IP Library › Granted Patent US 12,748,393
Granted Patent B2
US 12,748,393 · App. 19/415,774 · Granted Sep 29, 2026

Controlling a manufacturing process using causal models

Inventors: Brian E. Brooks (St. Paul, MN); Gilles J. Benoit (Minneapolis, MN); Peter O. Olson (Andover, MN); Tyler W. Olson (Woodbury, MN); Himanshu Nayar (St. Paul, MN); Frederick J. Arsenault (Stillwater, MN); Nicholas A. Johnson (Burnsville, MN)
Assignee: 3M INNOVATIVE PROPERTIES COMPANY
G05B13/042B60W40/064B60W40/08B60W40/105G05B13/021G05B13/024G05B13/0265G05B13/041G05B19/4065G05B19/41835G05B23/0229G05B23/0248G06F18/2193G06N5/043G06N5/046G06N7/01G06Q10/06315G06Q10/06395G06Q30/0202G05B2219/36301G06Q10/087
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,748,393
App. No.
19/415,774
Granted
Sep 29, 2026
Kind
B2
Abstract

An example method includes selecting treatment assignments for operation of a system in an environment based on a causal model mapped to a probability distribution over possible treatment assignments, where the causal model measures causal relationships between the treatment assignments and a performance metric associated with the environment. The method further includes mapping causal effects between the treatment assignments and measures of uncertainty associated with the environment in the causal model to corresponding probabilities using probability matching. The method further includes determining the performance metric using the treatment assignments and corresponding dependent variables obtained from the environment in response to the treatment assignments. The method further includes adjusting, based on the performance metric, the causal model, and re-computing the adjusted causal model by computing overall causal effects based on the measures of uncertainty around the overall causal effects.

Claims (65)

1 . A method comprising:

clustering procedural instances based on one or more characteristics of an environment to form two or more clusters;

identifying each of the procedural instances as one of a hybrid instance or a baseline instance for a controllable element associated with the environment;

selecting first control settings for the identified baseline instances;

selecting second control settings for the identified hybrid instances;

selecting treatment assignments for operation of a system in the environment based on a causal model mapped to a probability distribution over possible treatment assignments, wherein the causal model measures causal relationships between the treatment assignments and a performance metric associated with the environment, selection of the treatment assignments being for blocked groups according to a blocking scheme, the blocked groups formed based on a first blocked group including the baseline instances and a second blocked group including the hybrid instances;

mapping causal effects between the treatment assignments and measures of uncertainty associated with the environment in the causal model to corresponding probabilities using probability matching;

determining the performance metric using the treatment assignments and corresponding dependent variables obtained from the environment in response to the treatment assignments;

adjusting, based on the performance metric, the causal model; and

re-computing the adjusted causal model by computing overall causal effects based on the measures of uncertainty around the overall causal effects.

2 . The method of claim 1 , wherein the treatment assignments are selected for blocked groups according to the blocking scheme by selecting control settings for controllable elements for entities in the environment.

3 . The method of claim 1 , further comprising:

selecting a spatial extent for each respective entity of a plurality of entities associated with the environment;

selecting a temporal extent for each respective procedural instance of the procedural instances; and

generating the procedural instances based on the selected spatial extents and the selected temporal extents.

4 . The method of claim 1 , wherein clustering the procedural instances comprises:

computing d-scores for the first and second control settings, wherein each respective d-score represents an effect of the respective treatment assignment being active;

monitoring interaction terms between the causal effects and one or more external factors associated with the environment; and

applying a factorial analysis of variance (ANOVA) to generate the two or more clusters across space and time such that each respective cluster of the two or more clusters is representative of a distinct state of the one or more external factors.

5 . The method of claim 1 , wherein clustering the procedural instances comprises applying a machine learning technique.

6 . The method of claim 5 , wherein the machine learning technique comprises one of a decision tree-based technique, a conditional inference tree-based technique, a deep neural network, or a k-means clustering technique.

