# US 2026/0094103 A1 · Multi-layer prescriptive recommendations machine learning model

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0094103 A1
- Document ID: US20260094103A1
- Status: Published application
- Kind: A1
- Application number: 18/902,703
- Filed: 2024-09-30
- Published: 2026-04-02
- Earliest claimed: 2024-09-30
- Assignees: NCR Voyix Corporation
- Applicants: NCR Voyix Corporation
- Inventors: Itamar David Laserson; Shiran Abadi; Shay Marom; Elisheva Chocron
- Web page: https://patentgrep.com/patent/US20260094103A1

## CPC

- G06Q 10/06393: {Score-carding, benchmarking or key performance indicator [KPI] analysis}
- 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

## Abstract

A generic prescriptive recommendations machine learning model (MLM) for improving store operations uses a multi-layer architecture of MLMs. The first layer predicts metric values for a given key performance indicator (KPI), while the second layer predicts feature values for given metric values. This hierarchical structure enables the detection of underperforming KPIs, sets measurable and achievable improvement targets, and recommends specific actions. The multi-layer MLM integrates data analysis into practical applications, improving efficiency and decision-making in retail operations without requiring extensive technical expertise. This automated, data-driven solution addresses the challenges faced by store managers in identifying and resolving operational weaknesses across large retail organizations.

## Claims

### Claim 1

A method comprising:
identifying, by a first machine learning model (MLM) an underperforming key performance indicator (KPI) for a store;
predicting, by the first MLM, a predicted metric value for the underperforming KPI;
predicting, by a second layer of one or more MLMs, a predicted feature value for the predicted metric value, wherein each of the one or more MLMs in the second layer corresponds to a unique metric of the underperforming KPI; and
generating, based on the predicted feature value, a prescriptive action to enable improvement of the underperforming KPI at the store.

### Claim 2 (depends on 1)

The method of claim 1, wherein identifying the underperforming KPI further includes comparing KPI values of the store to other KPI values of other stores in a retail chain.

### Claim 3 (depends on 1)

The method of claim 1, wherein identifying the underperforming KPI further includes detecting specific KPIs where the store has an opportunity to improve based on historical data.

### Claim 4 (depends on 1)

The method of claim 1, wherein predicting the predicted metric value further includes applying a feature contribution analysis to detect an amount of contribution of each of a plurality of features to a change in the underperforming KPI.

### Claim 5 (depends on 4)

The method of claim 4, further comprising filtering out insignificant features based on a threshold of importance.

### Claim 6 (depends on 1)

The method of claim 1, wherein predicting the predicted feature value further includes optimizing each of selected business features to reach a target value for the predicted metric value.

### Claim 7 (depends on 6)

The method of claim 6, wherein optimizing the selected business features further include using an optimization algorithm that considers a potential optimization direction for each selected business feature.

### Claim 8 (depends on 1)

The method of claim 1, wherein generating the prescriptive action further includes setting a measurable and an actionable target to improve the underperforming KPI.

### Claim 9 (depends on 1)

The method of claim 1, further comprising validating a performance of the first MLM and each of the one or more MLMs of the second layer to ensure a proper fitting.

### Claim 10 (depends on 1)

The method of claim 1, further comprising filtering out one or more optimization values that result in a negligible change in the underperforming KPI.

### Claim 11 (depends on 1)

The method of claim 1, further comprising using the predicted feature value to predict the predicted metric value and using the predicted metric value to predict an improvement in the underperforming KPI.

### Claim 12

A method comprising:
receiving store operation data for a plurality of stores;
training a multi-layer architecture of regression machine learning models (MLMs) using the store operation data;
detecting, using a first layer of the multi-layer architecture, at least one underperforming key performance indicator (KPI) where a particular store is underperforming;
setting, using the first layer, a measurable and an achievable target to improve the at least one underperforming KPI;
determining, using a second layer of the multi-layer architecture, a specific action to achieve the measurable and the achievable target; and
providing the specific action as a recommendation to enable an improvement in operations of the particular store.

### Claim 13 (depends on 12)

The method of claim 12, wherein training the multi-layer architecture further includes training the first layer to predict metric values for a particular KPI and training the second layer to predict feature values for particular metric.

### Claim 14 (depends on 12)

The method of claim 12, wherein detecting the at least one underperforming KPI further includes applying a feature contribution analysis to identify features significantly contributing to KPI performance above a threshold.

### Claim 15 (depends on 12)

The method of claim 12, wherein setting the measurable and the achievable target further include optimizing metric values to reach a desired change in the at least one underperforming KPI.

### Claim 16 (depends on 12)

The method of claim 12, wherein determining the specific action further includes optimizing feature values to reach the measurable and the achievable target for metrics.

### Claim 17 (depends on 12)

The method of claim 12, further comprising validating performance of each regression MLM in the multi-layer architecture.

### Claim 18 (depends on 12)

The method of claim 12, further comprising discarding any recommendation that result in negligible improvements to at least one KPI.

### Claim 19

A system comprising:
a processor; and
a memory storing instructions that, when executed by the processor, cause the system to:
implement a multi-layer architecture of machine learning models (MLMs);
identify, using a first layer of the multi-layer architecture, at least one underperforming key performance indicator (KPI) for a store;
predict, using the first layer, at least one metric value for the at least one underperforming KPI;
predict, using a second layer of the multi-layer architecture, at least one feature value for the at least one metric value, wherein each regression MLM in the second layer corresponds to a unique metric of the at least one underperforming KPI; and
generate, based on the at least one feature value, a prescriptive action for improving the at least one underperforming KPI.

### Claim 20 (depends on 19)

The system of claim 19, wherein the instructions further cause the processor to optimize the at least one feature value using an optimization algorithm that considers a potential optimization direction for each feature of the at least one feature value.
