent signals of the erlvironment (a measlire of how desirable the achieved state is faking into consideration the performance criteria) due to the lack of an e x ~e r t scheduler.Conclusions are drawn and further research issues are discussed.