System and method for generating shipment forecasts for materials handling facilities

• US 8,407,151 B1
• Filed: 09/24/2010
• Issued: 03/26/2013
• Est. Priority Date: 09/24/2010
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Agarwal, Richa, Wan, Lin, Lai, Jin, Bhaskaran, Michael Mahesh, Helleboid, Thomas Yves Paul, Mishra, Shyam, Patrick, Simon M.
• Primary Examiner(s): Chen, George
• Application Number:
  US12/890,344
• Time in Patent Office: 914 Days
• Field of Search: None
• US Class Current: 705/330
• 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 10/08 Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
• CPC Class Codes: G06Q 10/083 Shipping

Abstract

Embodiments may include a forecasting component that, for each of different service levels offered to customers, generates a projection of the quantity of shipments to be shipped according to that service level during a time period. Based on an aggregate quantity of shipments projected to be shipped, projected quantities of shipments for the multiple service levels may be modified. The forecasting component may, for each service level, receive information specifying a distribution of different shipment methods that are projected to be utilized to ship shipments of a particular priority designation to meet requirements of that service level. Based on the distribution for each service level and the modified projected quantity of shipments for each service level, a forecast specifying quantities of shipments that are to be shipped during the time period according to each of the different shipment methods may be generated.

Claims

1. A system, comprising:

a memory; and

one or more processors coupled to the memory, wherein the memory comprises program instructions executable by the one or more processors to implement a forecasting component configured to;

for each of multiple different shipment service levels offered to customers, generate a projection of the quantity of shipments that will be shipped from a materials handling facility according to that shipment service level during a particular time period, wherein a different respective projection is generated for each of the multiple different shipment service levels;

based on an aggregate quantity of shipments projected to be shipped from the materials handling facility during the particular time period, modify the projected quantities of shipments for the multiple shipment service levels to provide a different respective modified projected quantity of shipments for each shipment service level;

for each shipment service level, receive information specifying a distribution of different shipment methods that are projected to be utilized during the particular time period in order to meet requirements of that shipment service level, wherein a different respective distribution is received for each different respective service level; and

based on the respective distribution for each different respective shipment service level and the respective modified projected quantity of shipments for each shipment service level, generate a forecast, wherein said forecast specifies respective quantities of shipments that are to be shipped during the particular time period according to each of the different shipment methods.

2. The system of claim 1, wherein one or more of the multiple different shipment service levels include one or more of:

a one-day shipment service level, a two-day shipment service level, or an economy shipment service level.

3. The system of claim 1, wherein to generate a given projection of the quantity of shipments that will be shipped from the materials handling facility according to a given shipment service level, the forecasting component is configured to:

utilize a time series forecasting model to generate the given projection from historical data.

4. The system of claim 1, wherein the forecasting component is configured to generate said aggregate quantity of shipments projected to be shipped from the materials handling facility, wherein to generate said aggregate quantity of shipments, the forecasting component is configured to:

receive information specifying a historical average of a quantity of units per shipment processed by the materials handling facility;

receive information specifying an aggregate quantity of units projected to be processed by the materials handling facility during the particular time period; and

divide said aggregate quantity of units by the historical average of the quantity of units per shipment to generate said aggregate quantity of shipments.

5. The system of claim 1, wherein to modify the projected quantities of shipments for the multiple shipment services levels based on said aggregate quantity of shipments, the forecasting component is configured to:

proportionately modify the projected quantities such that a summation of the projected quantities is equivalent to said aggregate quantity of shipments.

6. The system of claim 1, wherein the distribution of different shipment methods for a given shipment service level indicates, for each different shipment method, a projection of a frequency with which that shipment method will be utilized to ship shipments of the given shipment service level relative to a projection of the frequencies with which other shipment methods will be utilized to ship shipment of the given shipment service level.

7. The system of claim 1, wherein said distribution of different shipment methods that are projected to be utilized comprises a distribution of different shipment methods utilized to ship shipments of a particular priority designation, and wherein said respective forecast comprises a forecast specific to the shipments of the particular priority designation.

8. The system of claim 1, wherein said multiple different shipment service levels are not specific to any particular shipment carrier, wherein at least some of said different shipment methods are specific to a respective shipment carrier.

9. The system of claim 7, wherein the forecasting component is configured to:

for a particular shipment method, determine that said forecast specifies a quantity of shipments that exceeds a threshold capacity for that shipment method; and

generate a second forecast for the shipments of the particular priority designation;

wherein the generation of the second forecast excludes from consideration, the particular shipment method and any shipments assigned to the particular shipment method.

10. A method, comprising:

performing, by one or more computing devices;

for each of multiple different shipment service levels offered to customers, generating a projection of the quantity of shipments that will be shipped from a materials handling facility according to that shipment service level during a particular time period, wherein a different respective projection is generated for each of the multiple different shipment service levels;

based on an aggregate quantity of shipments projected to be shipped from the materials handling facility during the particular time period, modifying the projected quantities of shipments for the multiple shipment service levels to provide a different respective modified projected quantity of shipments for each shipment service level;

for each shipment service level, receiving information specifying a distribution of different shipment methods that are projected to be utilized during the particular time period in order to meet requirements of that shipment service level, wherein a different respective distribution is received for each different respective service level; and

;

based on the respective distribution for each different respective shipment service level and the respective modified projected quantity of shipments for each shipment service level, generating a forecast, wherein said forecast specifies respective quantities of shipments that are to be shipped during the particular time period according to each of the different shipment methods.

