System, method and computer program product for providing a multidimensional inventory management system

• US 20040093286A1
• Filed: 11/07/2002
• Published: 05/13/2004
• Est. Priority Date: 11/07/2002
• Status: Active Grant

• Current Assignee: Agentsmith, Inc., Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Agentsmith, Inc.
• Inventors: Cooper, Michael B., Gorman, R. Paul
• Granted Patent: US 7,761,346 B2
• Time in Patent Office: Days
• US Class Current: 705/28
• CPC Class Codes: G06Q 10/087 Inventory or stock management, e.g. order filling, procurement or balancing against orders

Abstract

A computer-implemented method of generating a package of advertising inventory items in response to a set of buyer requirements while optimizing both buyer preferences and a station'"'"'s revenue objectives given an expected demand can include: selecting a set of candidate inventory items from an inventory of inventory items available for presenting advertising spots, where the set of candidate inventory items satisfy a set of buyer preferences; and scoring the set of candidate inventory items with an affinity score.

Claims

1. A computer-implemented method of generating a package of advertising inventory items in response to a set of buyer requirements while optimizing both buyer preferences and a station'"'"'s revenue objectives given an expected demand, the method comprising the steps of:

selecting a set of candidate inventory items from an inventory of inventory items available for presenting advertising spots, wherein said set of candidate inventory items satisfy a set of buyer preferences; and

scoring said set of candidate inventory items with an affinity score.

2. The method according to claim 1, further comprising:

acquiring said set of buyer preferences in the form of an avail request.

3. The method according to claim 1, further comprising:

pricing said set of candidate inventory items according to an expected demand.

4. The method according to claim 1, further comprising:

generating a proposal in the form of a package of inventory items selected as the best inventory items from said set of candidate inventory items based on said affinity score;

submitting said proposal to a user for approval; and

communicating an inventory order of said package of inventory items to a billing process when said proposal is approved.

5. The method according to claim 1, further comprising:

reallocating said inventory according to said affinity score, including reallocating both available and committed inventory items to accommodate both a current sale and an expected demand across a plurality of demand segments.

6. The method according to claim 2, further comprising the steps of:

extracting a current allocation status of said candidate inventory items from a transaction database;

extracting station-specific candidate inventory selection criteria from a policy file;

selecting said set of candidate inventory items from said inventory of inventory items that conform to said station specific candidate inventory selection criteria and have an allocation status of available;

generating buyer attribute values from said set of buyer preferences, said avail request and buyer historical data;

generating candidate inventory item attribute values for said set of candidate inventory items;

computing said affinity score of a buyer to each of said set of candidate inventory items, based on said candidate inventory item attribute values and said buyer attribute values; and

ranking said set of candidate inventory items from a highest affinity score to a lowest affinity score.

7. The method according to claim 3, wherein said pricing step further comprises the steps of:

generating a short-term forecast from a sellout curve model using a transaction database including historical sales data;

generating a mid-term forecast from a price distribution using said transaction database;

generating a long-term forecast from a price trend model using said transaction database;

integrating said short-term, mid-term and long-term forecasts into a price distribution forecast;

generating a plurality of pricing tier boundaries from said price distribution forecast;

allocating said set of candidate inventory items into pricing tiers, wherein an allocation efficiency is maximized; and

pricing each of said set of candidate inventory items according to said affinity scores, said pricing tiers, and said expected demand.

8. The method of claim 4, wherein said step of generating a proposal comprises the steps of:

(a) extracting station-specific inventory item distribution criteria from a station policy file;

(b) generating an initial set of candidate inventory item packages;

(c) comparing each candidate inventory item package of said set of candidate inventory item packages with said set of buyer requirements and seller requirements;

(d) generating a package score for said each candidate inventory item package of said initial set of candidate inventory item packages;

(e) comparing each of said package scores with a score threshold; and

(f) if at least one of said package scores exceeds said score threshold, selecting a best package from said initial set of candidate inventory item packages wherein said package score of said best package exceeds said score threshold; and

(g) if none of said package scores exceeds said score threshold,

(i) removing said candidate packages that have low scores relative to said score threshold from said initial set of candidate inventory item packages;

(ii) generating a new set of candidate inventory item packages by combining remaining candidate inventory item packages in a candidate inventory item package pool; and

(iii) repeating steps (c) through (g) until step (f) occurs.