Chromatography Column Performance Monitoring via Multivariate Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current chromatography column monitoring methods are not sufficiently sensitive or comprehensive, leading to false positives, missed gradual trends, and unnecessary re-packing or waste of valuable product due to inadequate detection of column performance changes.

Innovation Solution

Applying multivariate analysis (MVA) methods to both process and transition analysis data for comprehensive evaluation of chromatography column performance, enabling near real-time monitoring of packing quality, integrity breaches, and subtle changes, and reducing operator subjectivity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional pulse test HETP methods are used for column monitoring, then the monitoring process is simple to implement, but the detection sensitivity is insufficient leading to false positives and missed gradual trends

Engineering Contradiction:
Improvedetection sensitivityVSAvoidmonitoring system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent combines multiple monitoring parameters (HETP, asymmetry, elution UV peak width, product yields) into a unified multivariate analysis system. This integration allows simultaneous evaluation of multiple column performance aspects, improving detection sensitivity while managing complexity through systematic data processing.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent transitions from univariate parameter monitoring to multivariate statistical analysis, adding dimensional depth to the monitoring approach. By analyzing multiple parameters simultaneously and their interrelationships, the system detects subtle column performance changes that single-parameter methods miss, resolving the contradiction between simplicity and sensitivity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Reliability

If conventional univariate parameter monitoring is used, then the analysis is straightforward and quick, but it fails to detect subtle changes and gradual trends in column performance

Engineering Contradiction:
Improvecolumn performance evaluation reliabilityVSAvoidperformance examination time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent implements preliminary action by establishing multivariate statistical models and control limits in advance based on historical data. Once configured, the system automatically evaluates new data against these pre-established criteria, providing rapid and reliable column performance assessment without requiring extensive real-time analysis, thus resolving the time-reliability contradiction.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system incorporates feedback mechanisms where monitoring results feed back into the statistical models for continuous refinement. This allows the system to learn from accumulated data, improving detection accuracy over time while maintaining efficient evaluation speeds through automated statistical processing.

Inventive Principle:
Principle #23Feedback

3Measurement precision

If frequent pulse test HETP measurements are performed, then detection sensitivity improves, but product waste increases due to false positives and unnecessary re-packing

Engineering Contradiction:
Improvecolumn performance detection accuracyVSAvoidproduct waste
Core Design Contradiction:
Measurement precisionVSLoss of substance

Solution Approach 1:

The patent creates a universal monitoring system that evaluates multiple column performance aspects simultaneously through multivariate analysis. This multi-functional approach provides comprehensive column health assessment in a single evaluation, improving detection accuracy without increasing measurement frequency or causing false positives that lead to product waste.

Inventive Principle:
Principle #6Universality (Multi-functionality)

4Loss of information

If comprehensive monitoring of multiple parameters is implemented, then detection capability improves, but the complexity of data analysis and processing increases

Engineering Contradiction:
Improveinformation completenessVSAvoiddata processing complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent replaces complex manual data analysis with automated multivariate statistical analysis. The system automatically processes multiple parameters, calculates statistical relationships, and generates performance evaluations without requiring extensive manual intervention, thus maintaining information completeness while managing data processing complexity through automation.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP2338049B1Methods for evaluating chromatography column performance
Publication Date: 2020.11.25 BIOGEN MA INC
  • EP2338049B1 patent drawingFigure 1
  • EP2338049B1 patent drawingFigure 2
  • EP2338049B1 patent drawingFigure 3

AI summary

Methods for evaluating and/or monitoring chromatography column performance are provided. Embodiments apply multivariate analysis (MVA) methods to process data as well as transition analysis data to provide a comprehensive evaluation of chromatography column performance. In embodiments, transition analysis data generated over extended periods of time can be analyzed together with process data to evaluate column performance. Further, embodiments enable a compact and robust tool for combining and presenting performance evaluation results, which allows for time-efficient performance examination. According to embodiments, MVA methods applied on transition analysis and process data provide (1) near real-time ability to comprehensively monitor column packing quality; (2) sensitive detection of column integrity breaches; (3) sensitive detection of subtle changes in column packing; (4) sensitive detection of different types of changes in column packing; (5) sensitive detection of fronting/tailing; and (6) sensitive detection of changes in process performance.