| CPC G06N 20/00 (2019.01) [G06N 7/01 (2023.01)] | 12 Claims |
|
1. A method for integrating a plurality of machine-learning models that impact different factor groups for generating a dynamic recommendation to collectively optimize a parameter, the method comprising:
processing, at a demand management server, a specification information of a demand management service, wherein the specification information comprises internal data of the impacted different factor groups and a meta-data model that corresponds to the impacted different factor groups, and operational data that corresponds to the demand management service obtained from at least one client device, wherein the at least one client device specifies the specification information of the group for which the demand management service is created;
training, at the demand management server, a machine learning (ML) model of the plurality of ML models using processed specification information and the operational data to obtain a trained ML model, wherein the plurality of ML models comprise at least one of an anticipation ML model that optimizes a demand parameter or a recommendation ML model that generates a recommendation to the impacted different factor groups for optimizing the parameter, wherein the impacted different factor groups comprise a first factor group, a second factor group and er a third factor group, wherein the first factor group comprises a pricing and promotions factor group, the second factor group comprises a sales and distribution factor group, and the third factor group comprises an inventory placement and allocation factor group;
integrating the trained ML model with the plurality of ML models by setting an output of a first ML model as a feature of a second ML model;
determining a demand of a product of the demand management service using integrated ML models by:
identifying an emerging signal that represents an impact that corresponds to at least one factor;
determining an elasticity of the at least one factor to the demand of the product and an effect of interaction between a plurality of factors associated with the factor group on the demand; and
quantifying probabilistic values that signify prediction of the demand, wherein the probabilistic values comprise the elasticity associated with the at least one factor and a demand forecast adjustment of the product;
updating in real-time, the recommendation ML model generates the recommendation for optimizing the parameter using quantified probabilistic values; and
generating the dynamic recommendation to the at least one client device to update values of the at least one factor of the impacted different factor groups using the recommendation ML model, wherein the dynamic recommendation is the recommendation associated with the impacted different factor groups and collectively optimizes the parameter of the demand management service, wherein the operational data corresponds to the pricing and promotions factor group and comprises data that corresponds to a promotion, at least one sales activity, at least one distribution activity history and at least one plan, wherein the operational data corresponds to the sales and distribution factor group and the operational data comprises the data corresponding to a sale of a product, an order history and a sales plan, wherein the operational data corresponds to the inventory placement and allocation factor group and comprises the data corresponding to an inventory history and an inventory plan.
|