US 12,190,265 B2
Forecasting system
Timothy Ryan Hodges, Bentonville, AR (US); and Christopher Wade Spencer, Garfield, AR (US)
Assigned to WALMART APOLLO, LLC, Bentonville, AR (US)
Filed by Walmart Apollo, LLC, Bentonville, AR (US)
Filed on Sep. 29, 2023, as Appl. No. 18/375,395.
Application 18/375,395 is a continuation of application No. 16/853,965, filed on Apr. 21, 2020, granted, now 11,810,015.
Claims priority of provisional application 62/836,789, filed on Apr. 22, 2019.
Prior Publication US 2024/0028989 A1, Jan. 25, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06Q 10/0631 (2023.01); G06Q 10/04 (2023.01); G06Q 10/0633 (2023.01); G06Q 10/0637 (2023.01)
CPC G06Q 10/063118 (2013.01) [G06Q 10/04 (2013.01); G06Q 10/0633 (2013.01); G06Q 10/06375 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A forecasting system using machine learning to reiteratively adjust forecast data, the forecasting system comprising:
a storage device storing a database including information associated with a plurality of facilities, the information including for each facility at least one of historical data or labor models indicating an amount of labor employed at the facility; and
a computer system having at least one processor in communication with the database storing a forecasting module comprising a machine learning algorithm that reiteratively adjust the forecast data, the computer system configured to:
execute the machine learning algorithm to generate slope data representing a calculated change in orders or items week over week for each facility of the plurality of facilities using inputs including the at least one of historical or labor model data retrieved from the database;
execute the machine learning algorithm to generate the forecast data for a specified amount of time including physical object predictive data associated with a quantity of physical objects to be disposed in each facility of the plurality of facilities and a quantity of physical objects to be retrieved from or delivered from each facility of the plurality of facilities, using the generated slope data, wherein the physical object predictive data is iteratively updated based on updates to the slope data;
execute the machine learning algorithm to generate a correction factor for comparison with previous correction factors associated with the facility to evaluate accuracy of prior forecast data for the facility;
execute the machine learning algorithm to apply the generated correction factor to the slope data to generate an updated forecast data; and
dynamically adjust, responsive to the forecasting module, schedule of workers scheduled working at the facility based on the updated forecast data.