Skip to Main content Skip to Navigation

Modélisation multi-contraintes d'un système de production flexible

Abstract : Flexible Manufacturing Systems are real opportunity for industrials to face the comptetion and the unpredictability of the market. The management implementation of this kind of manufacturing systems are complex difficult and expensive.In this thesis work, we propose a decision making tool for managing Flexible Manufacturing Systems. The latter is based on a simulation model with Hierarchical Colored Petri Nets. The simulation allows the observation of the parameters variation of a system, and the determination of performance criteria. The objectives of this tool are to enable the modeling, simulation and evaluation of production systems taking into account most of the constraints encountered in industrial systems. We used this simulator to test and validate a solution to the problem of routing products during production. The choice of Petri nets (PN) to the detriment of analytical models is justified by the fact that Petri Nets with its graphical representation allows a simplified description of processes as analytical models , Petri Nets can be used for simulation, and with simulation we can predict the behaviour of a systems during a production phase. Petri Nets allow, beyond modeling and simulation to prevent the dead locks of systems. These are the reasons why we use Petri Nets to develop our models. Among the models that we have developed we can mention : machine, stocks, transportation resources and products models. We insisted on product and machine modelling which are in the center of flexibility issues. In order to study the routing problem, We defined models to represent the transport activity. We validated these models on a real case namely the production chain of CESI of Rouen. Then to solve the routing problem, we developed a decision making method based on entropy and dispacthing rules. This method allows to modify the dispatching rules weight in function of the state of the manufacturing systems during the production (size of stocks; machine failure). We compared this method with an another one where the weight of dispatching rules are unchanged during the production phase. We obtained interesting result that we have presented and discussed.
Document type :
Complete list of metadata

Cited literature [93 references]  Display  Hide  Download
Contributor : ABES STAR :  Contact
Submitted on : Tuesday, January 17, 2017 - 10:25:07 AM
Last modification on : Wednesday, March 2, 2022 - 10:10:09 AM
Long-term archiving on: : Tuesday, April 18, 2017 - 12:44:16 PM


Version validated by the jury (STAR)


  • HAL Id : tel-01437152, version 1


José Eloundou. Modélisation multi-contraintes d'un système de production flexible. Modélisation et simulation. INSA de Rouen, 2016. Français. ⟨NNT : 2016ISAM0012⟩. ⟨tel-01437152⟩



Record views


Files downloads