# US 10,482,153 B2 · Analyzing method and analyzing system for manufacturing data > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,482,153 B2 - Document ID: US10482153B2 - Status: Granted - Kind: B2 - Application number: 15/905,263 - Filed: 2018-02-26 - Published: 2019-08-29 - Granted: 2019-11-19 - Earliest claimed: 2018-02-26 - Assignees: United Microelectronics Corp. - Applicants: United Microelectronics Corp. - Inventors: Da-Ching Liao; Li-Chin Wang; Ya-Ching Cheng; Chien-Hung Chen; Chun-Liang Hou - Web page: https://patentgrep.com/patent/US10482153B2 ## CPC - G05B 23/024: {Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks} - G06F 17/15: Correlation function computation {including computation of convolution operations (arithmetic circuits for sum of products per se, e.g. multiply-accumulators G06F7/5443; digital filters, e.g. FIR, IIR, adaptive filters H03H17/00)} - G06F 7/06: Arrangements for sorting, selecting, merging, or comparing data on individual record carriers - G06F 30/00: Computer-aided design [CAD] - G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract An analyzing method and an analyzing system for manufacturing data are provided. The analyzing method includes the following steps. A plurality of models each of which has a correlation value representing a relationship between at least one of a plurality of factors and a target parameter are provided. The models are screened according to the correlation values. A rank information and a frequency information of the factors are listed up according to the models. The factors are screened according to the rank information and the frequency information. The models are ranked and at least one of the models is selected. ## Claims ### Claim 1 An analyzing method for manufacturing data, comprising: providing a plurality of models each of which has a correlation value representing a relationship between at least one of a plurality of factors and a target parameter; screening, by a model screening circuit, the models according to the correlation values, wherein some of the models whose correlation values are less than a first predetermined value are filtered out; listing up a rank information and a frequency information of the factors according to the models; screening, by a factor screening circuit, the factors according to the rank information and the frequency information, wherein some of the factors whose frequencies for the models are less than a second predetermined value are filtered out and some of the factors whose best ranks in the models are less than a third predetermined value are filtered out; and ranking the models and selecting at least one of the models, wherein the at least one of the models is selected according to the frequencies and the best ranks of the models. ### Claim 2 (depends on 1) The analyzing method for manufacturing data according to claim 1, wherein the step of screening the models according to the correlation values includes: classifying the models into a plurality of groups and filtering out some of the models in at least one of the groups, wherein some of the models are filtered out according to the correlation values of the models. ### Claim 3 (depends on 1) The analyzing method for manufacturing data according to claim 1, wherein the step of screening the models according to the correlation values includes: filtering out some of the models whose correlation value is less than a predetermined value. ### Claim 4 (depends on 1) The analyzing method for manufacturing data according to claim 1, wherein the step of screening the factors according to the rank information and the frequency information includes: filtering out some of the factors whose frequency for three-factors models and four-factors models is less than a predetermined value. ### Claim 5 (depends on 1) The analyzing method for manufacturing data according to claim 1, wherein the step of screening the factors according to the rank information and the frequency information includes: filtering out some of the factors whose frequency for one-factor models to five-factors models is less than a predetermined value. ### Claim 6 (depends on 1) The analyzing method for manufacturing data according to claim 1, wherein the step of screening the factors according to the rank information and the frequency information includes: filtering out some of the factors whose best rank in one-factor models to three-factors models is larger than a predetermined value. ### Claim 7 (depends on 1) The analyzing method for manufacturing data according to claim 1, wherein the step of screening the factors according to the rank information and the frequency information includes: reserving some of the factors whose best rank in one-factor models to three-factors models is less than a predetermined value. ### Claim 8 (depends on 1) The analyzing method for manufacturing data according to claim 1, wherein the step of screening the factors according to the rank information and the frequency information includes: marking at least one of the factors whose frequency for two-factors model to four-factors model is larger than a predetermined value. ### Claim 9 (depends on 1) The analyzing method for manufacturing data according to claim 1, wherein the step of screening the factors according to the rank information and the frequency information includes: marking at least one of the factors whose best rank for one-factor model to four-factors model is less than a predetermined value. ### Claim 10 An analyzing system for manufacturing data, comprising: a model generating unit for providing a plurality of models each of which has a correlation value representing a relationship between at least one of a plurality of factors and a target parameter; a model screening circuit for screening the models according to the correlation values, wherein some of the models whose correlation values are less than a first predetermined value are filtered out; a factor information listing unit for listing up a rank information and a frequency information of the factors according to the models; a factor screening circuit for screening the factors according to the rank information and the frequency information, wherein some of the factors whose frequencies for the models are less than a second predetermined value are filtered out and some of the factors whose best ranks in the models are less than a third predetermined value are filtered out; and a reporting unit for ranking the models. ### Claim 11 (depends on 10) The analyzing system for manufacturing data according to claim 10, wherein the model screening circuit classifies the models into a plurality of groups and filters out some of the models in at least one of the groups, wherein some of the models are filtered out according to the correlation values of the models. ### Claim 12 (depends on 10) The analyzing system for manufacturing data according to claim 10, wherein the model screening circuit filters out some of the models whose correlation value is less than a predetermined value. ### Claim 13 (depends on 10) The analyzing system for manufacturing data according to claim 10, wherein the factor screening circuit filters out some of the factors whose frequency for three-factors models and four-factors models is less than a predetermined value. ### Claim 14 (depends on 10) The analyzing system for manufacturing data according to claim 10, wherein the factor screening circuit filters out some of the factors whose frequency for one-factor models to five-factors models is less than a predetermined value. ### Claim 15 (depends on 10) The analyzing system for manufacturing data according to claim 10, wherein the factor screening circuit filters out some of the factors whose best rank in one-factor models to three-factors models is larger than a predetermined value. ### Claim 16 (depends on 10) The analyzing system for manufacturing data according to claim 10, wherein the factor screening circuit reserves some of the factors whose best rank in one-factor models to three-factors models is less than a predetermined value. ### Claim 17 (depends on 10) The analyzing system for manufacturing data according to claim 10, wherein the factor screening circuit marks at least one of the factors whose frequency for two-factors model to four-factors model is larger than a predetermined value. ### Claim 18 (depends on 10) The analyzing system for manufacturing data according to claim 10, wherein the factor screening circuit marks at least one of the factors whose best rank for one-factor model to four-factors model is less than a predetermined value.