|
As semiconductor manufacturing wafer diameter becomes larger and larger, processing technology becomes more and more complex. Conducive to increase production, reduce costs and improve quality, large diameter wafers increasingly using automatic combination device for processing. The device performs automatic combination wafer processing includes the re-entry procedure. Because the wafer contains reentrant processing, the reentrant processing may scheduling situation there are many, so that the the automatic combining means scheduling becomes very difficult. Atomic layer deposition (Atomic layer deposition, ALD) processing the wafer fabrication by automatically combining means processing typical reentrant processing is very important to obtain optimal scheduling of the re-entrant processing. This article emphasizes ALD reentrant scheduling problem of automatic processing single-arm combination device, in the scheduling of the system, the most difficult part of the process scheduling reentrant. ALD processing, the number of re-entrant may require more than five and reentrant number of the larger, more complex scheduling problems. Reentrant process can be expressed as (a, b) k wherein a and b are two reentrant processing steps of the number of processing modules, k is the number of re-entrant. Because (1,2) k is the typical processing Therefore, this study re-entry the number of k = 3,4,5 ALD re-entry process (1,2) \established Petri net model is used to describe the re-entry process single-arm combination of device characteristics. Petri net model based on established research ALD processing process (1,2) reentrant k scheduling problem. studies have shown that, for a The given number of reentrant k optimal scheduling can be obtained by the k-th potential scheduling beat calculated beat analytical expression is obtained for each candidate scheduling Then, beat the minimum scheduling is optimal scheduling in this way, reentrant processing scheduling problem becomes very simple for scheduling wafer processing reentrant automatic combination device, which is a significant improvement. Petri net model based scheduling analysis process, Based eM-Plant simulation system is established, the system makes the proposed method becomes visualization ALD reentrant processing procedure (1,2) through the application of this simulation system BEST k scheduling beat can be obtained, whereby confirm the proposed method presented in this paper to verify the proposed method is feasible, and the correctness of the instances. Finally, ALD reentrant processing management system being developed all the analytical expressions are coded into the software system, when the processing parameters have been know when we enter these parameters, it is easy to get the optimal steady-state scheduling, and the minimum takt time.
|