Optimization Model Augmentation for Historical Data Uncertainty
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mathematical optimization models face challenges in efficiently generating accurate solutions due to uncertainties and inaccuracies from historical data, particularly when deriving models from restricted and non-representative decision sets, leading to suboptimal performance in applications like industrial control systems.
Innovation Solution
A system and method that utilize a processor to generate mathematical optimization models from historical data, incorporating a formal quality measure and target threshold to ensure the model quality exceeds predefined standards, addressing uncertainties through iterative improvement and data augmentation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If mathematical optimization models are generated from historical data, then model generation speed is improved, but model accuracy deteriorates due to uncertainties and inaccuracies in historical data
Solution Approach 1:
The patent introduces a formal quality measure as an intermediary between the historical data and the optimization model. This quality measure acts as a mediator that quantifies the reliability of historical data and guides the model generation process, allowing the system to maintain high generation speed while improving accuracy by filtering and weighting data based on its measured quality
Solution Approach 2:
The patent changes the parameter of data quality measurement by introducing a formal quality threshold parameter. By adjusting this parameter, the system can control the balance between using more historical data (faster generation) versus using only high-quality data (higher accuracy), enabling dynamic optimization of the contradiction based on specific application requirements
2Ease of operation
If optimization models are generated from restricted and non-representative decision sets, then ease of operation is improved, but solution quality deteriorates
Solution Approach 1:
The patent implements a feedback mechanism where the formal quality measure provides information back to the model generation process. The quality measurement results feed into the model generation algorithm, allowing it to adjust its behavior based on the quality of available historical data, thereby maintaining ease of operation while improving solution quality through data-driven adjustments
Solution Approach 2:
The patent performs preliminary quality assessment of historical data before generating the optimization model. By measuring the quality of historical data in advance and filtering or weighting it accordingly, the system prepares high-quality input data for model generation, ensuring that even restricted decision sets can produce reliable models without requiring complex post-processing
Data Source
AI summary
An example system includes a processor to receive historical data, a formal quality measure, a quality threshold, and a mathematical optimization model. At least part of the mathematical optimization model is generated from the historical data. The processor can measure a quality of the mathematical optimization model using the formal quality measure. The processor can then augment the mathematical optimization model such that the measured quality of the augmented mathematical optimization model exceeds the target quality threshold.