Process Model Generation for Guided Plant Manual Procedures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for guiding operators through manual procedures in plant operations, such as start-up or shut-down, rely heavily on operator experience and diligence, leading to inconsistent quality of execution.
Innovation Solution
A method to generate a process model by mining historical operational data to create a workflow model that guides operators and monitors their compliance, using statistical properties of process variables and set point changes, and integrating machine learning for precise recommendations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If special operator displays or event batch recipe systems are provided to guide operators, then the reliability of procedure execution is improved, but the device complexity increases
Solution Approach 1:
The patent creates a virtual copy of the procedural knowledge by generating a process model from historical log data. This digital twin captures the sequential order and statistical properties of manual operations, providing guidance without requiring complex physical displays or recipe systems. The process model serves as a lightweight virtual representation that guides operators while maintaining simplicity.
Solution Approach 2:
The system enables self-service by automatically learning procedural knowledge from historical operator actions stored in log data. Instead of requiring manual configuration of complex display systems or recipe systems by engineers, the process model is autonomously generated by analyzing past operational data, identifying sequential patterns and statistical properties of process variables and set point changes.
2Ease of operation
If manual procedure execution is left to operator experience and diligence, then the ease of operation is maintained, but the manufacturing precision of procedure execution deteriorates
Solution Approach 1:
The patent implements feedback by using the generated process model to monitor and evaluate operator actions during procedure execution. The system compares actual operator actions against the learned sequential order and statistical properties from historical data, providing guidance and monitoring that ensures consistent execution quality while maintaining operational simplicity through intuitive interfaces.
3Productivity
If a process model is generated by mining historical log data and applying machine learning, then the productivity of procedure guidance is improved, but the device complexity increases
Solution Approach 1:
The patent replaces complex mechanical or manual systems with data-driven computational methods. Instead of manually creating procedure models or configuring complex guidance systems, the approach uses machine learning algorithms to automatically analyze historical log data and generate process models. This substitution of mechanical/manual model creation with automated data mining and statistical analysis improves productivity while managing complexity through algorithmic approaches.
Data Source
AI summary
A method for generating a process model modeling a manual mode procedure instance of a plant process includes providing log events of operational actions; selecting related sequences of manual mode operational actions from the log events; filtering the related sequences according to an individual plant section; identifying a sequential order from the filtered related sequences; determining statistical properties of values of related process variables and/or statistical properties of values of related set point changes to each sequential ordered manual mode operational action from the filtered related sequences; generating the process model of the manual mode procedure instance by arranging related manual mode operational actions with the sequential order of each operational action assigned with the statistical properties of the values of related process variables and/or assigned with the statistical properties of the values of the related set point changes.