# US 12,223,535 B2 · Network-based inventory replenishment

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

- Number: US 12,223,535 B2
- Document ID: US12223535B2
- Status: Granted
- Kind: B2
- Application number: 17/458,928
- Filed: 2021-08-27
- Published: 2022-02-17
- Granted: 2025-02-11
- Earliest claimed: 2017-01-16
- Assignees: NCR Voyix Corporation
- Applicants: NCR Voyix Corporation
- Inventors: Yehoshua Zvi Licht; Brent Vance Zucker
- Web page: https://patentgrep.com/patent/US12223535B2

## CPC

- G06Q 30/0631: {Recommending goods or services}
- G06Q 10/083: Shipping
- G06Q 10/087: Inventory or stock management, e.g. order filling, procurement or balancing against orders
- G06Q 10/08726: {for replenishment processing, procedures, or recommendations using forecasting or optimisation}
- G06Q 30/0613: {using intermediate agents}
- G06Q 30/0635: {replenishment orders; recurring orders}
- 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 network-based inventory replenishment application maintains and analyzes item transactions on a per-consumer basis. Specific item purchases are calculated as being in need of replenishment by a specific consumer on specific future dates or a specific range of future dates. The consumer is proactively informed and asked if such items should be ordered. The network-based inventory replenishment application engages the consumer for item replenishment and item order processing over a plurality of disparate communication channels.

## Claims

### Claim 1

A method, comprising:
predicting, by a processor, at least one item that a customer frequently orders based on time gaps, non-linear sequences associated with the time gaps, frequency, and pattern analysis of transaction records from a transaction history associated with the customer, wherein predicting further includes predicting the at least one item without performing any clustering on the transaction records, wherein the pattern analysis includes identifying consumption patterns of the customer on a per item basis, identifying frequencies for each consumption pattern, identifying the time gaps and the non-linear sequences of the time gaps associated with each frequency and each corresponding consumption pattern;
maintaining, by the processor, a purchase history of regularly purchased items of the customer in a cloud-based inventory management system;
predicting, by the processor, when an order should be placed for a portion of the regularly purchased items by the a machine-learning model;
processing, by the processor, a recurrent neural network algorithm to identify patterns and frequencies of the patterns to predict a future date as being when the order should be placed for the portion of the regularly purchased items;
automatically ordering, by the processor, the at least one item on behalf of the customer within an order system using an Application Programming Interface (API) when a profile maintained for the customer provides customer authorization for the automatically ordering;
wherein the processor is further configured to execute a machine learning model that dynamically updates the prediction based on real-time transaction data and customer feedback; and
ordering, by the processor, the at least one item with the order system using the API after receiving the customer authorization through an interactive dialogue with the customer.

### Claim 2 (depends on 1)

The method of claim 1, wherein predicting further includes processing the recurrent neural network with machine learning to perform the time gaps, non-linear sequences associated with the time gaps, frequency, and pattern analysis on the transaction records from the transaction history.

### Claim 3 (depends on 1)

The method of claim 1, wherein predicting further includes identifying the at least one item as a plurality of grocery items that the customer purchases with regularity based on the time gaps, frequency, and pattern analysis.

### Claim 4 (depends on 1)

The method of claim 1, wherein predicting further includes periodically collecting new transaction records from a plurality of channels and updating the new transaction records to the transaction history.

### Claim 5 (depends on 1)

The method of claim 1, wherein predicting further includes predicting the at least one item without any input from the customer based entirely on the time gaps, non-linear sequences associated with the time gaps, frequency, and pattern analysis.

### Claim 6 (depends on 1)

The method of claim 1, wherein automatically ordering further includes automatically ordering the at least one item at a time that precedes a predicted date that the at least one item was likely to be purchased by the customer.

### Claim 7 (depends on 1)

The method of claim 1, wherein ordering further includes engaging the customer in the interactive dialogue at a time that precedes a predicted date that the at least one item was likely to be purchased by the customer.

### Claim 8 (depends on 1)

The method of claim 1, wherein ordering further includes modifying the at least one item that is ordered with the order system based on modifications received from the customer during the interactive dialogue.