首页> 外文会议>International symposium on semiconductor manufacturing >SIMULATING AMHS PERFORMANCE FOR SEMICONDUCTOR WAFER FABRICATION
【24h】

SIMULATING AMHS PERFORMANCE FOR SEMICONDUCTOR WAFER FABRICATION

机译:模拟半导体晶片制造的AMHS性能

获取原文
获取原文并翻译 | 示例

摘要

This paper presents the methods and results of modeling and simulating the performance of interbay automated material-handling systems (AMHS) for semiconductor wafer fabrication. Along with a case study of a wafer fab AMHS model, this paper includes an alternate approach to modeling interbay arrivals and delays and explores periods of peak transport demands.rnA From:To matrix was developed from a detailed simulation model with process flows to represent the hourly demand of bay-to-bay movements. A second From:To matrix was generated to represent the mean transport time of automated material handling. We examine the effect that these matrices have on simulation output and model performance. Three distributions, Constant, Normal, and Exponential, were used with each table. The comparison between detailed and matrix input results included the mean and 95th percentile of delivery times and lot cycle time differentials, as well as car utilization and model execution time. The Exponential distribution provided results closest to those of the detailed model.rnA challenging and often overlooked characteristic of wafer fab dynamics is the rush-hour occurrence usually experienced during meal breaks and shift changes. This paper provides an analysis of rush-hour movements in a wafer fab and their effect on interbay AMHS, while providing an approach to modeling this phenomenon. We developed a scalar array to represent the deviation from the mean hourly transport demand. This extra level of detail provided only a small improvement to the output statistics. Each of the three alternate approaches discussed provided at least a 50% reduction in model execution time.
机译:本文介绍了建模和模拟用于半导体晶圆制造的海湾间自动材料处理系统(AMHS)性能的方法和结果。除了对晶圆厂AMHS模型的案例研究外,本文还提供了另一种方法来建模海湾间到达和延误,并探讨了高峰运输需求的时期。每小时对海湾移动的需求。生成第二个From:To矩阵,以表示自动化物料搬运的平均运输时间。我们检查了这些矩阵对模拟输出和模型性能的影响。每个表使用了三个分布,即常数,正态和指数分布。详细输入结果与矩阵输入结果之间的比较包括交货时间的平均值和第95个百分位数,批次周期时间差异以及汽车利用率和模型执行时间。指数分布提供的结果与详细模型的结果最接近。晶圆厂动力学的一个具有挑战性且经常被忽视的特征是通常在进餐时间和班次变更期间出现的高峰时间。本文提供了晶圆厂高峰时间移动及其对海湾间AMHS的影响的分析,同时提供了对这种现象进行建模的方法。我们开发了一个标量数组来表示与平均每小时运输需求的偏差。这种额外的细节水平仅对输出统计数据提供了很小的改进。所讨论的三种替代方法中的每一种都至少减少了50%的模型执行时间。

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