7 . A system comprising:

a memory configured to store data associated with an environment; and

a processor in communication with the memory, the processor being configured to:

cluster procedural instances based on one or more characteristics of the environment to form two or more clusters;

identify each of the procedural instances as one of a hybrid instance or a baseline instance for a controllable element associated with the environment;

select first control settings for the identified baseline instances;

select second control settings for the identified hybrid instances;

select treatment assignments for operation of a system in the environment based on a causal model mapped to a probability distribution over possible treatment assignments, wherein the causal model measures causal relationships between the treatment assignments and a performance metric associated with the environment, selection of the treatment assignments being for blocked groups according to a blocking scheme, the blocked groups formed based on a first blocked group including the baseline instances and a second blocked group including the hybrid instances;

map causal effects between the treatment assignments and measures of uncertainty associated with the environment in the causal model to corresponding probabilities using probability matching;

determine the performance metric using the treatment assignments and corresponding dependent variables obtained from the environment in response to the treatment assignments;

adjust, based on the performance metric, the causal model; and

re-compute the adjusted causal model by computing overall causal effects based on the measures of uncertainty around the overall causal effects.

8 . The system of claim 7 , wherein the treatment assignments are selected for blocked groups according to the blocking scheme by selecting control settings for controllable elements for entities in the environment.

9 . The system of claim 7 , wherein the processor is further configured to:

select a spatial extent for each respective entity of a plurality of entities associated with the environment;

select a temporal extent for each respective procedural instance of the procedural instances; and

generate the procedural instances based on the selected spatial extents and the selected temporal extents.

10 . The system of claim 7 , wherein to cluster the procedural instances, the processor is configured to:

compute d-scores for the first and second control settings, wherein each respective d-score represents an effect of the respective treatment assignment being active;

monitor interaction terms between the causal effects and one or more external factors associated with the environment; and

apply a factorial analysis of variance (ANOVA) to generate the two or more clusters across space and time such that each respective cluster of the two or more clusters is representative of a distinct state of the one or more external factors.

11 . The system of claim 7 , wherein to cluster the procedural instances, the processor is configured to apply a machine learning technique.

12 . The system of claim 11 , wherein the machine learning technique comprises one of a decision tree-based technique, a conditional inference tree-based technique, a deep neural network, or a k-means clustering technique.

13 . A non-transitory computer-readable medium encoded with instructions that, when executed, cause a processor to:

cluster procedural instances based on one or more characteristics of an environment to form two or more clusters;

identify each of the procedural instances as one of a hybrid instance or a baseline instance for a controllable element associated with the environment;

select first control settings for the identified baseline instances;

select second control settings for the identified hybrid instances;

select treatment assignments for operation of a system in the environment based on a causal model mapped to a probability distribution over possible treatment assignments, wherein the causal model measures causal relationships between the treatment assignments and a performance metric associated with the environment, selection of the treatment assignments being for blocked groups according to a blocking scheme, the blocked groups formed based on a first blocked group including the baseline instances and a second blocked group including the hybrid instances;

map causal effects between the treatment assignments and measures of uncertainty associated with the environment in the causal model to corresponding probabilities using probability matching;

determine the performance metric using the treatment assignments and corresponding dependent variables obtained from the environment in response to the treatment assignments;

adjust, based on the performance metric, the causal model; and

re-compute the adjusted causal model by computing overall causal effects based on the measures of uncertainty around the overall causal effects.

14 . The non-transitory computer-readable medium of claim 13 , wherein the treatment assignments are selected for blocked groups according to the blocking scheme by selecting control settings for controllable elements for entities in the environment.

15 . The non-transitory computer-readable medium of claim 13 , the non-transitory computer-readable medium being further encoded with instructions that, when executed, cause the processor to:

select a spatial extent for each respective entity of a plurality of entities associated with the environment;

select a temporal extent for each respective procedural instance of the procedural instances; and

generate the procedural instances based on the selected spatial extents and the selected temporal extents.

