Managing item options for distribution efficiency

One or more non-transitory computer-readable media maintaining instructions that, when executed by one or more processors, program the one or more processors to:

• receive, from a plurality of respective buyer applications executing on a plurality of respective buyer devices associated with a plurality of respective buyers, buyer preferences of the respective buyers for respective menu items for a service that enables the respective buyers to receive the respective menu items for scheduled time periods;
• receive, from a plurality of respective merchant applications executing on a plurality of respective merchant devices associated with a plurality of respective merchants, item information about offered menu items of a plurality of offered menu items offered by the respective merchants, the plurality of offered menu items including the respective menu items;
• determine, based at least in part on comparing the buyer preferences with the item information about the plurality of offered menu items, a selected first menu item from a first merchant of the plurality of respective merchants that matches the buyer preferences of a threshold quantity of the respective buyers, wherein the threshold quantity of the respective buyers is a first set of multiple buyers of the plurality of buyers;
• based at least on determining the selected first menu item from the first merchant, determine a second set of multiple buyers of the plurality of respective buyers, exclusive of the first set, and having buyer preferences that match a second menu item from the first merchant for selecting the second menu item from the first merchant;
• send, to the respective buyer devices of the respective buyers of the first set whose buyer preferences match the selected first menu item from the first merchant, first scheduled item information that causes the respective buyer devices corresponding to the first set to present respective first user interfaces indicating that the selected first menu item is scheduled to be provided for an upcoming time period;
• send, to the respective buyer devices of the respective buyers of the second set whose buyer preferences match the selected second menu item from the first merchant, second scheduled item information that causes the respective buyer devices corresponding to the second set to present respective second user interfaces indicating that the selected second menu item is scheduled to be provided for the upcoming time period;
• use a trained machine learning model to determine a predicted quantity of cancelations predicted to be made by the respective buyers of the first set for the selected first menu item and the respective buyers of the second set for the selected second menu item for the upcoming time period, wherein the trained machine learning model is trained using a plurality of pieces of information of past order information for a plurality of a past orders, the plurality of pieces of information including buyer historic information related to past buyer-canceled orders;
• and
• based at least on a number of the respective buyers in the first set, a number of the respective buyers in the second set, and the predicted quantity of cancelations predicted to be made by the respective buyers of the first set for the selected first menu item and the respective buyers of the second set for the selected second menu item for the upcoming time period, request a quantity of orders of the selected first menu item and the selected second menu item from the first merchant for the upcoming time period.

A system comprising:

• one or more processors configured by executable instructions to perform operations comprising:
• receiving, by a computing device, from a plurality of respective buyer applications executing on a plurality of respective buyer devices associated with a plurality of respective buyers, buyer preferences of the respective buyers for respective menu items for a service that enables the respective buyers to receive the respective menu items for scheduled time periods;
• receiving, from a plurality of respective merchant applications executing on a plurality of respective merchant devices associated with a plurality of respective merchants, item information about offered menu items of a plurality of offered menu items offered by the respective merchants, the plurality of offered menu items including the respective menu items;
• determining, based at least in part on comparing the buyer preferences with the item information about the plurality of offered menu items, a selected first menu item from a first merchant of the plurality of respective merchants that matches the buyer preferences of a threshold quantity of the respective buyers, wherein the threshold quantity of the respective buyers is a first set of multiple buyers of the plurality of buyers;
• based at least on determining the selected first menu item from the first merchant, determining a second set of multiple buyers of the plurality of respective buyers, exclusive of the first set, and having buyer preferences that match a second menu item from the first merchant for selecting the second menu item from the first merchant;
• sending, to the respective buyer devices of the respective buyers of the first set whose buyer preferences match the selected first menu item from the first merchant, first scheduled item information that causes the respective buyer devices corresponding to the first set to present respective first user interfaces indicating that the selected first menu item is scheduled to be provided for an upcoming time period sending, to the respective buyer devices of the respective buyers of the second set whose buyer preferences match the selected second menu item from the first merchant, second scheduled item information that causes the respective buyer devices corresponding to the second set to present respective second user interfaces indicating that the selected second menu item is scheduled to be provided for the upcoming time period;
• using a trained machine learning model to determine a predicted quantity of cancelations predicted to be made by the respective buyers of the first set for the selected first menu item and the respective buyers of the second set for the selected second menu item for the upcoming time period, wherein the trained machine learning model is trained using a plurality of pieces of information of past order information for a plurality of a past orders, the plurality of pieces of information including buyer historic information related to past buyer-canceled orders;
• and
• based at least on a number of the respective buyers in the first set, a number of the respective buyers in the second set, and the predicted quantity of cancelations predicted to be made by the respective buyers of the first set for the selected first menu item and the respective buyers of the second set for the selected second menu item for the upcoming time period, requesting a quantity of orders of the selected first menu item and the selected second menu item from the first merchant for the upcoming time period.