×

Consumption-driven forecasting using multi-level heterogeneous input data

  • US 10,002,364 B2
  • Filed: 06/25/2014
  • Issued: 06/19/2018
  • Est. Priority Date: 06/25/2014
  • Status: Active Grant
First Claim
Patent Images

1. A method comprising:

  • identifying a computing system having a storage subsystem, the storage subsystem comprising at least one database component having a series of individual records stored on a computer-readable medium;

    receiving sales data as one or more individual records in a first data format, the first data format comprising an item identifier, a first address identifier, and a first date or first date range, wherein the sales data is received from a first plurality of sources, and the first address identifier is associated with the one or more individual records in the first data format, the first address identifier comprising at least one of an IP address, a network identifier, or a uniform resource locator;

    determining an origin of at least one of the one or more individual records in the first data format using at least the first address identifier;

    recording the at least one of the one or more individual records in the first data format in a first entry in a first relational database table using a selected first application programming interface of a first plurality of application programming interfaces, the first application programming interface being selected based on the origin of the at least one of the one or more individual records in the first data format, wherein different application programming interfaces of the first plurality of application programming interfaces are used for processing different formats of the first data format corresponding to the sales data;

    receiving distribution data as one or more individual records in a second data format, the second data format comprising the item identifier, a second address identifier, and a second date or second date range, wherein the distribution data is received from a second plurality of sources, the second address identifier associated with the one or more individual records in the second data format, the second address identifiers comprising at least one of the IP address, the network identifier, or the uniform resource locator, wherein the first data format comprises a plurality of data formats, and the second data format comprises a plurality of data formats;

    determining the origin of at least one of the one or more individual records in the second data format using at least the second address identifier;

    recording the at least one of the one or more individual records in the second data format in a second entry in a second relational database table using a selected second application programming interface of a second plurality of application programming interfaces, the second application programming interface being selected based on the origin of the at least one of the one or more individual records in the second data format, wherein different application programming interfaces of the of the second plurality of application programming interfaces are used for processing different formats of the second data format corresponding to the distribution data, wherein the sales data and the distribution data are in normalized formats to be combined;

    combining at least a portion of the sales data with at least a portion of the distribution data to generate a combined portion of data;

    generating a combined forecast of demand for at least an item using the combined portion of data;

    generating a graphical user interface comprising a selector component and a table structure,the selector component receives a selection of an option of a first set of options, and the table structure that is displayed depends at least in part on the selection received by the selector component, the table structure having a plurality of columns corresponding to each of a plurality of rows, a first subset of the plurality of columns corresponding to user-defined parameters, a second subset of the plurality of columns corresponding to parameters calculated for the selection, and a third subset of the plurality of columns having a parameter calculated based on at least the user-defined parameters in the first subset of the plurality of columns and the parameters calculated for the selection in the second subset of the plurality of columns; and

    generating a replenishment plan for the item based on at least the combined forecast of demand and one or more values displayed in the plurality of columns in a row corresponding to the item, wherein the first set of options comprise different types of safety stock strategy plans.

View all claims
  • 1 Assignment
Timeline View
Assignment View
    ×
    ×