US 11,734,749 B2
Online shopping system and method for selecting a warehouse for inventory based on predicted availability and predicted replacement machine learning models
Shishir Kumar Prasad, Fremont, CA (US); Sharath Rao Karikurve, Berkeley, CA (US); and Diego Goyret, Los Gatos, CA (US)
Assigned to Maplebear Inc., San Francisco, CA (US)
Filed by Maplebear, Inc., San Francisco, CA (US)
Filed on Apr. 14, 2021, as Appl. No. 17/230,816.
Prior Publication US 2022/0335505 A1, Oct. 20, 2022
Int. Cl. G06Q 30/0601 (2023.01); G06Q 10/087 (2023.01); G06Q 30/0204 (2023.01); G06N 20/00 (2019.01); G06Q 30/0201 (2023.01); G06N 5/04 (2023.01)
CPC G06Q 30/0639 (2013.01) [G06N 5/04 (2013.01); G06N 20/00 (2019.01); G06Q 10/087 (2013.01); G06Q 30/0201 (2013.01); G06Q 30/0205 (2013.01); G06Q 30/0619 (2013.01); G06Q 30/0629 (2013.01); G06Q 30/0633 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising, at an online concierge system comprising at least one processor and memory:
receiving a selection of a warehouse at the online concierge system from a user over a network, the warehouse having a plurality of locations offering items and the selection identifying a physical location;
identifying, by the online concierge system, locations of the warehouse within a threshold distance of the location identified by the selection;
retrieving, by the online concierge system, orders for one or more items the online concierge system previously received from the user;
selecting, by the online concierge system, a set of items from items included in the orders previously received by the online concierge system from the user;
for each identified location of the warehouse:
determining, by the online concierge system, a predicted availability of each item of the set by the online concierge system applying a machine-learned item availability model to characteristics of an item of the set and to characteristics of an identified location of the warehouse, wherein the machine-learned item availability model is trained by:
accessing, by the online concierge system, a dataset including a time and a warehouse associated with a previous order and item characteristics; and
updating, by the online concierge system, the machine-learned item availability model by:
identifying an item-warehouse pair;
determining a confidence score associated with a probability that the item is available at the warehouse;
collecting new information about the item responsive to the confidence score being lower than a threshold;
updating the dataset with the collected new information; and
retraining the machine-learned item availability model with the updated dataset;
generating, by the online concierge system, an availability value for the identified location of the warehouse from the predicted availabilities of each item of the set for the identified location of the warehouse; and
determining, by the online concierge system, a selected identified location of the warehouse from the availability values for each of the identified locations of the warehouse; and
retrieving, by the online concierge system, an inventory of items available from the selected identified location of the warehouse.