Computer program product and method for sales forecasting and adjusting a sales forecast

• US 8,392,228 B2
• Filed: 03/24/2010
• Issued: 03/05/2013
• Est. Priority Date: 03/24/2010
• Status: Active Grant

• Current Assignee: One Network Enterprises, Inc.
• Original Assignee: One Network Enterprises, Inc.
• Inventors: Mulukutla, Ramkartik, Notani, Ranjit
• Primary Examiner(s): Dixon, Thomas
• Assistant Examiner(s): Fields, Benjamin S
• Application Number:
  US12/730,805
• Publication Number:
  US 20110238461A1
• Time in Patent Office: 1,077 Days
• Field of Search: 705/10, 705/7.11, 705/7.42
• US Class Current: 705/7.11
• CPC Class Codes: 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
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 30/06 Buying, selling or leasing transactions

Abstract

A computer program product and method for sales forecasting and adjusting a sales forecast for an enterprise in a configurable region having one or more clusters of stores. The method includes periodically receiving a sales forecast for an enterprise over a configurable period of time, periodically receiving actual sales information, sales anomalies and anticipated events within the at least one of the clusters of stores over a computer network, determining positive and negative deviations from the anticipated sales of the sales forecast based on the sales information, determining whether one or more trends are occurring or have occurred using a pre-defined mathematical expression based on the sales information, the positive and negative deviations, and the sales anomalies, adjusting the anticipated sales of the sales forecast based on the sales anomalies, the trends and the anticipated events, and outputting the adjusted sales forecast to a user. The sales forecast includes anticipated sales for a plurality of items within at least one of the clusters of stores.

Claims

1. A computer program product embodied on a non-transitory computer readable medium for retail sales forecasting and adjusting a retail sales forecast for an enterprise in a configurable region having one or more clusters of stores, wherein the computer program product is implemented by one or more processors executing processor instructions, the computer program product comprising:

a first computer code for periodically receiving a retail sales forecast for an enterprise over a configurable period of time, wherein the retail sales forecast includes anticipated retail sales for a plurality of items within at least one of the clusters of stores;

a second computer code for periodically receiving actual retail sales information, retail sales anomalies and anticipated events within the at least one of the clusters of stores over a computer network;

a third computer code for determining positive and negative deviations from the anticipated retail sales of the retail sales forecast based on the retail sales information;

a fourth computer code for determining whether one or more trends are occurring or have occurred based on the retail sales information, the positive and negative deviations, and the retail sales anomalies, wherein determining whether one or more trends are occurring or have occurred uses a pre-defined mathematical expression;

a fifth computer code for adjusting the anticipated retail sales of the retail sales forecast based on the retail sales anomalies, the trends and the anticipated events, wherein adjusting the anticipated retail sales of the retail sales forecast uses a pre-defined mathematical expression; and

a sixth computer code for outputting the adjusted retail sales forecast to a user.

2. The computer program product of claim 1, wherein the fourth computer code for determining whether one or more trends are occurring or have occurred further comprises:

a seventh computer code for determining whether one or more wide-spread trends are occurring or have occurred during a wide-spread detection period based on the retail sales information, the positive and negative deviations, and the retail sales anomalies, wherein determining whether one or more wide-spread trends are occurring or have occurred uses a first predefined mathematical expression; and

an eighth computer code for determining whether one or more local trends are occurring or have occurred during a wide-spread detection period based on the retail sales information, the positive and negative deviations, and the retail sales anomalies, wherein determining whether one or more local trends are occurring or have occurred uses a second predefined mathematical expression.

3. The computer program product of claim 2, wherein the retail sales forecast includes a week of daily forecasts.

4. The computer program product of claim 3, wherein the pre-defined mathematical expression used to adjust the anticipated retail sales of the retail sales forecast comprises multiplying each daily forecast for the remainder of the week by a mathematical smoothing factor.

5. The computer program product of claim 2, wherein the seventh computer code for determining whether one or more wide-spread trends are occurring or have occurred includes determinations at multiple hierarchical levels.

6. The computer program product of claim 2, wherein the seventh computer code for determining whether one or more wide-spread trends are occurring or have occurred includes determining whether a configurable wide-spread trend confirmation ratio and a configurable wide-spread noise ratio have been met.

7. The computer program product of claim 2, wherein the eighth computer code for determining whether one or more local trends are occurring or have occurred includes determinations at multiple hierarchical levels.

8. The computer program product of claim 2, wherein the eighth computer code for determining whether one or more local trends are occurring or have occurred includes determining whether a configurable local trend confirmation count and a configurable local noise ratio have been met.

9. The computer program product of claim 2, wherein the second computer code for periodically receiving actual retail sales information, retail sales anomalies and anticipated events occurs in near real-time as the retail sales occur.

10. The computer program product of claim 2, wherein the second computer code for periodically receiving actual retail sales information, retail sales anomalies and anticipated events occurs hourly.

11. The computer program product of claim 2, wherein the second computer code for periodically receiving actual retail sales information, retail sales anomalies and anticipated events occurs daily.

12. The computer program product of claim 2, wherein the second computer code for periodically receiving actual retail sales information, retail sales anomalies and anticipated events occurs weekly.

13. The computer program product of claim 2, wherein the eighth computer code for determining whether one or more local trends are occurring or have occurred includes exponential convergence using a mathematical smoothing factor.

