# US 2019/0318271 A1 · Performance manager to autonomously evaluate replacement algorithms > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2019/0318271 A1 - Document ID: US20190318271A1 - Status: Published application - Kind: A1 - Application number: 15/953,639 - Filed: 2018-04-16 - Published: 2019-10-17 - Earliest claimed: 2018-04-16 - Assignees: General Electric Company - Applicants: General Electric Company - Inventors: Bradford MILLER; Kirk Lars BRUNS; Michael KINSTREY; Charles THEURER; Vrinda RAJIV - Web page: https://patentgrep.com/patent/US20190318271A1 ## CPC - G06N 20/00: Machine learning - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract An algorithm data store may contain information about a pool of available algorithms (e.g., to improve operation of an industrial asset). A deployment platform may be implemented in an edge portion at an industrial site associated with a live environment executing a current algorithm. A lifecycle manager of the deployment platform may manage execution of the current algorithm in the live environment creating source data. A performance manager may receive an indication of a selected at least one potential replacement algorithm from the pool of available algorithms and manage execution of the at least one potential replacement algorithm in a shadow environment using the source data. The performance manager may then report performance information associated with the at least one potential replacement algorithm. When appropriate, the potential replacement algorithm may replace the current algorithm. ## Claims ### Claim 1 A system associated with a live environment executing a current algorithm, comprising: an algorithm data store containing information about a pool of available algorithms; and a deployment platform implemented in an edge portion at an industrial site associated with the live environment, coupled to the algorithm data store, including: a lifecycle manager adapted to: manage execution of the current algorithm in the live environment creating source data, and a performance manager adapted to: receive an indication of a selected at least one potential replacement algorithm from the pool of available algorithms, manage execution of the at least one potential replacement algorithm in a shadow environment using the source data, and report performance information associated with the at least one potential replacement algorithm. ### Claim 2 (depends on 1) The system of claim 1, further comprising: an algorithm analysis engine coupled to the algorithm data store and deployment platform, including: a data connection to access the information about the pool of available algorithms, and at least one analysis computer processor coupled to the data connection and adapted to: compare the information about each of the pool of available algorithms with at least one requirement associated with the current algorithm executing in the live environment, based on said comparison, select at least one of the pool of available algorithms as a potential replacement algorithm, and transmit the indication of the selected at least one potential replacement algorithm. ### Claim 3 (depends on 1) The system of claim 1, wherein the current algorithm and the selected at least one potential replacement algorithm have at least one of: (i) different inputs, and (ii) different outputs. ### Claim 4 (depends on 1) The system of claim 1, wherein the reported performance information is based on at least one of: (i) an objective rating, and (ii) a subjective rating including an effect on future performance. ### Claim 5 (depends on 1) The system of claim 1, wherein data not included as an input to the current algorithm is analyzed to determine algorithm execution context information. ### Claim 6 (depends on 1) The system of claim 1, wherein the performance manager manages execution of multiple copies of a single potential replacement algorithm in the shadow environment. ### Claim 7 (depends on 1) The system of claim 1, wherein the performance manager further manages execution of a copy of the current algorithm in the shadow environment. ### Claim 8 (depends on 1) The system of claim 1, wherein at least some of the performance, prior to being transmitted to a cloud portion, undergoes a process associated with at least one of: (i) anonymization, (ii) de-identification, (iii) summarization, and (iv) normalization. ### Claim 9 (depends on 1) The system of claim 1, wherein the performance information includes at least one of: (i) an accuracy, (ii) a precision, and (iii) a level of confidence. ### Claim 10 (depends on 1) The system of claim 1, wherein the information about the pool of available algorithms comprises metadata including at least one of: (i) an identifier, (ii) a description, (iii) an implementation type, (iv) a version, (v) input data, (vi) output data, (vii) a resource requirement, (viii) context information, (ix) historical performance data, (x) an author, (xi) compliance information, and (xii) license information. ### Claim 11 A method associated with a live environment executing a current algorithm, comprising: managing, by a lifecycle manager of a deployment platform implemented in an edge portion at an industrial site associated with the live environment, execution of the current algorithm in the live environment to create a data source; receiving, at a performance manager of the deployment platform, an indication of a selected at least one potential replacement algorithm from a pool of available algorithms in an algorithm data; managing, by the performance manager, execution of the at least one potential replacement algorithm in a shadow environment using the source data; and reporting, by the performance manager, performance information associated with the at least one potential replacement algorithm. ### Claim 12 (depends on 1) The method of claim 1, wherein the current algorithm and the selected at least one potential replacement algorithm have at least one of: (i) different inputs, and (ii) different outputs. ### Claim 13 (depends on 11) The method of claim 11, wherein the reported performance information is based on at least one of: (i) an objective rating, and (ii) a subjective rating including an effect on future performance. ### Claim 14 (depends on 11) The method of claim 11, wherein data not included as an input to the current algorithm is analyzed to determine algorithm execution context information. ### Claim 15 (depends on 11) The method of claim 11, wherein the performance manager manages execution of multiple copies of a single potential replacement algorithm in the shadow environment ### Claim 16 (depends on 11) The method of claim 11, wherein the performance manager further manages execution of a copy of the current algorithm in the shadow environment. ### Claim 17 A non-transitory, computer-readable medium storing program code, the program code executable by a computer processor of an algorithm selection platform to cause the platform to perform a method associated with a live environment executing a current algorithm comprising: managing, by a lifecycle manager of a deployment platform implemented in an edge portion at an industrial site associated with the live environment, execution of the current algorithm in the live environment to create a data source; receiving, at a performance manager of the deployment platform, an indication of a selected at least one potential replacement algorithm from a pool of available algorithms in an algorithm data; managing, by the performance manager, execution of the at least one potential replacement algorithm in a shadow environment using the source data; and reporting, by the performance manager, performance information associated with the at least one potential replacement algorithm. ### Claim 18 (depends on 17) The medium of claim 17, wherein at least some of the performance, prior to being transmitted to a cloud portion, undergoes a process associated with at least one of: (i) anonymization, (ii) de-identification, (iii) summarization, and (iv) normalization. ### Claim 19 (depends on 17) The medium of claim 17, wherein the performance information includes at least one of: (i) an accuracy, (ii) a precision, and (iii) a level of confidence. ### Claim 20 (depends on 17) The medium of claim 17, wherein the information about the pool of available algorithms comprises metadata including at least one of: (i) an identifier, (ii) a description, (iii) an implementation type, (iv) a version, (v) input data, (vi) output data, (vii) a resource requirement, (viii) context information, (ix) historical performance data, (x) an author, (xi) compliance information, and (xii) license information.