Demand-model based price image calculation method and computer program therefor

• US 20040049470A1
• Filed: 08/07/2003
• Published: 03/11/2004
• Est. Priority Date: 05/21/1998
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: Khimetrics, Inc. (SAP SE)
• Inventors: Ouimet, Kenneth J.
• Granted Patent: US 7,603,286 B2
• Time in Patent Office: Days
• US Class Current: 705/400
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 30/0206 Price or cost determination based on market factors
• CPC Class Codes: G06Q 30/0283 Price estimation or determination

Abstract

A demand control process (42) includes a price-image computation subprocess (44) and an optimization subprocess (46). The price-image computation subprocess (44) tunes a demand model to both price (76) and non-price (78) demand-parameters. Baseline and trial pricing scenarios (98, 114) are then compiled for a selected set (86) of products and operated upon by the demand model to forecast demand (110, 118) at the specified prices. Revenues (122) are calculated at the baseline pricing scenario (98) using the forecast demand (110), and the revenues (122) are used to calculate weights for compiling summary statistics over the set (86) of products. Price image is then calculated using the values forecast by the demand model so that the resulting price image is responsive to price and non-price demand parameters (76, 78).

Claims

1. A method of computing a price image for a set of products of an enterprise comprising:

processing historical data describing activity of said enterprise to identify price and non-price factors of a demand model to tune said demand model to said historical data so that said demand model can forecast demand from price and non-price input parameters;

calculating weighting factors for each product in said set of products, said weighting factors being based upon at least a portion of said demand factors identified in said processing activity; and

computing said price image for said set of products to be responsive to said weighting factors.

2. A method as claimed in claim 1 additionally comprising:

operating said demand model to forecast demand, for each product in a set of products, said forecast demand being based upon said price and non-price factors identified in said processing activity.

3. A method as claimed in claim 2 wherein:

said processing activity determines reference prices from said historical data, said reference prices indicating prices that customers expect to pay; and

said calculating activity is configured so that said weighting factors are calculated based upon demand forecast at prices substantially equal to said reference prices.

4. A method as claimed in claim 2 wherein said processing activity calculates said reference prices to be predominantly responsive to said historical data.

5. A method as claimed in claim 2 wherein:

a portion of said historical data is historical pricing data; and

said processing activity calculates said reference prices as being different from actual historical prices in response to historical price changes.

6. A method as claimed in claim 2 wherein:

said computing activity calculates said price image to be responsive to differences in demand between said demand forecast for a trial pricing scenario and said demand forecast for a baseline pricing scenario.

7. A method as claimed in claim 6 wherein:

said demand is expressed as unit sales;

in said trial pricing scenario, at least some prices for selected ones of said products differ from corresponding prices in said baseline pricing scenario;

said differences in demand are configured as normalized changes in unit sales for each of said products between said baseline and trial pricing scenarios; and

said price image computing activity computes an average normalized change in unit sales over said set of products so that said price image is responsive to said average normalized change in unit sales.

8. A method as claimed in claim 1 wherein:

said processing activity comprises identifying price elasticities of demand for said set of products; and

said price image computing activity computes said price image to be responsive to said price elasticities of demand.

9. A method as claimed in claim 8 wherein:

said price image computing activity determines a weighted average price elasticity of demand over said set of products from said price elasticities of demand; and

said price image computing activity computes said price image to be responsive to said weighted average price elasticity of demand.

10. A method as claimed in claim 1 wherein said non-price demand factors and non-price input parameters characterize promotional activity so that said price image is responsive to said promotional activity.

11. A method as claimed in claim 1 wherein said non-price demand factors and non-price input parameters characterize product availability so that said price image is responsive to said product availability.

12. A method as claimed in claim 1 wherein said non-price demand factors and non-price input parameters characterize seasonality effects so that said price image is responsive to said seasonality effects.

13. A method as claimed in claim 1 wherein said non-price demand factors and non-price input parameters characterize customer types so that said price image is responsive to said customer types.

14. A method as claimed in claim 1 wherein said non-price demand factors and non-price input parameters characterize psychological price thresholds so that said price image is responsive to said price thresholds.

15. A method as claimed in claim 1 wherein said weighting factors are influenced by revenue for each of said products substantially in proportion to a total revenue for said set of products.