27. The computer program product of claim 2, wherein the retail sales forecasting includes manufacturing sales forecasting, the retail sales forecast includes a manufacturing sales forecast, the anticipated retail sales includes anticipated manufacturing sales, the actual retail sales information includes actual manufacturing sales, and wherein the retail sales anomalies include manufacturing sales anomalies.

14. A method for retail and manufacturing sales forecasting and adjusting a retail and manufacturing sales forecast for an enterprise in a configurable region having one or more clusters of stores, the method comprising:

periodically receiving a retail and manufacturing sales forecast for an enterprise over a configurable period of time, wherein the retail and manufacturing sales forecast includes anticipated retail and manufacturing sales for a plurality of items within at least one of the clusters of stores;

periodically receiving actual retail and manufacturing sales information, retail and manufacturing sales anomalies and anticipated events within the at least one of the clusters of stores over a computer network;

determining positive and negative deviations from the anticipated retail and manufacturing sales of the retail and manufacturing sales forecast based on the retail and manufacturing sales information;

determining whether one or more trends are occurring or have occurred based on the retail and manufacturing sales information, the positive and negative deviations, and the retail and manufacturing sales anomalies, wherein determining whether one or more trends are occurring or have occurred uses a pre-defined mathematical expression;

adjusting the anticipated retail and manufacturing sales of the retail and manufacturing sales forecast based on the retail and manufacturing sales anomalies, the trends and the anticipated events, wherein adjusting the anticipated retail and manufacturing sales of the retail and manufacturing sales forecast uses a pre-defined mathematical expression; and

outputting the adjusted retail and manufacturing sales forecast to a user.

15. The method of claim 14, wherein determining whether one or more trends are occurring or have occurred further comprises:

determining whether one or more wide-spread trends are occurring or have occurred during a wide-spread detection period based on the retail and manufacturing sales information, the positive and negative deviations, and the retail and manufacturing sales anomalies, wherein determining whether one or more wide-spread trends are occurring or have occurred uses a first predefined mathematical expression; and

determining whether one or more local trends are occurring or have occurred during a wide-spread detection period based on the retail and manufacturing sales information, the positive and negative deviations, and the retail and manufacturing sales anomalies, wherein determining whether one or more local trends are occurring or have occurred uses a second predefined mathematical expression.

16. The method of claim 15, wherein the retail and manufacturing sales forecast includes a week of daily forecasts.

17. The method of claim 16, wherein the pre-defined mathematical expression used to adjust the anticipated retail and manufacturing sales of the retail and manufacturing sales forecast comprises multiplying each daily forecast for the remainder of the week by a mathematical smoothing factor.

18. The method of claim 15, wherein determining whether one or more wide-spread trends are occurring or have occurred includes determinations at multiple hierarchical levels.

19. The method of claim 15, wherein determining whether one or more wide-spread trends are occurring or have occurred includes determining whether a configurable wide-spread trend confirmation ratio and a configurable wide-spread noise ratio have been met.

20. The method of claim 15, wherein determining whether one or more local trends are occurring or have occurred includes determinations at multiple hierarchical levels.

21. The method of claim 15, wherein determining whether one or more local trends are occurring or have occurred includes determining whether a configurable local trend confirmation count and a configurable local noise ratio have been met.

22. The method of claim 15, wherein periodically receiving actual retail and manufacturing sales information, retail and manufacturing sales anomalies and anticipated events occurs in near real-time as the retail and manufacturing sales occur.

23. The method of claim 15, wherein periodically receiving actual retail and manufacturing sales information, retail and manufacturing sales anomalies and anticipated events occurs hourly.

24. The method of claim 15, wherein periodically receiving actual retail and manufacturing sales information, retail and manufacturing sales anomalies and anticipated events occurs daily.

25. The method of claim 15, wherein periodically receiving actual retail and manufacturing sales information, retail and manufacturing sales anomalies and anticipated events occurs weekly.

26. The method of claim 15, wherein determining whether one or more local trends are occurring or have occurred includes exponential convergence using a mathematical smoothing factor.

28. A computer program product embodied on a non-transitory computer readable medium for manufacturing sales forecasting and adjusting a manufacturing sales forecast for an enterprise in a configurable region having one or more clusters of stores, wherein the computer program product is implemented by one or more processors executing processor instructions, the computer program product comprising:

a first computer code for periodically receiving a manufacturing sales forecast for an enterprise over a configurable period of time, wherein the manufacturing sales forecast includes anticipated manufacturing sales for a plurality of items within at least one of the clusters of stores;

a second computer code for periodically receiving actual manufacturing sales information, manufacturing sales anomalies and anticipated events within the at least one of the clusters of stores over a computer network;

a third computer code for determining positive and negative deviations from the anticipated manufacturing sales of the manufacturing sales forecast based on the manufacturing sales information;

a fourth computer code for determining whether one or more trends are occurring or have occurred based on the manufacturing sales information, the positive and negative deviations, and the manufacturing sales anomalies, wherein determining whether one or more trends are occurring or have occurred uses a pre-defined mathematical expression;

a fifth computer code for adjusting the anticipated manufacturing sales of the manufacturing sales forecast based on the manufacturing sales anomalies, the trends and the anticipated events, wherein adjusting the anticipated manufacturing sales of the manufacturing sales forecast uses a pre-defined mathematical expression; and

a sixth computer code for outputting the adjusted manufacturing sales forecast to a user.