Hybrid production systems

: logical modeling and simulation : a case study belonging to the fiber-glass industry / Tommaso Rossi. - Milano : Guerini scientifica, ©2004. - 169 p. : ill. ; 23 cm.

Aim of the paper is to formalize and to test an approach for the logical modeling of hybrid production contexts, which belong to the class of nophysical hybrid systems. In particular, considering a hybrid production context composed of continuous process and manufacturing sub-systems interacting with each other, focus of the paper is on the ones where the relations among the continuous and the discrete parts are given by the occurrence of minor stoppages on the latter due to the values assumed by the characteristic variables of the former. As a matter of fact, such kind of production systems are quite common (plants for the paper production, soaking pit furnaces, fibre-glass reels factories, etc. are examples of them) and the lack of an opportune logical modelling methodology does not allow, on one hand, sound simulation studies to be conducted on these systems and, on the other hand, the minor stoppages problem to be appropriately faced. In short, the proposed approach consists of disregarding the dynamics of the continuous process sub-systems and representing their effects on the manufacturing parts (i.e. minor stoppages) by means of the Autoregressive Conditional Duration (ACD) models developed by Engle and Russel in 1998 to treat high-frequency and irregularly spaced financial transaction data.


Introduction 9
Chapter 1 - Hybrid systems 15
     1.1 Hybrid system definitions, p. 15 - 1.2 Hybrid architectures, p. 18  
Chapter 2 - Hybrid systems modeling approaches 21
     2.1 Automata and transition systems, p. 23 - 2.2 Dynamical systems, p. 34 - 2.3 - Algebraic structures, p. 38 - 2.4 Programming languages, p. 39 - 2.5 Hybrid Petri nets, p. 41 - 2.6 Discrete abstractions, p. 45 - 2.7 Other techniques/theories, p. 53 - 2.8 Evaluation of the presented approaches, p. 58  
Chapter 3 - Autoregressive Conditional Duration models 63
     3.1 An introduction to the Autoregressive Conditional Duration models, p. 63 3.2 General structure of the ACD specification, p. 65 - 3.3 Linear ACD models, p. 66 - 3.4 Non-linear ACD models, p. 70  
Chapter 4 - The case study context 75
     4.1 The studied production system, p. 75 - 4.2 The studied production context as a hybrid system, p. 79  
Chapter 5 - The need of a new modeling approach 83
     5.1 Hybrid attributed Petri nets, p. 83 - 5.2 MR2002 modeling approach, p. 94 5.3 The need of a new modeling approach, p. 105  
Chapter 6 - The new approach for the logical modeling of hybrid production systems 109
     6.1 The proposed modeling method, p. 109 - 6.2 The application of the new approach, p. 112 - 6.3 Simulation model of the "furnace and spooling-bushing department" system, p. 118  
Chapter 7 - Results of the statistical analysis and concluding remarks 135
     7.1 Experimental campaign and statistical analysis of the outputs, p. 135 7.2 - Conclusions and further research, p. 139  
Appendix a 145
     Dynamical systems and automata, p. 145 - Hybrid Petri nets, p. 147  
Appendix b 155
References 165