Supply Chain Feature Importance Using LASSO-Pruned Bayesian Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing supply chain machine learning systems generate dense, convoluted probabilistic graphical models that require significant time and resources to interpret and do not effectively assess the relative importance of features and feature classes, hindering efficient decision-making.
Innovation Solution
A probabilistic graphical model feature importance system using LASSO regression analysis to prune and bin features, generating simplified networks that highlight important features and classes, enabling rapid inference and planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a Bayesian network PGM is used to model the flow of all variables through a supply chain network, then the model comprehensively represents all variables and their conditional dependencies, but the model becomes extremely dense and convoluted, requiring significant time and resources to interpret and derive meaningful inferences
Solution Approach 1:
The patent extracts only the important features and feature classes from the complete set of supply chain variables using LASSO regression analysis. By removing irrelevant or less important variables from the PGM, the model maintains comprehensive representation of critical factors while reducing overall density and complexity, making it more interpretable without sacrificing essential variable relationships
2Reliability
If a Bayesian network PGM is used to model the flow of all variables through a supply chain network, then the model represents all conditional dependencies, but the model does not assess and emphasize the relative importance of some features and feature classes relative to other features and feature classes
Solution Approach 1:
The patent applies LASSO regression analysis to assign different importance weights to different features and feature classes within the PGM. This creates local quality variations where critical features receive higher emphasis through larger coefficients, while less important features are downweighted or eliminated. The model thus maintains conditional dependency representation while simultaneously differentiating feature importance levels across the network
3Ease of operation
If LASSO regression analysis is used to prune and bin features, then the model complexity is reduced and interpretation is simplified, but the process requires additional computational steps for feature selection and regularization
Solution Approach 1:
The patent performs LASSO regression analysis as a preliminary step before constructing the final PGM. By conducting feature selection and regularization in advance, the system identifies and retains only the most important features, which are then used to build a simpler, more interpretable model. This preliminary action reduces the complexity burden from the main modeling process, making the overall system easier to operate despite the additional initial computational step
Data Source
AI summary
Embodiments of the following disclosure provide a feature importance system and method to identify features relevant to determining whether a target variable will achieve a particular value. One example method includes generating a probabilistic graphical model to represent the performance of one or more entities in a supply chain and selecting or more target variables. The method further includes collating a list of features pertaining to the one or more selected target variables, pruning at least one of the one or more features from the list and generating one or more bins in which to distribute the one or more features in the list. The method further includes modeling a network graph incorporating the one or more features in the list and bins and determining one or more inferences pertaining to the one or more supply chain entity target variables.