US 11,789,981 B2
Data processing device, data processing method, and non-transitory computer-readable recording medium
Atsushi Suzuki, Sapporo (JP); and Takahiko Kato, Miyagi (JP)
Assigned to TOKYO ELECTRON LIMITED, Tokyo (JP)
Appl. No. 16/957,958
Filed by Tokyo Electron Limited, Tokyo (JP)
PCT Filed Dec. 25, 2018, PCT No. PCT/JP2018/047519
§ 371(c)(1), (2) Date Jun. 25, 2020,
PCT Pub. No. WO2019/131608, PCT Pub. Date Jul. 4, 2019.
Claims priority of application No. 2017-253694 (JP), filed on Dec. 28, 2017.
Prior Publication US 2020/0372303 A1, Nov. 26, 2020
Int. Cl. G06F 18/15 (2023.01); G06F 16/28 (2019.01); G06V 30/242 (2022.01); G06F 18/21 (2023.01); G06F 18/231 (2023.01); G06F 18/24 (2023.01); G06F 119/18 (2020.01); G06F 18/23 (2023.01)
CPC G06F 16/285 (2019.01) [G06F 18/15 (2023.01); G06F 18/2163 (2023.01); G06F 18/231 (2023.01); G06F 18/24765 (2023.01); G06V 30/242 (2022.01); G05B 2219/45031 (2013.01); G06F 18/23 (2023.01); G06F 2119/18 (2020.01)] 7 Claims
OG exemplary drawing
 
1. A data processing device comprising:
an effect calculator configured to collect a plurality of data groups associated with a predetermined step of a process, and calculate effects in the predetermined step for each of the plurality of data groups;
a classifier configured to divide a feature space such that a distribution of each of the plurality of data groups associated with the predetermined step in the feature space is classified for each of the calculated effects; and
a Proxel calculator configured to calculate a Proxel that specifies respective regions of the divided feature space,
wherein the Proxel is a minimum data unit in a fine processing in a semiconductor manufacturing process,
wherein the classifier divides the feature space by performing a clustering processing on each of the plurality of data groups such that data groups having a same effect are classified into a same data group,
wherein the Proxel calculator is configured to calculate the Proxel by calculating a minimum value and a maximum value for each data included in each of the data groups classified into the same data group by the classifier,
wherein the Proxel calculator is configured to handle each of the data groups classified into the same data group by the classifier as one data group by calculating the Proxel, and
wherein the Proxel calculator is configured to interpolate an incomplete data group including a blank and handle the incomplete data group as one versatile data group including no blank by calculating the Proxel.