11. The method of claim 10, wherein one or more of the multiple different shipment service levels include one or more of:

a one-day shipment service level, a two-day shipment service level, or an economy shipment service level.

12. The method of claim 10, wherein generating a given projection of the quantity of shipments that will be shipped from the materials handling facility according to a given shipment service level comprises:

utilizing a time series forecasting model to generate the given projection from historical data.

13. The method of claim 10, wherein the method comprises generating said aggregate quantity of shipments projected to be shipped from the materials handling facility, wherein said generating comprises:

receiving information specifying a historical average of a quantity of units per shipment processed by the materials handling facility;

receiving information specifying an aggregate quantity of units projected to be processed by the materials handling facility during the particular time period; and

dividing said aggregate quantity of units by the historical average of the quantity of units per shipment to generate said aggregate quantity of shipments.

14. The method of claim 10, wherein modifying the projected quantities of shipments for the multiple shipment services levels based on said aggregate quantity of shipments comprises:

proportionately modifying the projected quantities such that a summation of the projected quantities is equivalent to said aggregate quantity of shipments.

15. The method of claim 10, wherein the distribution of different shipment methods for a given shipment service level indicates, for each different shipment method, a projection of a frequency with which that shipment method will be utilized to ship shipments of the given shipment service level relative to a projection of the frequencies with which other shipment methods will be utilized to ship shipment of the given shipment service level.

16. The method of claim 10, wherein said shipments of a particular priority designation comprise shipments that are expected to be delivered by a particular delivery deadline.

17. The method of claim 10, wherein said distribution of different shipment methods that are projected to be utilized comprises a distribution of different shipment methods utilized to ship shipments of a particular priority designation, and wherein said respective forecast comprises a forecast specific to the shipments of the particular priority designation.

18. The method of claim 17, comprising:

for a particular shipment method, determining that said forecast specifies a quantity of shipments that exceeds a threshold capacity for that shipment method; and

generating a second forecast for the shipments of the particular priority designation;

wherein the generation of the second forecast excludes from consideration, the particular shipment method and any shipments assigned to the particular shipment method.

19. A non-transitory computer-readable storage medium, storing program instructions computer-executable on a computer system to implement a forecasting component configured to:

for each of multiple different shipment service levels offered to customers, generate a projection of the quantity of shipments that will be shipped from a materials handling facility according to that shipment service level during a particular time period, wherein a different respective projection is generated for each of the multiple different shipment service levels;

based on an aggregate quantity of shipments projected to be shipped from the materials handling facility during the particular time period, modify the projected quantities of shipments for the multiple shipment service levels to provide a different respective modified projected quantity of shipments for each shipment service level;

for each shipment service level, receive information specifying a distribution of different shipment methods that are projected to be utilized during the particular time period in order to meet requirements of that shipment service level, wherein a different respective distribution is received for each different respective service level

based on the respective distribution for each different respective shipment service level and the respective modified projected quantity of shipments for each shipment service level, generate a forecast, wherein said forecast specifies respective quantities of shipments that are to be shipped during the particular time period according to each of the different shipment methods.

20. The non-transitory computer-readable storage medium of claim 19, wherein one or more of the multiple different shipment service levels include one or more of:

a one-day shipment service level, a two-day shipment service level, or an economy shipment service level.

21. The non-transitory computer-readable storage medium of claim 19, wherein to generate a given projection of the quantity of shipments that will be shipped from the materials handling facility according to a given shipment service level, the forecasting component is configured to:

utilize a time series forecasting model to generate the given projection from historical data.

22. The non-transitory computer-readable storage medium of claim 19, wherein the forecasting component is configured to generate said aggregate quantity of shipments projected to be shipped from the materials handling facility, wherein to generate said aggregate quantity of shipments, the forecasting component is configured to:

receive information specifying a historical average of a quantity of units per shipment processed by the materials handling facility;

receive information specifying an aggregate quantity of units projected to be processed by the materials handling facility during the particular time period; and

divide said aggregate quantity of units by the historical average of the quantity of units per shipment to generate said aggregate quantity of shipments.

23. The non-transitory computer-readable storage medium of claim 19, wherein to modify the projected quantities of shipments for the multiple shipment services levels based on said aggregate quantity of shipments, the forecasting component is configured to:

proportionately modify the projected quantities such that a summation of the projected quantities is equivalent to said aggregate quantity of shipments.

24. The non-transitory computer-readable storage medium of claim 19, wherein the distribution of different shipment methods for a given shipment service level indicates, for each different shipment method, a projection of a frequency with which that shipment method will be utilized to ship shipments of the given shipment service level relative to a projection of the frequencies with which other shipment methods will be utilized to ship shipment of the given shipment service level.

25. The non-transitory computer-readable storage medium of claim 19, wherein said distribution of different shipment methods that are projected to be utilized comprises a distribution of different shipment methods utilized to ship shipments of a particular priority designation, and wherein said respective forecast comprises a forecast specific to the shipments of the particular priority designation.

26. The non-transitory computer-readable storage medium of claim 19, wherein said multiple different shipment service levels are not specific to any particular shipment carrier, wherein at least some of said different shipment methods are specific to a respective shipment carrier.

27. The non-transitory computer-readable storage medium of claim 25, wherein the forecasting component is configured to:

for a particular shipment method, determine that said forecast specifies a quantity of shipments that exceeds a threshold capacity for that shipment method; and

generate a second forecast for the shipments of the particular priority designation;

wherein the generation of the second forecast excludes from consideration, the particular shipment method and any shipments assigned to the particular shipment method.