首页> 外文会议>AEC/APC Symposium >Adaption of Multivariate Control for Semiconductor Processes - (PPT)
【24h】

Adaption of Multivariate Control for Semiconductor Processes - (PPT)

机译:半导体过程的多变量控制适应 - (PPT)

获取原文

摘要

One of the advantages to real time multivariate process monitoring is the opportunity to make concurrent adjustments to the process to positively affect the outcome. Umetrics online batch analysis software utilizes this by predicting the outcome of a batch after each observation is available. This methodology stops being advantageous however when the process time is short; less than 15 minutes for example. In this case, there is no time for manual intervention, and the real time process monitoring objective changes to stopping the next process from running and notifying the appropriate systems and people. Of course, the large majority of semiconductor processes fall into this category. To support this need, a separate software application called Fabstat is available. Fabstat accepts all of the process data for the batch, a wafer in a single-wafer chamber, at one time and calculates the results immediately for action to be taken. The architecture is very flexible allowing the host system to make any type of decision based on the outcome including inhibiting a chamber, notifying process techs, and plotting the results.
机译:实时多变量过程监测的一个优点是有机会对过程进行正面影响结果的同时调整。 UMetrics在线批量分析软件通过预测每次观察后的批次结果来利用这一点。然而,当处理时间短时,该方法停止有利;例如,不到15分钟。在这种情况下,没有时间进行手动干预,并且实时流程监控客观的变化,以阻止下一个进程运行并通知适当的系统和人员。当然,大多数半导体过程属于此类别。为了支持这种需求,可以使用一个单独的软件应用程序,称为fabstat。 Fabstat接受批次的所有过程数据,单晶片腔室中的晶片,一次性地计算用于采取的动作的结果。该架构非常灵活,允许主机系统基于包括抑制腔室的结果,通知过程技术以及绘制结果的任何类型的决定。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号