SYSTEMS AND METHODS FOR DYNAMIC DELIVERY ROUTING

1 . A system, comprising:

an interface configured to receive a request for a package;

a database configured to store shipping data, wherein the database is updated in real-time during package transit; and

a processing system configured to:

retrieve, from the database, shipping data for the package;

based on the shipping data, determine an initial shipping route using a set of machine learning models;

retrieve, from the database, updated shipping data for the package during package transit;

select updated training data based on the updated shipping data;

retrain the set of machine learning models using the updated training data;

determine an updated shipping route using the retrained set of machine learning models; and

automatically reroute the package according to the updated shipping route.

2 . The system of claim 1 , wherein the updated shipping data comprises at least one of: updated scan data for the package, updated weather data, or updated delivery parameters.

3 . The system of claim 1 , wherein each machine learning model is associated with a shipping carrier in a set of shipping carriers, wherein selecting the updated training data comprises, for each machine learning model, selecting historical shipping data corresponding to the updated shipping data and the shipping carrier associated with the machine learning model, wherein determining the updated shipping route comprises, for each shipping carrier in the set of shipping carriers, predicting a delivery time using the retrained machine learning model associated with the shipping carrier, and selecting a shipping carrier based on the predicted delivery times and a set of delivery parameters associated with the package.

4 . The system of claim 3 , wherein the set of shipping carriers comprises a set of last-mile shipping carriers.

5 . The system of claim 1 , wherein the updated shipping data comprises a set of updated delivery parameters associated with the package, wherein each updated delivery parameter is associated with a machine learning model in the set of machine learning models, wherein selecting updated training data comprises, for each updated delivery parameter, selecting historical shipping data corresponding to the updated delivery parameter.

6 . The system of claim 5 , wherein the set of updated delivery parameters comprises a set of addresses.

7 . The system of claim 5 , wherein determining the updated shipping route comprises:

predicting a delivery time for each updated delivery parameter in the set of updated delivery parameters; and

determining the updated shipping route based on the predicted delivery times and a set of delivery conditions associated with the set of updated delivery parameters.

8 . The system of claim 1 , wherein the set of machine learning models are retrained in response to retrieving the updated shipping data.

9 . The system of claim 1 , wherein the updated shipping route comprises at least one of: delivery address, delivery time, or shipping carrier.

10 . The system of claim 1 , wherein the initial shipping route comprises a partial shipping route, wherein the updated shipping route comprises a complete shipping route.

11 . A system, comprising:

an interface configured to receive a request for a package and a recipient identifier; and

a processing system configured to:

retrieve a set of delivery preferences associated with the recipient identifier;

based on the set of delivery preferences, determine an initial shipping route using a set of machine learning models;

receive updated shipping data associated with the package during package transit;

based on the updated shipping data, determine an updated shipping route using the set of machine learning models; and

automatically reroute the package according to the updated shipping route.

12 . The system of claim 11 , wherein a delivery vehicle is automatically controlled to transport the package according to the updated shipping route.

13 . The system of claim 11 , wherein the updated shipping data comprises at least one of: updated weather data, updated scan data for the package, or updated delivery preferences associated with the recipient identifier.

14 . The system of claim 11 , wherein the set of delivery preferences comprise a set of addresses and a set of delivery conditions for each address.

15 . The system of claim 11 , wherein determining the updated shipping route using the set of machine learning models comprises selecting updated training data based on the updated shipping data and retraining the set of machine learning models using the updated training data.

16 . The system of claim 11 , wherein determining the updated shipping route comprises:

for each machine learning model in the set of machine learning models, predicting a delivery time based on the updated shipping data; and

determining the updated shipping route based on the predicted delivery times and the set of delivery preferences.

17 . The system of claim 16 , wherein the updated shipping route is further determined based on a confidence parameter for each predicted delivery time.

18 . The system of claim 1 , wherein determining the initial shipping route comprises determining a first subset of delivery parameters from the set of delivery parameters, wherein determining the updated shipping route comprises determining a second subset of delivery parameters from the set of delivery parameters, wherein automatically rerouting the package according to the updated shipping route comprises providing the second subset of delivery parameters to a shipping carrier.

19 . The system of claim 18 , wherein the first subset of delivery parameters comprises a first address, wherein the second subset of delivery parameters comprises a second address.

20 . The system of claim 1 , wherein the updated shipping route comprises at least one of: delivery address, delivery time, or shipping carrier.