Machine Learning Plan Validation for Industrial Process Schedules
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The process of operational planning in process industries is costly and inefficient due to the need for multiple experienced resources to review complex operating plans and schedules, exacerbated by demographic gaps where junior resources take over experienced planners' responsibilities.
Innovation Solution
A method and system that uses feature extraction technology to identify a subset of critical process variables from historical data, filtering and grouping them based on importance and correlation, allowing for efficient plan validation and scheduling through Machine Learning technology.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple experienced resources review operating plans and schedules, then validation accuracy and reliability improve, but labor costs and time consumption increase
Solution Approach 1:
The patent extracts and isolates the critical validation function from the manual review process by implementing an automated validation system that uses machine learning models to identify and flag anomalies in operating plans and schedules, separating the analytical task from human reviewers
Solution Approach 2:
The patent replaces the mechanical system of manual human review with an automated electronic validation system that uses algorithms and machine learning to perform the same validation function, eliminating the need for multiple experienced resources to manually review plans
2Measurement precision
If experienced resources are used for plan review, then detection of anomalies and suggestions for improvement improve, but resource costs increase
Solution Approach 1:
The patent implements a self-service validation system where the operating plans and schedules validate themselves through automated comparison against historical data and constraints, eliminating the need for external human reviewers
Solution Approach 2:
The patent replaces human expertise with an automated electronic system that uses machine learning models trained on historical data to detect anomalies and suggest improvements, substituting human cognitive resources with computational resources
3Reliability
If a comprehensive review of all process variables is conducted, then validation thoroughness improves, but complexity and time consumption increase
Solution Approach 1:
The patent extracts only the critical subset of variables that need validation by using machine learning to identify and isolate the most significant parameters, eliminating the need to review all variables comprehensively
Solution Approach 2:
The patent segments the validation process into distinct automated steps including data extraction, anomaly detection, and report generation, simplifying the overall complexity while maintaining thoroughness
4Productivity
If Machine Learning technology is used for variable selection, then efficiency and robustness of validation improve, but implementation complexity increases
Solution Approach 1:
The patent performs preliminary action by pre-training machine learning models on historical operating data before deployment, so that when the system is implemented, the variable selection and anomaly detection capabilities are already optimized, reducing the apparent complexity of implementation
Data Source
AI summary
Disclosed are methods and systems that help identify critical variables for more efficient and robust plan validation process. An example embodiment is a computer implemented method of industrial process control. The example method includes receiving in computer memory a dataset including initial process parameters representing operational data of a subject industrial process, and, using filtering operations and grouping operations on the dataset, identifying a subset of the process parameters indicative of control data for controlling the subject industrial process. The example method further includes automatically applying the identified subset of process parameters controlling the subject industrial process.