16. A method as claimed in claim 1 additionally comprising processing historical data describing activity of said enterprise to identify factor values associated with price and non-price demand parameters of said demand model to tune said demand model to said historical data.

17. A method as claimed in claim 1 wherein:

said method additionally comprises operating said demand model to forecast demand based upon said price and non-price input parameters; and

said computing activity computes said price image to be responsive to change in demand for said set of products.

18. A method of controlling demand for products offered by an enterprise through the use of price image, said method comprising:

processing historical data describing activity of said enterprise to identify factors of a demand model that cause said demand model to forecast demand from price and non-price input parameters;

calculating a price image to be responsive to said demand model factors identified in said processing activity; and

reporting said price image.

19. A method as claimed in claim 18 wherein said processing activity calculates reference prices from said historical data, said reference prices characterizing prices that customers expect to pay.

20. A method as claimed in claim 18 wherein:

a portion of said historical data is actual historical pricing data; and

said processing activity calculates said reference prices as being different from actual historical prices in response to historical price changes.

21. A method as claimed in claim 18 wherein said processing activity step calculates said reference prices to be predominantly responsive to said historical data.

22. A method as claimed in claim 18 wherein:

said processing activity identifies a price elasticity of demand for each of said products; and

said calculating activity calculates said price image to be responsive to said price elasticities of demand.

23. A method as claimed in claim 22 wherein:

said calculating activity determines a weighted average price elasticity of demand over said products from said price elasticities of demand; and

said calculating activity calculates said price image to be responsive to said weighted average price elasticity of demand.

24. A method as claimed in claim 23 wherein:

said method additionally comprises calculating revenues for each of said products; and

said weighted average price elasticity of demand is influenced by said price elasticity of demand for each of said products in proportion to revenues for each of said products.

25. A method as claimed in claim 18 wherein:

said method additionally comprises establishing a trial pricing scenario for said products in which at least some prices for selected ones of said products differ from corresponding prices for said selected products in a baseline pricing scenario;

said method additionally comprises determining a normalized change in demand between said trial and baseline pricing scenarios in response to operation of said demand model; and

said calculating activity calculates an average normalized change in demand over said products so that said price image is responsive to said average normalized change in demand.

26. A method as claimed in claim 25 wherein:

said processing activity identifies price elasticities of demand of said products; and

said calculating activity calculates an average price elasticity of demand over said products and calculates said price image to be responsive to said average price elasticity of demand in combination with said average normalized change in demand.

27. A method as claimed in claim 25 wherein said average normalized change in demand is a weighted average.

28. A method as claimed in claim 27 wherein:

said method additionally comprises operating said demand model to forecast demand;

said method additionally comprises calculating revenues for each of said products in response to said forecast demand; and

said weighted average of said average normalized change in demand is influenced by said forecast normalized change in demand for each of said products in proportion to revenues for each of said products.

29. A method as claimed in claim 27 wherein said weighted average dynamically responds to changes in forecast normalized change in demand for each of said products by said demand model.

30. A method as claimed in claim 18 additionally comprising establishing prices for said products in response to said price image reported in said reporting activity.

31. A method as claimed in claim 18 additionally comprising promoting said products in response to said price image reported in said reporting activity.

32. A computer program residing upon a computer-readable medium and configured to compute a price image for a set of products of an enterprise, said computer program comprising:

a first code segment that processes price and non-price historical data describing activity of said enterprise to tune factors for a demand model;

a second code segment that identifies price elasticities of demand for said set of products from said demand model;

a third code segment that causes said demand model to forecast a trial demand for said set of products based upon a trial price scenario for said set of products and to forecast a baseline demand for said set of products based upon a baseline price scenario for said set of products;

a fourth code segment that calculates normalized demand changes for said set of products between said trial and baseline price scenarios; and

a fifth code segment that calculates said price image for said set of products in response said normalized demand changes and in response to said price elasticities of demand.

33. A computer program as claimed in claim 32 additionally comprising:

a sixth code segment that calculates revenues for said set of products in response to one of said baseline and trial demands forecast in response to operation of said third code segment; and

a seventh code segment that calculates weighting factors for each of said products in said set of products, said weighting factors being applied by said fifth code segment in calculating said price image, and said weighting factors being responsive to said revenues.

34. A computer program as claimed in claim 32 additionally comprising a sixth code segment that generates said baseline pricing scenario predominately in response to said historical price data for said enterprise.