Production modeling and simulation of surface mounting technology assembly industries using distributed agent oriented Petri net

摘要

The surface mounting technology (SMT) is widely used in the printed circuit board (PCB) assembly industries. In general, the SMT assembly is categorized into the processes of chip feeding of surface mountable component (SMC), nuzzle exchange, chip pickup and chip placement. To increase the productivity and throughput, the SMT mounting machine is capable of pickup and placement of chips in the batch manner. In the case study, four nuzzles are allocated for assembly concurrently. Each nuzzle is desired for a group of SMC components with the geometry and weight considerations. In order to increase the assembly chip types of the SMT mounting machine, nuzzles are capable of automatic nuzzle change (ANC). The number of components in a batch is variable according to the nuzzle allocations of each nuzzle exchange. Finally, the components of each pickup batch are placed on the PCB board according to the chip position and direction. Therefore, the behaviors of the SMT assembly are characterized as the concurrent operations of the discrete event systems, and they can be modeled using the high level Petri nets. In this paper, the distributed agent-oriented Petri net (DAOPN) is used to model the production activities of the SMT assembly. The models of "initializing lane buffers", "allocating the ANC", "component pickup" and "component placement" are all modeled. The assembly cycle time can also calculated in terms of the simulation results.