US 12,140,937 B2
Method and system for reducing work-in-process
Po-Yi Wang, Tainan (TW); and Chao-Ming Cheng, Pingtung County (TW)
Assigned to Taiwan Semiconductor Manufacturing Company, Ltd., Hsinchu (TW)
Filed by Taiwan Semiconductor Manufacturing Company, Ltd., Hsinchu (TW)
Filed on Jun. 18, 2023, as Appl. No. 18/337,021.
Application 18/337,021 is a continuation of application No. 17/852,383, filed on Jun. 29, 2022, granted, now 11,726,462.
Application 17/852,383 is a continuation of application No. 15/983,100, filed on May 18, 2018, granted, now 11,402,828, issued on Aug. 2, 2022.
Prior Publication US 2023/0359183 A1, Nov. 9, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G05B 19/418 (2006.01)
CPC G05B 19/4188 (2013.01) [G05B 19/4187 (2013.01); G05B 2219/45031 (2013.01); G05B 2219/49071 (2013.01); Y02P 90/02 (2015.11)] 20 Claims
OG exemplary drawing
 
1. A method for reducing a work-in-progress (WIP) in a process of a product, the method comprising:
collecting process profile data from a plurality of tool groups running the process, and calculating a standard deviation of an output of a stage of a bottleneck tool group of the plurality of tool groups according to the collected process profile data to serve as values of at least one of a plurality of key-performance-indicators (KPIs) of each tool group;
feeding the values of the KPIs and the WIP of each tool group into a neural network model in order to output an impact on the WIP for each KPI of each tool group by the neural network model, wherein the neural network model is trained by feeding the values of the KPIs and the WIP of each tool group in the process profile data of a plurality of days collected in advance, and performing sensitivity analysis on the values of the KPIs and the WIP so as to output the impact showing a rate of change of the WIP resulting from a predetermined change of each KPI;
selecting a predetermined percentage of the KPIs for each tool group as a set of major KPIs according to the impact on the WIP for each KPI of each tool group output by the neural network model, wherein the standard deviation of the output of the stage of the bottleneck tool group is selected into the set of major KPIs of the bottleneck tool group; and
controlling the plurality of the tool groups according to the impact of the set of major KPIs of each tool group in order to reduce a total WIP, wherein the controlling comprises adjusting a dispatching priority of the bottleneck tool group to reduce the standard deviation of the output of the stage of the bottleneck tool group and the total WIP is a summation of the WIP of each tool group.