16 . The non-transitory computer-readable medium of claim 13 , wherein to cluster the procedural instances, the processor is configured to:

compute d-scores for the first and second control settings, wherein each respective d-score represents an effect of the respective treatment assignment being active;

monitor interaction terms between the causal effects and one or more external factors associated with the environment; and

apply a factorial analysis of variance (ANOVA) to generate the two or more clusters across space and time such that each respective cluster of the two or more clusters is representative of a distinct state of the one or more external factors.

17 . The non-transitory computer-readable medium of claim 13 , wherein to cluster the procedural instances, the processor is configured to apply a machine learning technique.

18 . The non-transitory computer-readable medium of claim 17 , wherein the machine learning technique comprises one of a decision tree-based technique, a conditional inference tree-based technique, a deep neural network, or a k-means clustering technique.

Continuity (4)
Continuation 19283736 · Jul 29, 2025
Continuation 17437563 · Sep 11, 2019
Provisional Application 62818816 · Mar 15, 2019
Related Publication 20260104682A1 · Apr 16, 2026
References Cited (134)
US 5196997A · Kurtzberg et al. · 1993 [cited by applicant]
US 5882774A · Jonza et al. · 1999 [cited by applicant]
US 5976424A · Weber et al. · 1999 [cited by applicant]
US 6151582A · Huang et al. · 2000 [cited by applicant]
US 6266853B1 · Ho · 2001 [cited by applicant]
US 6594024B1 · Singh et al. · 2003 [cited by applicant]
US 6952688B1 · Goldman et al. · 2005 [cited by applicant]
US 7050873B1 · Discenzo et al. · 2006 [cited by applicant]
US 7164954B2 · Lefebvre et al. · 2007 [cited by applicant]
US 7401057B2 · Eder · 2008 [cited by applicant]
US 7461040B1 · Goldman et al. · 2008 [cited by applicant]
US 7680752B1 · Clune, III et al. · 2010 [cited by applicant]
US 8126574B2 · Discenzo et al. · 2012 [cited by applicant]
US 8239070B1 · Schlueter · 2012 [cited by examiner]
US 8594828B2 · Stephenson et al. · 2013 [cited by applicant]
US 9679258B2 · Mnih et al. · 2017 [cited by applicant]
US 9947018B2 · Brooks et al. · 2018 [cited by applicant]
US 10133983B1 · Chelian et al. · 2018 [cited by applicant]
US 10190792B2 · Jacobson et al. · 2019 [cited by applicant]
US 10229026B1 · Vijendra et al. · 2019 [cited by applicant]
US 10628733B2 · Schaul et al. · 2020 [cited by applicant]
US 11113605B2 · Czarnecki et al. · 2021 [cited by applicant]
US 20010047846A1 · Currens et al. · 2001 [cited by applicant]
US 20030229407A1 · Yasui et al. · 2003 [cited by applicant]
US 20040099993A1 · Jackson et al. · 2004 [cited by applicant]
US 20040230397A1 · Chadwick · 2004 [cited by applicant]
US 20050015217A1 · Weidl et al. · 2005 [cited by applicant]
US 20050055110A1 · Tuszynski · 2005 [cited by applicant]
US 20050288892A1 · Martini · 2005 [cited by applicant]
US 20060242288A1 · Masurkar · 2006 [cited by applicant]
US 20060274244A1 · Battiato et al. · 2006 [cited by applicant]
US 20070031853A1 · Stanton, Jr. et al. · 2007 [cited by applicant]
US 20070100475A1 · Korchinski · 2007 [cited by applicant]
US 20070156294A1 · Tipping et al. · 2007 [cited by applicant]
US 20070156382A1 · Graham, II et al. · 2007 [cited by applicant]
US 20070219736A1 · Okita · 2007 [cited by applicant]
US 20080013821A1 · Macgregor et al. · 2008 [cited by applicant]
US 20080027841A1 · Eder · 2008 [cited by applicant]
