Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/123768
Title: quotTransform and Intercept Models for Bivariate Cascade High Sigma Manufacturing Processesquot
Researcher: Srinivasan Lakshminarasimhan
Guide(s): S.M.Kannan
Keywords: Manufacturing Variations, High Sigma Control Charts, Transformation,
University: Birla Institute of Technology and Science
Completed Date: 
Abstract: When two manufacturing process are serially dependant or cascade in nature, newlineregression based cause selecting control charts pioneered by Zhang in 1984 are newlineadopted for process monitoring. The objective is to identify the assignable causes of newlinevariation in the downstream process on account of an assignable cause variation in the newlineupstream process. This unique characteristic of cause selecting control charts make it newlinedistinctly different from other multivariate process monitoring control charts. newlineIn literature, cause selecting control charts have been discussed for 3 sigma newlinemanufacturing processes. Shewhart type control charts are not adoptable to processes newlinewith metric above 3 sigma, due to rare occurrence of defects. The type of control newlinecharts that is adaptable for high sigma process is a time between event control charts newlineknown as cumulative count of conforming items between two nonconforming items newlinecontrol chart. newlineIn this work a high sigma cause selecting control chart has been designed for newlineapplication of cascade process, thus upgrading the utility of cause selecting control newlinecharts from 3 sigma to high sigma. For brevity a two stage high sigma manufacturing newlineprocess has been considered in this work. The defect counts in the high sigma newlineprocesses above 3 sigma follow a geometric distribution. A new power transformation newlinehas been proposed to convert geometric data into normal form. A design flow newlinedetailing the stage wise procedure to draw the control chart has been established. The newlinedesign methodology has been demonstrated using data from a high sigma pin newlinemanufacturing process. newline
Pagination: 4.40MB
URI: http://hdl.handle.net/10603/123768
Appears in Departments:Mechanical Engineering

Files in This Item:
File Description SizeFormat 
mythesisfinal241108.zipAttached File4.51 MBUnknownView/Open


Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.