A Colaborative Fuzzy CPN System for Conflict Solution of Flexible Manufacturing System Felipe B. Reis, Vinicius F. Caridá, O. Morandin Jr., Renan L. Castro, Carlos C.M. Tuma Department of Computer Science Federal University of São Carlos (UFSCar) São Carlos, Brazil {felipe_reis, vinicius_carida, orides, renan_cafe, carlos_tuma}@dc.ufscar.br Abstract—The flexible manufacturing systems (FMS) is composed of machines, robots and automated transportation systems. Make the control of an FMS is a complex task due to the many subsystems and elements that compose it. Due to the large number of variables to be analyzed, the modeling and control of these systems become complex tasks, difficult to maintain and adapt to different layouts. The purpose of this paper is to indicate a strategy for modeling FMS based on Petri nets (PN) allowing the control to be performed by an external system. Thereby the modeling is easy to understand, requires low efforts to adapt to different production plans, and allows control of the FMS model take into consideration information current state of the system in decision-making for conflict resolution. To support control system is built multi criteria fuzzy rules. CPN Tools will be used for modeling, simulating and Matlab will be used for creating the fuzzy system. Keywords—manufacturing modeling, fuzzy system, colored Petri nets, conflict. I. INTRODUCTION In the current context of the economy, manufacturing industries are facing a crucial challenge: to increase the productive capacity of their systems to answer the volatile demands and production cycles getting shorter. Flexible Manufacturing Systems (FMS) have been designed to attend market needs that demand products that have low cost, quality, and prompt delivery. The complexity of FMS control increases with the needs of production and technological resources that cannot be designed using sequences of discrete events, simple and / or interlocks basic. The FMS must meet the needs of the production plan, alternative routes of production, operating system and the availability of resources, beyond basic control [4]. The control problem is related to the characterization of the system as being from a discrete event. The purpose of a controller for these systems is to establish and maintaining a sequence of events desired. The system must be free of deadlocks and the controller should not create new situations like this. The controller must also satisfy a set of restrictions [5]. System modeling and its representation have been considered as one of the most important challenges in many fields of engineering. A deep analysis of the manufacturing system model is an essential step before construction of a real system [1]. Colored Petri Nets (CPN's) have been used as a promising technique for modeling automated manufacturing environments. There is software that can generate graphical models of CPN's. Some offer the possibility of applying simulation on the generated models and even make mathematical analysis. Many works have been developed using CPN modeling, such as [6], [7], [19] ensuring their efficiency. Petri nets can efficiently model a manufacturing system with conflicts, buffer size, precedence relationships, resource sharing and structural interactions [2]. When two or more processes share a resource in a manufacturing system, this may result in a conflict situation. Conflicts modeled in a Petri net belong to free choice net class [3], where two or more arcs leaving the same input place are destined to different transitions. This generates different execution paths and only one transition among the enabled ones can be fired at a certain instant of time. Uncertainty about the continuation of the transition firing requires a decision. In turn, the automatic control plays a vital role in the advance of engineering and science and has become an integral and important industrial processes and manufacturing [8]. A technique of artificial intelligence widely used for decision-making systems is fuzzy logic. This paper proposes the modeling of a manufacturing system using CPN and fuzzy control. The production routing, AGVs routes, machines and buffers are modeled in CPN and a rule-based system for decision making is built using Matlab. It was developed a system which has the purpose communicate the modeling with the control, i.e., communicate the fuzzy system with Petri net and combining the efficiency of fuzzy systems for decision making and modeling in Petri nets. Using this approach it is possible to settle a dynamic scheduling, control the system and ensure that the system is caught when unexpected events occur, such as machine breakdowns or changes in production management, because only real time ＹＷＸＭＱＭＴＷＹＹＭＰＲＲＳＭＱＯＱＳＯＤＳＱＮＰＰ＠ﾩＲＰＱＳ＠ｉｅｅｅ ＳＲＱＴ