Assortment decisions

• US 7,006,981 B2
• Filed: 04/04/2001
• Issued: 02/28/2006
• Est. Priority Date: 04/04/2001
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: ProfitLogic, Inc. (Oracle Corporation)
• Inventors: Woo, Jonathan W., Leven, Steven M., Rose, Dawn M.
• Primary Examiner(s): Hapiz, Tariq R.
• Assistant Examiner(s): BOSWELL, BETH V
• Application Number:
  US09/826,378
• Publication Number:
  US 20020147630A1
• Time in Patent Office: 1,791 Days
• Field of Search: 705/10, 705/22, 705/28, 705/7, 705/9, 705/8, 235/375, 235/376, 235/377, 700/99, 700/100
• US Class Current: 705/7.31
• CPC Class Codes: G06Q 10/0635 Risk analysis of enterprise or organisation activities
• CPC Class Codes: G06Q 10/06375 Prediction of business process outcome or impact based on a proposed change
• CPC Class Codes: G06Q 10/087 Inventory or stock management, e.g. order filling, procurement or balancing against orders
• CPC Class Codes: G06Q 20/203 Inventory monitoring
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities

Abstract

Historical information is stored about demand behavior associated with different sizes of styles of items of commerce. A size assortment recommendation is generated based on the stored historical information and on objectives with respect to purchase or allocation decisions about assortments of the different sizes.

Claims

1. A method comprising

A. obtaining historical information about sales associated with each of one or more different sizes of styles of items of commerce,

B. determining historical demand behavior associated with the different sizes of styles of items of commerce by filtering the historical information about sales of those styles of items of commerce to remove sales data for each of one or more styles in each of one or more commerce locations that correspond to one or more time periods in which not all sizes of that style were presented to customers in that time period at that commerce location

C. generating a size assortment recommendation based on

i) the historical demand behavior and on

ii) one or more objectives with respect to purchase or allocation decisions about assortments of the different sizes,

D. the step of generating a size assortment recommendation including

identifying a plurality of size profiles, each maximizing on a per commerce location, per time period, per class basis an expected value of realized profit minus a weighting factor times a customer service function,

where the customer service function is a function of a number of time periods between total stockout of a style and a first time period that any size of that style stocks out, and

where a class comprises one or more styles.

2. The method of claim 1 also including

generating the size assortment recommendation based also on operational constraints associated with the purchase or allocation decisions.

3. The method of claim 1 in which the historical information about sales is stored separately for each style offered at each of multiple commerce locations.

4. The method of claim 1 in which the historical information about sales is stored separately for each style offered in multiple time periods.

5. The method of claim 1 in which the generation of a size assortment recommendation includes determining size profiles, each size profile representing a distribution of percentages of sales by size within one of the styles.

6. The method of claim 2 in which the operational constraints include a limitation on the granularity of purchase and allocation decisions that can be implemented.

7. The method of claim 6 in which the generation of a size assortment recommendation includes

determining size profiles, each size profile representing a distribution of percentages of sales by size within one of the styles, and

optimizing the number of size profiles by assigning them to groups.

8. The method of claim 7 also including substituting a representative size profile for all of the size profiles of each of the groups.

9. The method of claim 7 in which the optimizing is based on the objectives.

10. The method of claim 9 in which the optimizing is based on a genetic algorithm.

11. The method of claim 7 in which the optimizing yields a determination of the optimal number of groups.

12. The method of claim 1 in which the size assortment recommendation is an optimal assortment, the method also including the step of performing simulations on the optimal assortment to derive a risk profile of outcomes.

13. The method of claim 5 also including

generating error bound information from size profiles for multiple styles and multiple time periods.

14. The method of claim 13 also including

using the error bound information to optimize size profiles based on the objectives.

15. The method of claim 1, comprising deriving a risk profile for a set of size profiles determined in the identifying step.