# US 11,113,652 B2 · System and method for a recommendation mechanism regarding store remodels > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,113,652 B2 - Document ID: US11113652B2 - Status: Granted - Kind: B2 - Application number: 16/569,892 - Filed: 2019-09-13 - Published: 2020-03-19 - Granted: 2021-09-07 - Earliest claimed: 2018-09-14 - Assignees: Walmart Apollo, LLC - Applicants: Walmart Apollo, LLC - Inventors: Nikesh Srivastava; Sameer Shah; Uruj Fatima - Web page: https://patentgrep.com/patent/US11113652B2 ## CPC - G06Q 10/04: Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" (market predictions or forecasting for commercial activities G06Q30/0202) - G06F 18/24155: {Bayesian classification} - G06Q 10/06375: {Prediction of business process outcome or impact based on a proposed change} - G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract An iterative, tiered system for identifying which assets need to be remodeled. This tiered system uses a Bayesian Structural Time Series, followed by an ensemble classification and cost estimation. The results are then input into an optimization model, where the best possible set of stores is selected according to the constraints. Remodeling of the store then commences. ## Claims ### Claim 1 A computer-implemented method, comprising: generating, via a processor, a trained Bayesian Structural Time Series machine learning model, wherein the trained Bayesian Structural Time Series machine learning model is iteratively trained by a first training data set comprising sales data for one or more stores; analyzing, via the processor, a profitability impact of a remodel on a target store using the trained Bayesian Structural Time Series machine learning model; generating, via the processor, a trained non-parametric regression machine learning model, wherein the trained non-parametric regression machine learning model is iteratively trained by a second training data set; performing, in parallel via the processor: generating a lift classification of the target store using the profitability impact; and generating a cost estimate for the remodel using the trained non-parametric regression machine learning model; generating, via the processor, a remodel score for the target store based on the lift classification and the cost estimate of the remodel; and determining, via the processor, that the remodel score is above a threshold, to yield a determination. ### Claim 2 (depends on 1) The method of claim 1, wherein: the first training data set comprises historical sales data of the target store; wherein the trained Bayesian Structural Time Series machine learning model implements a Kalman filter to iteratively predict future sales data of the target store based on the historical sales data of the target store; and wherein generating the trained non-parametric regression machine learning model includes assigning weights to expenditures for the target store based on historical data of previously remodeled stores. ### Claim 3 (depends on 2) The method of claim 2, wherein the determining that the remodel score is above the threshold is based on a linear optimization model, where the linear optimization model uses, as input, a total budget. ### Claim 4 (depends on 1) The method of claim 1, further comprising: applying the analyzing, determining, and the generating the lift classification and the remodel score steps to a plurality of stores to yield a plurality of remodel scores; ranking the plurality of remodel scores, to yield ranked remodel scores; and identifying, as part of the determining that the remodel score is above the threshold, that the target store is ranked highest among the ranked remodel scores. ### Claim 5 (depends on 1) The method of claim 1, wherein the cost estimate identifies costs for: HVAC (Heating, Ventilation, and Air Conditioning), Refrigeration, Fixtures, and Flooring. ### Claim 6 (depends on 1) The method of claim 1, wherein the generating of the remodel score comprises a weighted combination of a measure of profit for the target store and a measure of a need of remodeling the target store. ### Claim 7 A system comprising: a processor; and a computer-readable storage medium having instructions stored which, when executed by the processor, cause the processor to perform operations comprising: generating a trained Bayesian Structural Time Series machine learning model, wherein the trained Bayesian Structural Time Series machine learning model is iteratively trained by a first training data set comprising sales data for one or more stores; analyzing a profitability impact of a remodel on a target store using the trained Bayesian Structural Time Series machine learning model; generating a trained non-parametric regression machine learning model, wherein the trained non-parametric regression machine learning model is iteratively trained by a second training data set; performing, in parallel via the processor: determining a cost estimate for the remodel using the trained non-parametric regression machine learning model; and generating, using the profitability impact, a lift classification of the target store; generating a remodel score for the target store based on the lift classification and the cost estimate of the remodel; and determining that the remodel score is above a threshold, to yield a determination. ### Claim 8 (depends on 7) The system of claim 7, wherein the first training data set comprises historical sales data of the target store; wherein the trained Bayesian Structural Time Series machine learning model implements a Kalman filter to iteratively predict future sales data of the target store based on the historical sales data of the target store; and wherein generating the trained non-parametric regression machine learning model includes assigning weights to expenditures for the store based on historical data of previously remodeled stores. ### Claim 9 (depends on 8) The system of claim 8, wherein the determining that the remodel score is above the threshold is based on a linear optimization model, where the linear optimization model uses, as input, a total budget. ### Claim 10 (depends on 7) The system of claim 7, the computer-readable storage medium having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising: applying the analyzing, the determining, and the generating the lift classification and the remodel score steps to a plurality of stores, to yield a plurality of remodel scores; ranking the plurality of remodel scores, to yield ranked remodel scores; and identifying, as part of the determining that the remodel score is above the threshold, that the target store is ranked highest among the ranked remodel scores. ### Claim 11 (depends on 7) The system of claim 7, wherein the cost estimate identifies costs for: HVAC (Heating, Ventilation, and Air Conditioning), Refrigeration, Fixtures, and Flooring. ### Claim 12 (depends on 7) The system of claim 7, wherein the generating of the remodel score comprises a weighted combination of a measure of profit for the target store and a measure of a need of remodeling the target store. ### Claim 13 A non-transitory computer-readable storage medium having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising: generating a trained Bayesian Structural Time Series machine learning model, wherein the trained Bayesian Structural Time Series machine learning model is iteratively trained by a first training data set comprising sales data for one or more stores; analyzing a profitability impact of a remodel on a target store using the trained Bayesian Structural Time Series machine learning model; generating a trained non-parametric regression machine learning model, wherein the trained non-parametric regression machine learning model is iteratively trained by a second training data set; performing, in parallel via the computing device: determining a cost estimate for the remodel using the trained non-parametric regression machine learning model; and generating, using the profitability impact, a lift classification of the target store; generating a remodel score for the target store based on the lift classification and the cost estimate of the remodel; and determining that the remodel score is above a threshold, to yield a determination. ### Claim 14 (depends on 13) The non-transitory computer-readable storage medium of claim 13, wherein the first training data set comprises historical sales data of the target store; wherein the trained Bayesian Structural Time Series machine learning model implements a Kalman filter to iteratively predict future sales data of the store based on the historical sales data of the store; and wherein generating the trained non-parametric regression machine learning model includes assigning weights to expenditures for the store based on historical data of previously remodeled stores. ### Claim 15 (depends on 14) The non-transitory computer-readable storage medium of claim 14, wherein the determining that the remodel score is above the threshold is based on a linear optimization model, where the linear optimization model uses, as input, a total budget. ### Claim 16 (depends on 13) The non-transitory computer-readable storage medium of claim 13, having additional instructions stored which, when executed by the computing device, cause the computing device to perform operations comprising: applying the analyzing, the determining, and the generating the lift classification and the remodel score steps to a plurality of stores, to yield a plurality of remodel scores; ranking the plurality of remodel scores, to yield ranked remodel scores; and identifying, as part of the determining that the remodel score is above the threshold, that the target store is ranked highest among the ranked remodel scores. ### Claim 17 (depends on 13) The non-transitory computer-readable storage medium of claim 13, wherein the cost estimate identifies costs for: HVAC (Heating, Ventilation, and Air Conditioning), Refrigeration, Fixtures, and Flooring.