Authors:
            
                    Ahmed Khoumsi
                    
                        
                    
                     and
                
                    Hicham Chakib
                    
                        
                    
                    
                
        
        
            Affiliation:
            
                    
                        
                    
                    University of Sherbrooke, Canada
                
        
        
        
        
        
             Keyword(s):
            Discrete Event Systems, Multi-Decision Control, C&P-control, D&A-control, Inference-Based Control.
        
        
            
                Related
                    Ontology
                    Subjects/Areas/Topics:
                
                        Control and Supervision Systems
                    ; 
                        Distributed Control Systems
                    ; 
                        Informatics in Control, Automation and Robotics
                    ; 
                        Intelligent Control Systems and Optimization
                    ; 
                        Robotics and Automation
                    
            
        
        
            
                Abstract: 
                Some years ago, a decentralized architecture, qualified as inference-based, has been developed for supervisory
control of discrete event systems. One of its essential principles is to associate ambiguity levels to local decisions.
More recently, a decentralized control architecture, qualified as multi-decision, has been developed. Its
main principle is to use several decentralized control architectures in parallel. So far, multi-decision control
has been mostly studied as a solution to generalize inference-based control, by using several inference-based
architectures in parallel. In the present study, we regard multi-decision control with a different perspective.
Instead of using multi-decision control to generalize inference-based control, we will rather use it as an alternative
to inference-based control. More precisely, our objective is to avoid using inference-based architectures,
by using instead several simpler architectures running in parallel.