Agent using detailed predictive model

• US 7,644,863 B2
• Filed: 10/24/2002
• Issued: 01/12/2010
• Est. Priority Date: 11/14/2001
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP AG (SAP SE)
• Inventors: Renz, Alexander, Chen, Ye
• Primary Examiner(s): RUDY, ANDREW J
• Application Number:
  US10/281,074
• Publication Number:
  US 20030126103A1
• Time in Patent Office: 2,637 Days
• Field of Search: 705/28, 705/7, 705/10, 705/22, 705/29, 235/375, 235/383, 235/385, 700/106
• US Class Current: 235/385
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/08 Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
• CPC Class Codes: G06Q 10/087 Inventory or stock management, e.g. order filling, procurement or balancing against orders

Abstract

Methods and apparatus, including computer program products, implementing and using techniques for monitoring and predicting various data related to an inventory. A data processing apparatus interacts with one or more business computers to obtain one of replenishment and consumption activity data related to an inventory. A model for variations in the inventory is created. The model is used to determine estimates of variation for one of currently planned replenishment activities and currently planned consumption activities. The estimates of variation are used to predict future inventory levels. A system including one or more agent computers is also described.

Claims

1. A computer-readable storage medium having stored therein instructions, which when executed cause a machine to perform a set of operations comprising:

interacting with one or more business computers to obtain one of replenishment and consumption activity data related to an inventory;

creating a model for variations in the inventory;

using the model to determine estimates of variation for one of currently planned replenishment activities and currently planned consumption activities; and

using the estimates of variation to predict future inventory levels.

2. The computer-readable storage medium of claim 1, wherein creating the model comprises applying machine learning to create the model.

3. The computer-readable storage medium of claim 1, wherein using the model for variations in inventory comprises using an algorithm including a decision tree.

4. The computer-readable storage medium of claim 3, wherein using the algorithm including a decision tree comprises using a classification and regression tree.

5. The computer-readable storage medium of claim 4, wherein the classification and regression tree is constructed using a growing stage and a pruning stage to provide one or more splitting points.

6. The computer-readable storage medium of claim 5, wherein the growing stage comprises instructions operable to recursively split data using criterion for maximizing the information resulting from each of the one or more splitting points.

7. The computer-readable storage medium of claim 5, wherein the pruning stage comprises instructions operable to remove splitting points.

8. The computer-readable storage medium product of claim 5, wherein the criterion for determining splitting points comprises applying analysis of variance techniques.

9. The computer-readable storage medium of claim 4, wherein a size of the decision tree is selected based on a process comprising instructions operable to:

determine a data size;

split data into multiple substantially equal-sized groups;

construct multiple test decision trees, each test decision tree including all of the groups but one such that test decision trees are created in which each group is excluded once;

prune each test decision tree at one or more degrees;

for each degree, test each of the test decision trees by evaluating each test decision tree with the data excluded from creating that test decision tree;

select a degree of pruning based on a maximization of the average performance for the excluded data; and

prune the decision tree to the selected degree of pruning.

10. The computer-readable storage medium of claim 1, wherein the model comprises one or more of data relating to:

an update number specifying how many times an activity has been updated, a creation date specifying when an activity was first planned, a current date specifying a date when the information in a record was extracted from the activity, an expected date specifying the date at which the activity is expected to be fulfilled, an expected quantity specifying an amount of material required by the activity, source of an order specifying an entity making the order, demand at a source in a temporal window prior to the expected date, actual delivered quantity, actual delivery date, time spread for use with activities that are completed in more than one piece, and activity cancelled information relating to cancellation of an activity.

11. The computer-readable storage medium of claim 10, further comprising instructions that create a model to predict variability on actual delivery quantity.

12. The computer-readable storage medium of claim 10, further comprising instructions that create a model for predicting variability of an actual delivery date.

13. The computer-readable storage medium of claim 1, further comprising instructions that create a model for predicting an order cancellation probability.

14. The computer-readable storage medium of claim 10, further comprising instructions:

determine if a qualitative shift in behavior has occurred; and

if a qualitative shift in behavior has occurred

discard data prior to the qualitative shift in behavior; and

create a decision tree using data generated after the qualitative shift in behavior.

15. The computer-readable storage medium of claim 1, further comprising instructions to process the data based on the quantity of data.

16. The computer-readable storage medium of claim 15, wherein processing the data includes sampling the data if the quantity of the data exceeds a threshold value.

17. The computer-readable storage medium of claim 16, wherein processing the data includes using additional data if the quantity of the data is below a threshold value.

18. The computer-readable storage medium of claim 1, further comprising instructions to display predicted future inventory levels.

19. The computer-readable storage medium of claim 18, further comprising instructions to display the predicted future inventory levels as a bicumulative display.

