Online Program Home
My Program

Abstract Details

Activity Number: 229 - Statistical Process Monitoring of High-Volume Data Streams
Type: Topic Contributed
Date/Time: Monday, July 30, 2018 : 2:00 PM to 3:50 PM
Sponsor: Quality and Productivity Section
Abstract #328967
Title: Multi-Stage Processes Monitoring and Diagnostics Using Timeslides
Author(s): Emmanuel Yashchin*
Companies: IBM Research
Keywords: Change-points; Detection; False Alarm Rate; Segmentation; SPC
Abstract:

We consider multi-stage manufacturing processes in which downstream variables undergo changes that are caused by events in one of the upstream stages. For example, in the course of a semiconductor manufacturing process, reliability degradation observed in mid-process can be caused by early-stage events that occurred a month earlier. In the absence of a model that connects upstream and downstream variables, timeslides are often useful for both monitoring and diagnostics. A timeslide is a set of observations that is ordered in accordance with product passage time through a given stage. In this paper, we discuss methodology for detecting unfavorable changes in downstream variables and using timeslides to identify problematic stages in the manufacturing process. We illustrate the methodology using examples from semiconductor industry and supply chain management.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program