Design of One Computationally Improved Deadlock Prevention Based on MFFP Technology in Flexible Manufacturing System

Today, the robots have been widely used in manufacturing industry, especially in flexible manufacturing system. Therefore, how to control them seems important issues since such competition for shared resources by concurrent job processes can lead to a system deadlock. Accordingly, many pioneers devote solving deadlock problem. In existing literature, the maximal number of forbidding FBM problem (MFFP) is proposed to obtain optimal controllers so as to systems' maximally permissive states can be held. Most important, the number of controllers is the least among all proposed methods. However, as complexity of Petri net models increases, computational burden usually raises. For reducing computational cost, we try to propose one novel concept named PIP (Pre-Idle Places) once it can be found from one system and bypassed in the process of solving MFFP. Based on experimental results, the improved MFFP seems more efficient than the conventional MFFP in obtaining such same optimal controllers.