US 20080121729A1 · Gray · 2008 [cited by applicant]
US 20090006125A1 · Angell et al. · 2009 [cited by applicant]
US 20090204234A1 · Sustaeta et al. · 2009 [cited by applicant]
US 20090204267A1 · Sustaeta et al. · 2009 [cited by applicant]
US 20090281896A1 · Brooks · 2009 [cited by applicant]
US 20100103521A1 · Smith et al. · 2010 [cited by applicant]
US 20100191361A1 · Mccready et al. · 2010 [cited by applicant]
US 20100201242A1 · Liu et al. · 2010 [cited by applicant]
US 20100330155A1 · Berry et al. · 2010 [cited by applicant]
US 20110022193A1 · Panaitescu · 2011 [cited by applicant]
US 20110112997A1 · Sabe · 2011 [cited by applicant]
US 20120010867A1 · Eder · 2012 [cited by applicant]
US 20130038939A1 · Walker, Jr. et al. · 2013 [cited by applicant]
US 20130069792A1 · Blevins et al. · 2013 [cited by applicant]
US 20130197675A1 · Mccarthy et al. · 2013 [cited by applicant]
US 20130197698A1 · Shah et al. · 2013 [cited by applicant]
US 20140025358A1 · Hill et al. · 2014 [cited by applicant]
US 20140289174A1 · Statnikov et al. · 2014 [cited by applicant]
US 20140379101A1 · Büchmann et al. · 2014 [cited by applicant]
US 20150012255A1 · Li · 2015 [cited by examiner]
US 20150142704A1 · London · 2015 [cited by applicant]
US 20160041541A1 · Drees et al. · 2016 [cited by applicant]
US 20160147203A1 · Di Cairano et al. · 2016 [cited by applicant]
US 20160287693A1 · Norman et al. · 2016 [cited by applicant]
US 20160350796A1 · Arsenault et al. · 2016 [cited by applicant]
US 20170031354A1 · Tyber et al. · 2017 [cited by applicant]
US 20170050590A1 · List et al. · 2017 [cited by applicant]
US 20170102678A1 · Nixon et al. · 2017 [cited by applicant]
US 20170109642A1 · Kawale et al. · 2017 [cited by applicant]
US 20170278114A1 · Renders · 2017 [cited by applicant]
US 20170359418A1 · Sustaeta et al. · 2017 [cited by applicant]
US 20180013320A1 · Brooks · 2018 [cited by examiner]
US 20180025288A1 · Piche · 2018 [cited by examiner]
US 20180121817A1 · Datta et al. · 2018 [cited by applicant]
US 20180136617A1 · Xu · 2018 [cited by examiner]
US 20180181808A1 · Sridharan · 2018 [cited by applicant]
US 20180242429A1 · Ashdown et al. · 2018 [cited by applicant]
US 20180364657A1 · Luo et al. · 2018 [cited by applicant]
US 20190026754A1 · Miltonberger · 2019 [cited by applicant]
US 20190078801A1 · Turney et al. · 2019 [cited by applicant]
US 20190244110A1 · Qiu et al. · 2019 [cited by applicant]
US 20200027585A1 · Russell, II et al. · 2020 [cited by applicant]
US 20200103858A1 · Sayyarrodsari et al. · 2020 [cited by applicant]
CA 2977272A1 · 2018 [cited by applicant]
CN 102937784A · 2013 [cited by applicant]
CN 103745273A · 2014 [cited by applicant]
CN 108710289B · 2021 [cited by applicant]
EP 3316140A1 · 2018 [cited by applicant]
JP 2006277370A · 2006 [cited by applicant]
WO 2005124580A1 · 2005 [cited by applicant]
WO 2012151499A2 · 2012 [cited by applicant]
WO 2013130956A1 · 2013 [cited by applicant]
WO 2016086665A1 · 2016 [cited by applicant]
WO 2020109937A1 · 2020 [cited by applicant]
WO 2020188327A1 · 2020 [cited by applicant]
WO 2020188328A1 · 2020 [cited by applicant]
WO 2020188330A1 · 2020 [cited by applicant]
WO 2020188331A1 · 2020 [cited by applicant]
WO 2020188333A1 · 2020 [cited by applicant]
WO 2020188334A1 · 2020 [cited by applicant]
WO 2020188335A1 · 2020 [cited by applicant]
WO 2020188336A1 · 2020 [cited by applicant]
WO 2020188338A1 · 2020 [cited by applicant]
WO 2020188341A1 · 2020 [cited by applicant]
WO 2020190324A1 · 2020 [cited by applicant]
WO 2020190325A1 · 2020 [cited by applicant]
