Mutual Information Graphs for Supply Chain Resolution Actions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing supply chain modeling systems fail to separate supply chain features and attributes, lack the ability to recommend options based on these features, and do not provide graphical displays, hindering user understanding and translation of modeling data.
Innovation Solution
A probabilistic graphical model (PGM) resolution system that generates PGMs, identifies probability relationships and mutual information between supply chain features and attributes, and provides graphical visualizations to recommend resolution actions that improve key performance indicators (KPIs) and service level agreements (SLAs) with minimal changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If supply chain modeling systems use comprehensive modeling data to represent all variables and conditional dependencies, then the completeness of the model is improved, but the complexity of the system increases and the ability to separate and analyze specific features is lost
Solution Approach 1:
The patent segments the comprehensive supply chain modeling data into distinct features and attributes that can be separately analyzed. The system divides the complex model into manageable components, allowing users to focus on specific supply chain aspects without being overwhelmed by the entire system's complexity.
Solution Approach 2:
The patent extracts specific features and attributes from the comprehensive modeling data using feature extraction techniques. This allows the system to isolate and analyze individual supply chain elements while maintaining the ability to reconstruct the complete picture when needed.
2Loss of information
If existing supply chain modeling systems process and store all modeling data without separation, then data completeness is maintained, but the ability to recommend options based on specific features is lost
Solution Approach 1:
The patent segments modeling data into distinct features and attributes, enabling the system to generate targeted recommendations for specific supply chain challenges while maintaining access to the complete data set for comprehensive analysis.
Solution Approach 2:
The patent introduces feature extraction and probabilistic graphical models as intermediary layers between the raw modeling data and the recommendation engine. These intermediaries process the data to identify meaningful patterns and relationships, enabling intelligent recommendations without losing the underlying data completeness.
3Quantity of substance
If existing supply chain modeling systems store comprehensive modeling data without graphical display capabilities, then data retention is improved, but user understanding and translation of the data is hindered
Solution Approach 1:
The patent creates graphical copies and visual representations of the supply chain modeling data. These visualizations include network diagrams, flow charts, and other graphical displays that mirror the underlying data structure, allowing users to understand complex relationships without sacrificing data completeness.
Solution Approach 2:
The patent transforms the tabular or structured modeling data into graphical dimensions, adding a visual layer that makes the data more accessible and understandable while preserving the original data for analytical purposes.
Data Source
AI summary
A system and method are disclosed for training a probabilistic graphical model based on historical attributes of a supply chain to represent supply chain performance, selecting supply chain entity target variables, collating with the use of machine learning models, a list of features and classes pertaining to selected supply chain entity target variables, calculating first and second level features associated with the list of features and classes, generating supply chain predictions based on the trained probabilistic graphical model, where the supply chain predictions are based on test data, comparing the supply chain predictions to desired supply chain outputs to determine a delta distance, and generating resolution actions, to decrease the delta distance between the supply chain output predictions and the desired supply chain outputs.