Systems and methods of controlling quantities of denominations of currency at a retail shopping facility

Patent metadata

Field	Value
Publication number	US-10402764-B2
Application number	US-201715474032-A
Country	US
Kind code	B2
Filing date	Mar 30, 2017
Priority date	Apr 1, 2016
Publication date	Sep 3, 2019
Grant date	Sep 3, 2019

Abstract

In some embodiments, systems and methods are provided to forecast denomination needs at one or more retail shopping facilities. The system comprises a control circuit configured to determine a predicted optimized quantity of each of multiple different denominations of currency to be available for use at a retail shopping facility over a future period of time as a function of historic needs; select for each of the multiple denominations one of multiple different models; forecast a quantity of each of the multiple denominations to be present at the termination of the first future period of time by applying the selected one of the models; and determine, for each of the multiple denominations, whether the forecasted quantity of the denomination is greater than a threshold quantity of that denomination of the currency.

First claim

What is claimed is:

1. A system to forecast denomination needs at a retail shopping facility, comprising:
a control circuit coupled with a transceivers; and
a memory coupled to the control circuit and storing computer instructions that when executed by the control circuit cause the control circuit to:
determine a predicted optimized quantity of each of multiple different denominations of currency to be available for use at a first retail shopping facility over a first future period of time as a function of historic needs of each of the multiple denominations, wherein the first future period of time is at least twenty four hours in the future;
select for each of the multiple denominations one of multiple different models;
forecast a quantity of each of the multiple denominations to be present at the termination of the first future period of time by applying the selected one of the models;
determine, for each of the multiple denominations of the currency, whether the forecasted quantity of the denomination of the currency is greater than a threshold quantity of that denomination of the currency;
cause the optimized quantity of each of multiple different denominations to be maintained by the retail shopping facility as beginning quantities at the beginning of the first future period of time and at least portions of each of the denominations distributed to point of sale systems of the retail shopping facility for use during the first future period of time; and
force out an additional denomination of the currency that has a value between two of the multiple denominations from use at the first shopping facility disregarding the additional denomination of the currency, and does not determine a predicted optimized quantity for the additional denomination.

2. The system of claim 1 , wherein the control circuit, in selecting for each of the multiple denominations one of the multiple different models, is further configured to:
access historic data over a historic period of time;
for each denomination:
train each of the multiple models using a first percentage of the historic data corresponding to a first percentage of the historic period of time;
apply each of the multiple trained models to a second percentage of the historic data corresponding to a second percentage of the historic period of time and generate a plurality of historic forecasts of a quantity of the denomination over the second percentage of the historic period of time; and
identify the selected model producing the historic forecasts of the quantity of the denomination that corresponds to actual quantity needs of the denomination during the second percentage of the historic period of time.

3. The system of claim 2 , wherein the control circuit is further configured to determine, for each forecasted quantity of the denomination, an error factor from the difference between the historic forecasted quantity of the denomination and the actual quantity need of the denomination, and determine minimum average percentage error of the differences.

4. The system of claim 2 , wherein the control circuit is further configured to identify when there is insufficient historic data corresponding to the first retail shopping facility, identify a secondary retail shopping facility that has sales consistent with the first shopping facility, and access the historic data of the historic period of time corresponding to the secondary retail shopping facility.

5. The system of claim 1 , wherein the control circuit is further configured to determine a future change order of a specified quantity of a first denomination to be received at the first shopping facility during the first future period of time from a currency source, wherein the future change order is determined as a function of a difference between the forecasted quantity of the first denomination at the termination of the first future period of time and the threshold quantity corresponding the first denomination.

6. The system of claim 1 , wherein the control circuit, in determining the predicted optimized quantity of each of the multiple different denominations, is further configured to evaluate cost difference between holding an excess quantity of one or more of the multiple denominations versus depositing the excess quantity of the one or more of the multiple denominations with a banking entity.

7. The system of claim 6 , wherein the control circuit, in evaluating the cost difference, is further configured to calculate expected interest income from the depositing of the excess quantity, and an identified cost associated with receiving from a currency source a change order comprising an additional quantity of one or more of the multiple denominations for use at the first shopping facility during the first future period of time.

8. A method of forecasting denomination needs at a retail shopping facility, comprising:
by a denomination forecasting control circuit comprising a processor implementing code to perform the steps of:
determining a predicted optimized quantity of each of multiple different denominations of currency to be available for use at a first retail shopping facility over a first future period of time as a function of historic needs of each of the multiple denominations, wherein the first future period of time is at least twenty four hours in the future;
selecting for each of the multiple denominations one of multiple different models;
forecasting a quantity of each of the multiple denominations to be present at the termination of the first future period of time by applying the selected one of the models;
determining, for each of the multiple denominations of the currency, whether the forecasted quantity of the denomination of the currency is greater than a threshold of the forecasted quantity of that denomination of the currency;
causing the optimized quantity of each of multiple different denominations to be maintained by the retail shopping facility as beginning quantities at the beginning of the first future period of time and at least portions of each of the denominations distributed to point of sale systems of the retail shopping facility for use during the first future period of time; and
forcing out an additional denomination of the currency, which has a value between two of the multiple denominations, from use at the first shopping facility disregarding the additional denomination of the currency, and not determining a predicted optimized quantity for the additional denomination.

9. The method of claim 8 , further comprising:
accessing historic data over a historic period of time;
for each denomination:
training each of the multiple models using a first percentage of the historic data corresponding to a first percentage of the historic period of time;
applying each of the multiple trained models to a second percentage of the historic data corresponding to a second percentage of the historic period of time and generate a plurality of historic forecasts of a quantity of the denomination over the second percentage of the historic period of time; and
identifying the selected model producing the historic forecasts of the quantity of the denomination that corresponds to actual quantity needs of the denomination during the second percentage of the historic period of time.

10. The method of claim 9 , further comprising:
determining, for each forecasted quantity of the denomination, an error factor from the difference between the historic forecasted quantity of the denomination and the actual quantity need of the denomination; and
determining minimum average percentage error of the differences.

11. The method of claim 9 , wherein the accessing the historic data over

Assignees

• Walmart Apollo Llc

Inventors

• Nail Betsy Goosman
• Johnson Christopher M
• Nazarian Ehsan

Classifications

• G06Q10/06315 Primary

  Needs-based resource requirements planning or analysis · CPC title

• G06Q30/0601

  Electronic shopping [e-shopping] · CPC title

• G06N20/00

  Machine learning · CPC title