WO 2020190326A1 · 2020 [cited by applicant]
WO 2020190327A1 · 2020 [cited by applicant]
WO 2020190328A1 · 2020 [cited by applicant]
Agrawal, “Analysis of Thompson Sampling for the Multi-armed Bandit Problem”, 25th Conference on Learning Theory, 2012, vol. 23, pp. 39.1-39.26. [cited by applicant]
Alvarez, “Introduction to Adaptive Experimental Design”, Center for Quantitative Sciences, Vanberbilt University School of Medicine, Oct. 12, 2012, 96 pages. [cited by applicant]
Auer, “Finite-time Analysis of the Multiarmed Bandit Problem”, Machine Learning, 2002, vol. 47, pp. 235-256. [cited by applicant]
Cao, “Nearly Optimal Adaptive Procedure with Change Detection for Piecewise-Stationary Bandit”, 22nd International Conference on Artificial Intelligence and Statistics, 2019, vol. 89, 10 pages. [cited by applicant]
Cesa-Bianchi, “Boltzmann Exploration Done Right”, 31st Conference on Neural Information Processing Systems, 2017, pp. 1-15. [cited by applicant]
Gomes, “Machine-Learning Maestro Michael Jordon on the Delusions of Big Data and Other Huge Engineering Efforts”, IEEE Spectrum, Oct. 2014, 24 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/IB19/57662 mailed on May 1, 2020, 3 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/IB19/58424 mailed on Jan. 17, 2020, 3 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/IB2019/050701, mailed on Jan. 13, 2020, 4 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/IB2019/057646, mailed on Dec. 17, 2019, 4 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/IB2019/057648, mailed on Jan. 24, 2020, 7 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/IB2019/057664, mailed on May 4, 2020, 3 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/IB2019/057673, mailed on May 12, 2020, 2 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/IB2019/058423 mailed on Jan. 17, 2020, 3 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/IB2019/058428, mailed on Jan. 17, 2020, 3 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/IB2019/059227, mailed on Jan. 27, 2020, 3 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/IB2019/58426 mailed on, Jan. 17, 2020, 3 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/IB2019/58438 mailed on Apr. 24, 2020, 3 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/US2019/050691, mailed on Dec. 11, 2019, 3 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/US2019/050699, mailed on Dec. 5, 2019, 3 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/US2019/50695 mailed on Nov. 27, 2019, 3 pages. [cited by applicant]
International Search Report for PCT International Application No. PCT/US2019/50703 mailed on Dec. 4, 2019, 3 pages. [cited by applicant]
Jordan, “Machine Learning: Trends, Perspectives, and Prospects”, Journal of Science, Jul. 17, 2015, vol. 349, No. 6245, pp. 255-260. [cited by applicant]
Li, “Knowledge Discovery From Observational Data For Process Control Using Causal Bayesian Networks,” IIE Transactions, 2007, vol. 39, pp. 681-690. [cited by applicant]
Li, “A Contextual-Bandit Approach to Personalized News Article Recommendation”, Proceedings of the 19th International World Wide Web Conference (WWW), Apr. 2010, pp. 661-670. [cited by applicant]
Pike-Burke, “Bandits with Delayed, Aggregated Anonymous Feedback”, Proceedings of the 35th International Conference on Machine Learning, 2018, 9 pages. [cited by applicant]
Shi, “Quality Control and Improvement for Multistage Systems: A survey”, HE Transactions, Apr. 2009, vol. 41, No. 9, pp. 744-753. [cited by applicant]