Dynamic Regulatory Change Management System

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Enterprises with multiple lines of business face challenges in maintaining compliance with varying regulatory requirements, as manual analysis of regulatory changes is resource-intensive and often results in irrelevant data being analyzed across different business lines.

Innovation Solution

A dynamic regulatory change management system that continuously monitors regulatory data sources, uses machine learning to identify coverage area indicator terms and phrases, and determines impact values to automatically alert relevant business lines, thereby filtering out irrelevant changes and providing a compliance action plan.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual analysis of regulatory changes is performed across all business lines, then comprehensive compliance coverage is achieved, but resource consumption increases significantly

Engineering Contradiction:

Improvecompliance coverageVSAvoidresource consumption

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The system segments regulatory changes by identifying coverage area indicator terms and phrases to divide the enterprise into different business lines. Each regulatory change is analyzed and assigned to specific business lines that are actually affected, rather than distributing all changes to all business lines. This segmentation reduces resource consumption while maintaining compliance coverage for affected areas.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system applies local quality by providing targeted regulatory change analysis to specific business lines based on their relevance. Instead of uniform manual analysis across all business lines, the system identifies and delivers regulatory changes only to the specific business lines where they apply, optimizing resource allocation while ensuring appropriate compliance coverage.

Inventive Principle:

Principle #3Local quality

2Reliability

If manual parsing of all regulatory documentation is performed, then no relevant changes are missed, but analysis time and resources increase

Engineering Contradiction:

Improvedetection completenessVSAvoidanalysis time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary action by pre-identifying coverage area indicator terms and phrases in regulatory documentation before full analysis. The machine learning model is trained beforehand to recognize these indicators, enabling rapid filtering and classification of regulatory changes. This preliminary processing reduces analysis time while maintaining detection completeness through systematic identification of relevant changes.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system replaces manual mechanical parsing with automated machine learning-based analysis. The machine learning model automatically identifies coverage area indicator terms and phrases, determines impact values, and classifies regulatory changes without manual intervention. This substitution dramatically reduces analysis time while maintaining or improving detection completeness through consistent automated application of classification criteria.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Loss of energy

If automated parsing is implemented to reduce manual effort, then resource consumption decreases, but accuracy in identifying relevant changes may deteriorate

Engineering Contradiction:

Improvemanual effortVSAvoididentification accuracy

Core Design Contradiction:

Loss of energyVSMeasurement precision

Solution Approach 1:

The system implements feedback by using machine learning models that can be trained and refined based on performance metrics. The identification of coverage area indicator terms and phrases is continuously improved through feedback loops that adjust the model's sensitivity and specificity. This feedback mechanism ensures that automated parsing maintains high identification accuracy while reducing manual effort, as the system learns from past performance to improve future accuracy.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system applies parameter changes by adjusting the machine learning model's parameters such as impact value thresholds, coverage area indicator weights, and classification criteria. These parameter adjustments optimize the balance between resource consumption and identification accuracy. By tuning parameters like the impact value threshold and the importance weights of different coverage area indicators, the system achieves accurate identification of relevant regulatory changes while minimizing manual effort.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10872206B2System and framework for dynamic regulatory change management

Publication Date: 2020.12.22 BANK OF AMERICA CORP

AI summary

Embodiments of the present invention provide a system for dynamic regulatory change management for an enterprise. The system continuously monitors multiple regulatory data sources to identify regulatory change management documentation. The system then scans descriptive fields within each identified regulatory change documentation for coverage area indicator terms and phrases. A machine learning system then determines an impact value for each identified regulatory change documentation for one or more regulatory inventories of the enterprise by analyzing the coverage area indicator terms and phrases. If the impact value is above a predetermined threshold, the system alerts a user dashboard associated with a particular regulatory inventory to the regulatory change documentation. If the impact value is below the predetermined threshold, the system tags the regulatory change documentation as not being relevant to the regulatory inventory.