20. The computer-readable storage medium of claim 18, further comprising instructions to display the predicted future inventory levels as a relative density display.

21. The computer-readable storage medium of claim 1, further comprising instructions for an early warning agent to:

receive predicted future inventory levels; and

use the predicted future inventory levels to determine whether to order additional stock for the inventory.

22. The computer-readable storage medium of claim 1, further comprising instructions to generate the data using a radio-frequency identification device.

23. The computer-readable storage medium of claim 1, further comprising instructions to generate the data based upon the sale of stock in an inventory.

24. The computer-readable storage medium of claim 1, further comprising instructions to generate the data based upon the transfer of stock in an inventory.

25. The computer-readable storage medium of claim 1, wherein the computer-readable storage medium is implemented as one or more agents in a supply chain management system.

26. A system comprising:

a business computer storing one of replenishment and consumption activity data related to an inventory;

an agent computer coupled to the business computer over a network, the agent computer including an agent to interact with the business computer to obtain one of replenishment and consumption activity data related to the inventory, to create a model for variations in the inventory, to sue the model to determine estimates of variation for one of currently planned replenishment activities and currently planned consumption activities and to use the estimates of variation to predict future inventory levels; and

an agent framework to provide framework services to the agent.

27. The system of claim 26, wherein the agent comprises:

an activity risk agent to apply machine learning to create the model.

28. The system of claim 26, wherein using the model for variations in inventory comprises using an algorithm including a decision tree.

29. The system of claim 28, wherein using the algorithm including the decision tree comprises using a classification and regression tree.

30. The system of claim 29, wherein the classification and regression tree is constructed by the agent computers using a growing stage and a pruning stage to provide one or more splitting points.

31. The system of claim 30, wherein the growing stage recursively splits data using criterion for maximizing the information resulting from each of the one or more splitting points.

32. The system of claim 30, wherein the pruning stage removes splitting points.

33. The system of claim 30, wherein the criterion for determining splitting points comprises applying analysis of variance techniques.

34. The system of claim 29, wherein a size of the decision tree is selected based on a process comprising instructions that:

determine a data size;

split data into multiple substantially equal-sized groups;

construct multiple test decision trees, each test decision tree including all of the groups but one such that test decision trees are created in which each group is excluded once;

prune each test decision tree at one ore more degrees;

for each degree, test each of the test decision trees by evaluating each test decision tree with the data excluded from creating that test decision tree;

select a degree of pruning based on a maximization of the average performance for the excluded data; and

prune the decision tree to the selected degree of pruning.

35. The system of claim 26, wherein the model comprises one or more of data relating to:

an update number specifying how many times an activity has been updated, a creation date specifying when an activity was first planned, a current date specifying a date when the information in a record was extracted from the activity, an expected date specifying the date at which the activity is expected to be fulfilled, an expected quantity specifying an amount of material required by the activity, source of an order specifying an entity making the order, demand at a source in a temporal window prior to the expected date, actual delivered quantity, actual delivery date, time spread for use with activities that are completed in more than one piece, and activity cancelled information relating to cancellation of an activity.

36. The system of claim 35, wherein the agent creates another model to predict variability on actual delivery quantity.

37. The system of claim 35, wherein the agent creates another model for predicting variability of an actual delivery date.

38. The system of claim 35, wherein the agent creates another model for predicting an order cancellation probability.

39. The system of claim 26, wherein the agent determines if a qualitative shift in behavior has occurred, and if a qualitative shift in behavior has occurred;

discards data prior to the qualitative shift in behavior and creates a decision tree using data generated after the qualitative shift in behavior.

40. The system of claim 26, wherein the agent processes the data based on the quantity of data.

41. The system of claim 40, wherein processing the data includes sampling the data if the quantity of the data exceeds a threshold value.

42. The system of claim 41, wherein processing the data includes using additional data if the quantity of the data is below a threshold value.

43. The system of claim 26, wherein the agent displays predicted future inventory levels.

44. The system of claim 43, wherein the agent displays the predicted future inventory levels as a bicumulative display.

45. The system of claim 43, wherein the agent displays the predicted future inventory levels as a relative density display.

46. The system of claim 26, further comprising:

an early warning agent computer coupled to the agent computer over the network that receives predicted future inventory levels and uses the predicted future inventory levels to determine whether to order additional stock for the inventory.

47. The system of claim 26, further comprising a radio-frequency identification device that generate data.

48. The system of claim 26, wherein the data is generated upon the sale of stock in an inventory.

49. The system of claim 26, wherein the data is generated upon the transfer of stock in an inventory.

50. The system of claim 26, wherein the business and agent computers are implemented in a supply chain management system.