| CPC G06Q 30/0625 (2013.01) [G06F 16/211 (2019.01); G06F 16/9537 (2019.01); G06F 17/18 (2013.01); G06F 40/205 (2020.01); G06Q 10/083 (2013.01); G06Q 30/0267 (2013.01); G06Q 30/0271 (2013.01); G06Q 30/0631 (2013.01); G06Q 30/0633 (2013.01); H04L 51/046 (2013.01); H04W 4/021 (2013.01); H04W 4/14 (2013.01)] | 20 Claims |
|
1. A method, comprising:
processing data received at an adaptive distribution platform implementing a multi-stack application configured to identify a point of time at one or more levels of a programming stack of the multi-stack application, the point of time being evaluated against a date range associated with a zone of time, the point of time being a function of a usage rate based on a rate of consumption;
initiating automatic replenishment of the item by the adaptive distribution platform if the point of time is substantially within the date range;
after the initiating, transmitting a message to a client associated with a first location via a transport level of the programming stack, the message including a characteristic of the item and a control user input configured to adapt a scheduled delivery to replenish the item;
receiving a response to the message at the adaptive distribution platform;
classifying the item based on application of a machine learning application or a deep learning application to automatically predict consumption of the item as a function of the point of time;
determining automatically one or more retrieval options proximate to the client at the first location to select a message format protocol associated with the transport level of the programming stack;
implementing a portion of the multi-stack application to generate automatically location predictive retrieval data using a networked multi-stack computing environment to determine available inventory at a second location proximate to that of a third location associated with a mobile device associated with an account identified in account data, the available inventory associated with other locations associated with one or more inventory management systems;
processing the response to determine whether to automatically adjust the scheduled delivery to replenish the item based on the proximate location that is predicted automatically relative to the device associated with the account identified to form an adjusted scheduled delivery;
generating a confirmation message to the client, the confirmation message comprising other data configured, when parsed, to confirm a scheduled delivery of the item;
transmitting a control signal from the adaptive distribution platform to a system, the control signal being configured to initiate the scheduled delivery of the item;
adapting a predicted distribution event associated with the item and another scheduled delivery by modifying the zone of time based on evaluating the response and the confirmation message to form an adapted predicted distribution event;
using a geo data analysis engine to generate geographic data associated with the mobile device to detect proximity to the other locations associated with the available inventory;
detecting a threshold distance between the second location proximate to that of a third location to determine whether to automatically generate an alternate retrieval option to pick up the item rather than delivery;
preventing delivery of the item to the first location in favor of the alternate retrieval option the other locations, thereby canceling the scheduled delivery;
implementing the machine learning application in the networked multi-stack computing environment to access stored data representing a predictive model to automatically modify model data as function of the adapted predicted distribution event to optimize accuracy of predicting future scheduled deliveries, the machine learning application further configured to adjust the model data to include the second location and at least one of the other locations to predict whether to generate future messages to the mobile device of based on implementation of the alternate retrieval option.
|