Production Recipe Optimization Using Multi-Factor Utility Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current processes for producing chemical, pharmaceutical, and biotechnological products fail to accurately account for the complex interplay of factors affecting production, often relying on simplistic approaches that underestimate non-linear dynamics and lack integration of comprehensive knowledge.
Innovation Solution
A computer-implemented method that determines optimal recipes by simulating production processes using variable recipe and evolution parameters, incorporating process evolution information, likelihood functions, and performance indicators to optimize utility scores.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional simplistic optimization approaches are used to focus on identifying and optimizing for a particular set-point, then the optimization process is easier to implement, but the complex interplay between different factors characterizing process performance is underestimated
Solution Approach 1:
The patent transforms the optimization approach from focusing on single set-points to optimizing multiple performance indicators simultaneously. The system changes the parameters being optimized from simple target values to comprehensive utility functions that incorporate multiple factors (product titre, quality attributes, process robustness), thereby resolving the contradiction between ease of implementation and measurement precision.
Solution Approach 2:
The patent creates a composite optimization framework that integrates multiple performance indicators into a unified utility function. This composite approach combines various process outcomes (yield, quality, robustness) into a single comprehensive metric, allowing simultaneous optimization of multiple factors while maintaining systematic implementation.
2Device complexity
If available knowledge about the process is taken into consideration only in a rudimental manner, then the process design is simpler, but the optimization accuracy is reduced
Solution Approach 1:
The patent applies preliminary action by systematically incorporating available process knowledge (from theoretical assumptions and previous executions) into the optimization framework before the actual optimization begins. The system pre-integrates existing data and expertise into the utility function definitions and performance indicator selections, thereby improving optimization accuracy without significantly increasing design complexity.
Solution Approach 2:
The patent implements feedback mechanisms where simulation results and actual process outcomes are continuously fed back into the optimization system. This feedback loop allows the system to refine its understanding of process dynamics and improve parameter optimization accuracy by learning from both theoretical models and empirical data.
3Productivity
If maximisation of product titre is pursued as the primary goal, then the product yield is improved, but the overall process performance may not be optimized
Solution Approach 1:
The patent applies universality by creating a multi-functional optimization framework that simultaneously evaluates multiple performance indicators including product titre, quality attributes, and process robustness. The utility function serves multiple purposes: maximizing yield while ensuring quality standards and maintaining process reliability, thereby resolving the contradiction between productivity and reliability.
Data Source
Figure 1
Figure 2
Figure 3a
AI summary
Summarizing the invention, a computer-implemented method of determining at least one recipe for a production process to produce a chemical, pharmaceutical and/or biotechnological product is provided, wherein the production process is defined by a plurality of steps specified by recipe parameter(s) controlling an execution of the production process and a recipe comprises the plurality of steps defining the production process.