Back to results
US 12,182,744 B2Granted
Systems and methods for supply management
AssigneeE. & J. Gallo Winery
InventorsSuraj Ponnaganti, Ambarish M. Acharya
Life of this patent
Estimating the term…
Abstract
Systems and methods are described for managing a supply of building blocks for one or more products. An example method includes: obtaining demand data for each building block from a plurality of building blocks; obtaining substitution data for the plurality of building blocks; providing the demand data and the substitution data as input to a predictive model; receiving as output from the predictive model a predicted optimal production quantity for each building block; and facilitating a production of the plurality of building blocks according to the predicted optimal production quantities.
Claims21
A computer-implemented method of managing a supply of building blocks for one or more products, the method comprising:
- providing a predictive model on a server side of a client-server system, the client-server system comprising a client side communicatively coupled to the server side via an interface layer, the server side comprising (i) a model layer comprising the predictive model and (ii) an input data layer for providing input data to the predictive model, wherein the predictive model is configured to perform two-stage stochastic programming with recourse in which:
- a first stage comprises determining supply target levels for a plurality of building blocks used to form one or more products, wherein building blocks from the plurality of building blocks are to be produced and then aged before being combined to form the one or more products;
- and
- a second stage comprises determining permissible substitutions among the plurality of building blocks prior to being combined to form the one or more products;
- providing input data via the input data layer to the predictive model in the model layer, the input data comprising:
- demand data for each building block from the plurality of building blocks, the demand data comprising a projected demand distribution for each building block, the projected demand distribution comprising a probability distribution comprising forecasted probabilities for a range of possible demands for the respective building block;
- and
- substitution data for the plurality of building blocks, the substitution data comprising an identification of one or more permissible building block substitutions in which one building block is substituted for another building block;
- in the model layer of the sever side, generating with the predictive model output data comprising a predicted optimal production quantity for each building block from the plurality of building blocks;
- storing at least a portion of the output data in a computer data file readable by a spreadsheet software application;
- and
- facilitating a production of the plurality of building blocks according to the predicted optimal production quantities generated by the predictive model, wherein at least a portion of the input data is obtained from an enterprise software system, and wherein at least one of (i) the predictive model is configured to execute an optimization solver software application, or (ii) the predictive model is configured to utilize open source computer programming libraries.
9 more elements
A system comprising:
- one or more computer systems programmed to perform operations comprising:
- providing a predictive model on a server side of a client-server system, the client-server system comprising a client side communicatively coupled to the server side via an interface layer, the server side comprising (i) a model layer comprising the predictive model and (ii) an input data layer for providing input data to the predictive model, wherein the predictive model is configured to perform two-stage stochastic programming with recourse in which:
- a first stage comprises determining supply target levels for a plurality of building blocks used to form one or more products, wherein building blocks from the plurality of building blocks are to be produced and then aged before being combined to form the one or more products;
- and
- a second stage comprises determining permissible substitutions among the plurality of building blocks prior to being combined to form the one or more products;
- providing input data via the input data layer to the predictive model in the model layer, the input data comprising:
- demand data for each building block from the plurality of building blocks, the demand data comprising a projected demand distribution for each building block, the projected demand distribution comprising a probability distribution comprising forecasted probabilities for a range of possible demands for the respective building block;
- and
- substitution data for the plurality of building blocks, the substitution data comprising an identification of one or more permissible building block substitutions in which one building block is substituted for another building block;
- in the model layer of the sever side, generating with the predictive model output data comprising a predicted optimal production quantity for each building block from the plurality of building blocks;
- storing at least a portion of the output data in a computer data file readable by a spreadsheet software application;
- and
- facilitating a production of the plurality of building blocks according to the predicted optimal production quantities generated by the predictive model, wherein at least a portion of the input data is obtained from an enterprise software system, and wherein at least one of (i) the predictive model is configured to execute an optimization solver software application, or (ii) the predictive model is configured to utilize open source computer programming libraries.
10 more elements
A non-transitory computer-readable medium having instructions stored thereon that, when executed by one or more computer processors, cause the one or more computer processors to perform operations comprising:
- providing a predictive model on a server side of a client-server system, the client-server system comprising a client side communicatively coupled to the server side via an interface layer, the server side comprising (i) a model layer comprising the predictive model and (ii) an input data layer for providing input data to the predictive model, wherein the predictive model is configured to perform two-stage stochastic programming with recourse in which:
- a first stage comprises determining supply target levels for a plurality of building blocks used to form one or more products, wherein building blocks from the plurality of building blocks are to be produced and then aged before being combined to form the one or more products;
- and
- a second stage comprises determining permissible substitutions among the plurality of building blocks prior to being combined to form the one or more products;
- providing input data via the input data layer to the predictive model, the input data comprising:
- demand data for each building block from the plurality of building blocks, the demand data comprising a projected demand distribution for each building block, the projected demand distribution comprising a probability distribution comprising forecasted probabilities for a range of possible demands for the respective building block;
- and
- substitution data for the plurality of building blocks, the substitution data comprising an identification of one or more permissible building block substitutions in which one building block is substituted for another building block;
- in the model layer of the sever side, generating with the predictive model output data comprising a predicted optimal production quantity for each building block from the plurality of building blocks;
- storing at least a portion of the output data in a computer data file readable by a spreadsheet software application;
- and
- facilitating a production of the plurality of building blocks according to the predicted optimal production quantities generated by the predictive model, wherein at least a portion of the input data is obtained from an enterprise software system, and wherein at least one of (i) the predictive model is configured to execute an optimization solver software application, or (ii) the predictive model is configured to utilize open source computer programming libraries.
9 more elements
Dependencies are inferred from the claim text.