US 11,842,910 B2
Detecting outliers at a manufacturing system using machine learning
Bharath Ram Sundar, Chennai (IN); Raman K Nurani, Chennai (IN); Ramkishore Sankarasubramanian, Kanchipuram (IN); Ramachandran Subramanian, Chennai (IN); Bharath Muralidharan, Chennai (IN); Ramaswamy Melatoor Narayanan, Chennai (IN); and Ganapathi Raman Sankaranarayanan, Chennai (IN)
Assigned to APPLIED MATERIALS, INC., Santa Clara, CA (US)
Filed by APPLIED MATERIALS, INC., Santa Clara, CA (US)
Filed on Feb. 4, 2021, as Appl. No. 17/168,041.
Prior Publication US 2022/0246457 A1, Aug. 4, 2022
Int. Cl. H01L 21/66 (2006.01); G06N 5/04 (2023.01); H01L 21/67 (2006.01); G06N 20/00 (2019.01)
CPC H01L 21/67288 (2013.01) [G06N 5/04 (2013.01); G06N 20/00 (2019.01); H01L 22/12 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
providing, as input to a trained machine learning model, data collected by a plurality of sensors at a manufacturing system during a performance of a current process for a first set of substrates at the manufacturing system;
obtaining one or more outputs from the trained machine learning model;
extracting, from the one or more outputs:
a first amount of drift of a first set of parameter values for the first set of substrates from a target set of parameter values for the first set of substrates, and
a second amount of drift of each of the first set of parameter values for the first set of substrates from a corresponding parameter value of a second set of parameter values for a second set of substrates processed according to the current process at the manufacturing system prior to the performance of the current process for the first set of substrates;
assigning a substrate health rating for each of the first set of substrates based on the first amount of drift and a sensor health rating for each of the plurality of sensors at the manufacturing system based on the second amount of drift; and
at least one of: (a) transmitting an indication of the substrate health rating and the sensor health rating for each of the plurality of sensors to a client device connected to the manufacturing system, or (b) displaying the substrate health rating and the sensor health rating.