Risk Lifecycle Management Using Tracing Matrices for Causality Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional risk management technologies fail to accurately reflect actual risks by disregarding relationships between failure modes and misidentify risk profiles, and are cumbersome and costly to track changes across different versions and iterations in manufacturing and product development environments.
Innovation Solution
A risk based lifecycle management system that utilizes a machine learning model to determine potential multi-variant causes of adverse events by associating process steps with objects and attributes through a tracing matrix, enabling continuous improvement and regulatory compliance by tracking changes and applying corrective actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional risk management technologies use FMEA to identify possible failures, then failure modes can be cataloged, but the actual risk of failures is not accurately reflected because relationships between failure modes are disregarded
Solution Approach 1:
The patent combines multiple failure modes into composite risk profiles that capture relationships between failures. Instead of treating each failure mode independently as in traditional FMEA, the system merges related failure modes into unified risk profiles that represent actual risk scenarios, thereby improving measurement precision while preserving information about failure relationships.
Solution Approach 2:
The system creates a universal risk profile framework that can represent multiple failure modes and their relationships in a unified structure. This multi-functional approach allows the same risk profile to capture various aspects of failure relationships (causality, correlation, etc.) without requiring separate analysis for each relationship type.
2Productivity
If RPN numbers are used to represent risk priority levels, then risks can be ranked, but RPN numbers of the same value may not reflect the same level of actual risk
Solution Approach 1:
The patent transforms the traditional single-value RPN parameter into a multi-dimensional risk profile structure. Instead of relying on a single RPN number that loses information, the system uses multiple parameters (failure modes, relationships, severity weights) to represent risk levels, maintaining productivity through automated calculation while dramatically improving measurement precision.
3Reliability
If tracking is performed between different versions and iterations of FMEA changes, then regulatory requirements can be met, but the process becomes cumbersome and costly
Solution Approach 1:
The system creates digital copies of risk profiles across different versions and iterations, allowing automated tracking of changes. Instead of manual tracking processes, the patent uses digital replication and version control mechanisms that automatically maintain regulatory compliance while reducing complexity and cost of change tracking.
Data Source
AI summary
Risk based lifecycle management systems are presented herein. A system determines a process ontology of a process including process steps of the process and objects including attributes corresponding to performances of the process steps; associates a process step with respective objects using a tracing matrix; detects an event corresponding to a performance of the process step; and associates the event with the process step and the respective objects using the tracing matrix. In response to determining, utilizing a machine learning model, that the event corresponds to a defined risk profile including defined failure modes, the system selects a group of defined failure modes as candidates of causality of the event representing potential multi-variant causes of the event, and sends the candidates of causality of the event directed to a user identity to facilitate mitigation of effects of the potential multi-variant causes of the event on the process.