Granted Patent B2

US 12,229,791 · App. 18/371,215 · Granted Feb 18, 2025

Inventor: Sawyer Bateman (Lehi, UT)

Assignee: Simpler Postage, Inc.

G06Q30/0202G06Q10/067G06Q10/08G06Q10/0833G06Q10/08345G06Q10/0838G06Q40/08

Abstract

Claims (45)

1. A method, comprising:

determining a set of models, wherein each model in the set of models is trained using a different subset of historical delivery data;

retrieving parcel data for a parcel;

determining parcel features based on the parcel data;

selecting a model from the set of models based on the parcel features; and

using the selected model, predicting a delivery estimate for the parcel.

2. The method of claim 1 , further comprising transforming the historical delivery data into delivery features, wherein each different subset of historical delivery data corresponds to a subset of delivery features selected using a feature selection rule.

3. The method of claim 2 , wherein, for each model, the selected subset of delivery features corresponds to historical parcels satisfying the feature selection rule, wherein the feature selection rule is based on at least one of: origin address, destination address, ship date, or carrier information.

4. The method of claim 2 , wherein the subset of delivery features used to train the selected model matches at least a portion of the parcel features.

5. The method of claim 1 , further comprising:

retrieving second parcel data for a second parcel;

determining second parcel features based on the second parcel data;

selecting a second model from the set of models based on the second parcel features; and

using the selected second model, predicting a delivery estimate for the second parcel.

6. The method of claim 1 , wherein the parcel features comprise at least one of: origin address, destination address, date, or carrier information.

7. The method of claim 1 , wherein the delivery estimate is predicted based on a subset of the parcel features.

8. The method of claim 1 , wherein the model is further selected from the plurality of models based on an accuracy criterion.

9. The method of claim 1 , wherein each model in the set of models comprises a machine learning model.

10. The method of claim 1 , further comprising:

receiving a shipment request for the parcel;

providing a response to a user endpoint, the response comprising the delivery estimate.

11. A system, comprising:

a database storing a plurality of models, wherein each model in the plurality of models is trained using a subset of historical delivery data; and

a processing system configured to:

receive a shipment request for a parcel;

retrieve parcel data for the parcel;

select a model from the plurality of models based on the parcel data and the subset of historical delivery data used to train the selected model;

using the selected model, predict a delivery estimate for the parcel; and

provide a response to a user endpoint, the response comprising the delivery estimate for the parcel.

12. The system of claim 11 , wherein the model is further selected based on contextual data.

13. The system of claim 12 , wherein the contextual data comprises at least one of: weather, traffic patterns, or theft statistics.

14. The system of claim 11 , wherein the processing system is further configured to:

detect a trigger event during a delivery of the parcel; and

update the parcel data based on the trigger event, wherein the updated parcel data is provided to the user endpoint.

15. The system of claim 14 , wherein the trigger event comprises at least one of a delivery disruption or a user query.

16. The system of claim 11 , wherein the processing system is further configured to:

receive a shipment request for a second parcel;

retrieve second parcel data for the second parcel;

select a second model from the plurality of models based on the second parcel data and the subset of historical delivery data used to train the second model;

using the selected second model, predict a delivery estimate for the second parcel; and

provide a second response to the user endpoint, the response comprising the delivery estimate for the second parcel.

17. The system of claim 11 , wherein the processing system is further configured to transform the historical delivery data into delivery features, wherein each different subset of historical delivery data corresponds to a subset of delivery features selected using a feature selection rule, wherein the feature selection rule is based on at least one of: origin address, destination address, ship date, or carrier information.

18. The system of claim 11 , wherein the model is further selected from the plurality of models based on an accuracy criterion.

19. The system of claim 11 , wherein selecting the model from the plurality of models comprises matching at least a portion of the parcel data to the subset of historical delivery data used to train the selected model.

20. The system of claim 11 , wherein the database is configured to store each of the plurality of models in association with an identifier associated with the subset of historical delivery data used to train the model, wherein selecting the model comprises retrieving the model based on the identifier.

Assignments (2)

Continuity (7)

References Cited (116)