# US 2021/0174284 A1 · System and method for determining and implementing sales clusters for stores

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

- Number: US 2021/0174284 A1
- Document ID: US20210174284A1
- Status: Published application
- Kind: A1
- Application number: 17/172,185
- Filed: 2021-02-10
- Published: 2021-06-10
- Earliest claimed: 2017-12-28
- Assignees: Walmart Apollo, LLC
- Applicants: Walmart Apollo, LLC
- Inventors: Somedip Karmakar; Sourit Manna; Gayatri Pal
- Web page: https://patentgrep.com/patent/US20210174284A1

## CPC

- G06Q 10/06315: {Needs-based resource requirements planning or analysis}
- G06Q 10/08: Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q 10/087: Inventory or stock management, e.g. order filling, procurement or balancing against orders
- G06Q 10/08724: {by shelf level inventory management, e.g. planograms}
- G06Q 30/02: Marketing; Price estimation or determination; Fundraising
- G06Q 30/0204: Market segmentation
- 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

Initial sales cluster is divided by the control circuit into a plurality of velocity buckets. Subsequently, each velocity bucket is divided into a plurality of micro-clusters. The micro-clusters are defined according to demographic information or store characteristic information. A importance score and a performance score for each of the micro-clusters is determined. An optimal sales cluster and a corresponding optimal planogram for each retail store in each micro-cluster are determined based upon the importance score and the performance score.

## Claims

### Claim 1

A system that determines and implements optimal planograms in retail stores to maximize store performance, the system comprising:
a plurality of retail stores;
shelving units disposed in each of the plurality of retail stores, the shelving units configured to display products to customers;
a plurality of sales entry devices that are configured to collect sales data at the plurality of retail stores;
a transceiver circuit disposed at a central processing center, the transceiver circuit being configured to receive the sales data, and additionally receive product characteristic data for products stocked in the plurality of retail stores;
a database disposed at the central processing center, the database configured to store the sales data and the product characteristic data;
a control circuit disposed at the central processing center, the control circuit being coupled to the transceiver circuit and the database, the control circuit being configured to:
form initial sales clusters of retail stores from the plurality of retail stores according to one of a plurality of algorithms using the product characteristic data and the sales data, each of the initial sales clusters including a subset of the plurality of retail stores, formation of the initial sales clusters of retail stores from the plurality of retail stores being effective to maximize the compactness of the initial sales clusters and to minimize the total number of initial sales clusters, each of the initial sales clusters being assigned an initial planogram;
divide each initial sales cluster into a plurality of velocity buckets, each of the velocity buckets relating to a velocity of product sales according to the sales data for retail stores in the velocity bucket;
subsequently divide each velocity bucket into a plurality of micro-clusters, the micro-clusters being defined according to demographic information or store characteristic information;
using the sales data, determine an importance score and a performance score for each of the micro-clusters, the importance score indicating the relative ranking a micro-cluster compared to other micro-clusters in accomplishing store performance goals, and the performance score indicating the absolute operational performance of the retail stores in the micro-cluster in accomplishing the store performance goals;
selectively determine an optimal sales cluster and a corresponding optimal planogram for each retail store in each micro-cluster based upon the importance score and the performance score, the optimal planogram being selected to maximize store performance;
determine and implement an action based upon the optimal planogram, the action including stocking the shelving units at the stores according to the optimal planogram, the stocking being accomplished by moving or rearranging products within the retail stores to the shelving units in the retail stores, the action also including reconfiguring at least one of the shelving units in at least one of the plurality of retail stores;
wherein the performance score indicates that the micro-cluster is generally ideal in performance, over-performing, under-performing, or at-risk of under-performing, and the importance score indicates that the micro-cluster is significant relative to other micro-clusters or not significant relative to other micro-clusters;
wherein when a micro-cluster is under-performing and significant, the optimal planogram is a newly created planogram;
wherein when a micro-cluster is over-performing and not significant, then the optimal planogram is the same as an original modular planogram of the sales cluster;
wherein the controller is configured to cause the display of significant micro-clusters, importance scores associated with each of the significant micro-clusters, and the performance scores associated with each of the significant micro-clusters.

### Claim 2 (depends on 1)

The system of claim 1, wherein the store performance goal is one or more of:
year-over-year sales trends, year-over-year category growth, market or customer penetration, or category share of a department.

### Claim 3 (depends on 1)

The system of claim 1, wherein the performance score and the importance score reflect differences between a micro-cluster average and a national average.

### Claim 4 (depends on 1)

The system of claim 1, wherein the optimal sales cluster is a newly created sales cluster with a newly created planogram.

