US 12,410,013 B1
Automated capacity recommendation engine for shipping networks
Edward King Armah Amartey-Tagoe, Seattle, WA (US); Michael Behrman, Roswell, GA (US); and Vibhor Kaushik, Bothell, WA (US)
Assigned to Amazon Technologies, Inc., Seattle, WA (US)
Filed by Amazon Technologies, Inc., Seattle, WA (US)
Filed on Mar. 1, 2021, as Appl. No. 17/188,514.
Int. Cl. B65G 1/137 (2006.01); G06Q 10/0631 (2023.01); G06Q 10/083 (2024.01)
CPC B65G 1/1373 (2013.01) [G06Q 10/06312 (2013.01); G06Q 10/0838 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A computer-implemented method, comprising:
obtaining, by a computer system comprising one or more processors, first information identifying capacity changes associated with shipping a plurality of packages from one or more entities, each capacity change of the capacity changes associated with a reason code that identifies a reason for the capacity change;
receiving, by the computer system, second information that includes weather information for one or more geographic areas that correspond to a delivery area for the one or more entities;
receiving third information comprising text associated with the reason for each capacity change;
determining, based on the text, the reason codes for the capacity changes;
obtaining, by the computer system, historical information that includes historical capacity changes and associated historical reason codes from the one or more entities;
implementing, by the computer system, a machine learning algorithm based at least in part on the first information, the second information, the reason code for each capacity change, and the historical information, the machine learning algorithm previously trained using prior capacity changes and associated prior reason codes, prior event data, and the historical information to identify discrepancies between an estimated capacity requirement and a predicted capacity requirement for the one or more entities;
generating, by the computer system, a recommendation for an entity of the one or more entities using the machine learning algorithm, the recommendation including a modification of an autonomous mobile robot of a fulfillment center associated with a shipping network of the entity for a future time period to accommodate at least one discrepancy of the identified discrepancies, and at least a portion of the autonomous mobile robot modifiable at an entity computer associated with the entity; and
responsive to the recommendation:
generating, by the computer system, instructions for modifying the autonomous mobile robot based at least in part on the recommendation, the instructions comprising a layout of the plurality of packages stored in the fulfillment center and routes for the autonomous mobile robot to use to move the plurality of packages to correspond to the layout, the layout of the plurality of packages accommodating the at least one discrepancy;
transmitting, by the computer system, the instructions to the entity computer associated with the entity;
modifying at least the portion of the autonomous mobile robot at the entity computer in accordance with the instructions; and
causing the autonomous mobile robot to operate according to the routes.