US 12,223,237 B2
Generation method, estimation method, generator, and estimator
Hiroko Yoshida, Osaka (JP); Nobuo Hara, Osaka (JP); Koichi Wakitani, Toyama (JP); and Akihisa Nakahashi, Osaka (JP)
Assigned to PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO., LTD., Osaka (JP)
Filed by Panasonic Intellectual Property Management Co., Ltd., Osaka (JP)
Filed on May 18, 2021, as Appl. No. 17/323,941.
Claims priority of application No. 2020-101907 (JP), filed on Jun. 11, 2020.
Prior Publication US 2021/0390227 A1, Dec. 16, 2021
Int. Cl. G06F 30/20 (2020.01); B23K 26/244 (2014.01); B23K 26/32 (2014.01); B23K 101/18 (2006.01); G06F 111/10 (2020.01)
CPC G06F 30/20 (2020.01) [B23K 26/244 (2015.10); B23K 26/32 (2013.01); B23K 2101/18 (2018.08); G06F 2111/10 (2020.01)] 9 Claims
OG exemplary drawing
 
1. A generation method that is executed by a processor executing instructions stored in a memory, the method comprising:
performing an experiment of a processing operation on a device in order to acquire type 1 information, type 2 information, type 3 information, and type 4 information, if the performed experiment is a simulation experiment the type 1 information and the type 2 information each indicating estimated processing conditions associated with the device during the processing operation, and the type 3 information and the type 4 information each indicating estimated results of the processing operation performed on the device, and if the performed experiment is an actual machine experiment, the type 1 information and the type 2 information each indicating actual processing conditions associated with the device during the processing operation, and the type 3 information and the type 4 information each indicating actual results of the processing operation performed on the device;
deriving a first statistical model equation expressing a first mathematical relationship between the type 1 information, the type 2 information, and the type 3 information, and deriving a second statistical model equation expressing a second mathematical relationship between the type 1 information, the type 2 information, and the type 4 information; and
generating and outputting an estimation model that estimates information indicating a target result of an inline processing operation to be performed on the device by using the first statistical model equation and the second statistical model equation and the type 2 information and the type 3 information as inputs,
wherein the information output by the estimation model is used to perform the inline processing operation on the device in a manufacturing environment that is not the experiment in order to achieve the target result.