Closed-Loop Planning and Scheduling Under Industrial Uncertainty
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current predictive systems for industrial planning and scheduling under uncertainty are inefficient, particularly in high-dimensional domains, as they fail to accurately account for uncertain data, leading to ineffective plans and increased computational and storage resources, and lack adequate feedback mechanisms for correcting prediction errors.
Innovation Solution
A computer-implemented method and system that generates optimized plans by receiving uncertain input data, applying uncertainty-based modification, and using a non-linear model predictive control optimization model with a feedback mechanism to ensure accurate predictions and automatic reconfiguration of industrial plant operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current predictive systems are used for industrial planning under uncertainty, then computational resources and storage are consumed, but prediction accuracy is insufficient and plans are ineffective
Solution Approach 1:
The system segments the planning problem into multiple scenarios based on different uncertainty realizations. Each scenario represents a possible future state, allowing the system to evaluate plans under diverse conditions separately, improving prediction accuracy without requiring a single computationally intractable optimization
Solution Approach 2:
The system performs preliminary actions by generating uncertainty-based modification data and creating multiple scenarios before final plan execution. This advance preparation allows the optimization model to work with pre-processed data structures, reducing computational burden during actual plan generation while maintaining high accuracy
2Reliability
If current predictive systems process high-dimensional uncertain data, then computational resources increase, but plan effectiveness decreases
Solution Approach 1:
The system incorporates a feedback mechanism where the optimization model's outputs are evaluated against multiple uncertainty scenarios. Results feed back into refining the plan, ensuring it remains effective under various conditions. This iterative feedback loop improves reliability without requiring excessive computational resources by focusing on critical scenario evaluations
Solution Approach 2:
The system changes parameters by applying uncertainty-based modifications to input data, transforming deterministic values into scenario-based distributions. This parameter transformation allows the system to handle high-dimensional uncertain data more efficiently by working with modified data representations that capture uncertainty without requiring proportional computational resources
3Measurement precision
If uncertainty-based modification is applied to input data, then prediction accuracy improves, but data processing complexity increases
Solution Approach 1:
The system applies dynamic adjustments to input data based on uncertainty characteristics. Rather than static data processing, the system dynamically modifies data representations according to uncertainty levels and scenario requirements, improving prediction accuracy while managing processing complexity through adaptive rather than uniformly complex operations
4Reliability
If feedback mechanism is implemented for correcting prediction errors, then plan reliability improves, but system complexity increases
Solution Approach 1:
The feedback mechanism compares optimization results against multiple uncertainty scenarios and uses discrepancies to refine plans. This targeted feedback approach improves reliability by focusing computational effort on correcting prediction errors in critical scenarios rather than uniformly increasing system complexity across all operations
Data Source
AI summary
Embodiments of the present disclosure provide for improved optimized plan predictions. Such embodiments, utilize optimized model(s) that accounts for uncertainty in input data to the model. Some example embodiments, receive input data associated with one or more industrial plants. At least a portion of the input data may include uncertain input data. The noted example embodiments, generate uncertainty-based modification data, and apply the uncertainty-based modification data to the input data to generate updated input data. The noted example embodiments generate, based at least in part on applying the updated input data to an optimization model, predicted optimized plan. The predicted optimized plan comprises optimized plan data. Further, the noted example embodiments initiate the performance of one or more prediction-based actions based at least in part on the predicted optimized plan.