### Claim 5 (depends on 1)

The system of claim 1, wherein the optimal sales cluster is one of the initial sales clusters with the initial planogram for the initial sales cluster.

### Claim 6 (depends on 1)

The system of claim 1, further comprising rendering profile information concerning the initial sales clusters or the micro-clusters at a user interface.

### Claim 7 (depends on 1)

The system of claim 1, wherein the importance score is a weighted average of the absolute relative changes of store traits from a cluster average and a national average multiplied by a variable importance.

### Claim 8 (depends on 1)

The system of claim 1, wherein the variable importance is obtained from a Random forest model.

### Claim 9 (depends on 1)

The system of claim 1, wherein the algorithms are selected from the group consisting of: a K-means algorithm, a K-medoids algorithm, a Ward's Clustering algorithm, and a Convex Clustering algorithm.

### Claim 10

A method that determines and implements optimal planograms in retail stores to maximize store performance, the method comprising:
providing a plurality of shelving units disposed in each of the plurality of retail stores, the shelving units configured to display products to customers;
collecting sales data at a plurality of sales entry devices at the plurality of retail stores;
receiving the sales data and product characteristic data for products stocked in the plurality of retail stores at a central processing center;
storing the sales data and the product characteristic data at a database at the central processing center;
at control circuit disposed at the central processing center, forming initial sales clusters of retail stores from the plurality of retail stores according to one of a plurality of algorithms using the product characteristic data and the sales data, each of the initial sales clusters including a subset of the plurality of retail stores, the forming of the initial sales clusters of retail stores from the plurality of retail stores being effective to maximize the compactness of the initial sales clusters and to minimize the total number of initial sales clusters, each of the initial sales clusters being assigned an initial planogram, dividing each initial sales cluster into a plurality of velocity buckets, each of the velocity buckets relating to a velocity of product sales according to the sales data for retail stores in the velocity bucket, subsequently dividing each velocity bucket into a plurality of micro-clusters, the micro-clusters being defined according to demographic information or store characteristic information, using the sales data, determining an importance score and a performance score for each of the micro-clusters, the importance score indicating the relative ranking a micro-cluster compared to other micro-clusters in accomplishing store performance goals, and the performance score indicating the absolute operational performance of the retail stores in the micro-cluster in accomplishing the store performance goals, selectively determining an optimal sales cluster and a corresponding optimal planogram for each retail store in each micro-cluster based upon the importance score and the performance score, the optimal planogram maximizing store performance, and determining and implementing an action based upon the optimal planogram, the action including stocking the shelving units of the stores according to the optimal planogram, the stocking being accomplished by moving or rearranging products within the retail stores to the shelving units in the retail stores, the action also including reconfiguring at least one of the shelving units in at least one of the plurality of retail stores;
wherein the performance score indicates that the micro-cluster is ideal in performance, over-performing, under-performing, or at-risk of under-performing, and the importance score indicates that the micro-cluster is significant relative to other micro-clusters or not significant relative to other micro-clusters;
wherein when a micro-cluster is under-performing and significant, the optimal planogram is a newly created planogram;
wherein when a micro-cluster is over-performing and not significant, then the optimal planogram is the same as an original modular planogram of the sales cluster;
wherein the controller is configured to cause the display of significant micro-clusters, importance scores associated with each of the significant micro-clusters, and the performance scores associated with each of the significant micro-clusters.

### Claim 11 (depends on 10)

The method of claim 10, wherein the store performance goal is one or more of: year-over-year sales trends, year-over-year category growth, market or customer penetration, or category share of a department.

### Claim 12 (depends on 10)

The method of claim 10, wherein the performance score and the importance score reflect differences between a micro-cluster average and a national average.

### Claim 13 (depends on 10)

The method of claim 10, wherein the optimal sales cluster is a newly created sales cluster with a newly created planogram.

### Claim 14 (depends on 10)

The method of claim 10, wherein the optimal sales cluster is one of the initial sales clusters with the initial planogram for the initial sales cluster.

### Claim 15 (depends on 10)

The method of claim 10, further comprising rendering profile information concerning the initial sales clusters or the micro-clusters at a user interface.

### Claim 16 (depends on 10)

The method of claim 10, wherein the importance score is a weighted average of the absolute relative changes of store traits from a cluster average and a national average multiplied by a variable importance.

### Claim 17 (depends on 10)

The method of claim 10, wherein the variable importance is obtained from a Random forest model.

### Claim 18 (depends on 10)

The method of claim 10, wherein the algorithms are selected from the group consisting of: a K-means algorithm, a K-medoids algorithm, a Ward's Clustering algorithm, and a Convex Clustering algorithm.
