Fault Tree Back-Tracing for Closed-Loop Failure Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for fault tree analysis in safety-critical systems struggle to effectively handle closed loops, which are problematic due to their complexity and the limitations of Boolean logic, leading to inexact results and inefficiencies in automated analysis.
Innovation Solution
A computer-implemented method that models multi-component systems using fault trees, identifies closed loops by back-tracing failure propagation paths, sets loop interconnections to Boolean TRUE, replaces problematic gates, and cuts off unnecessary inputs to remove loops, allowing for efficient analysis without remaining loops.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional fault tree analysis methods are used to analyze safety-critical systems, then the analysis can identify system failure states and causes, but the method cannot effectively handle closed loops in multi-component systems, leading to inexact results
Solution Approach 1:
The patent segments the fault tree analysis process into distinct phases: initial Boolean analysis, loop detection through back-tracing, and iterative refinement. By dividing the analysis into manageable segments, the method can handle closed loops systematically while maintaining accuracy in safety-critical system assessment
Solution Approach 2:
The patent implements a dynamic, iterative approach where the fault tree analysis is refined through multiple passes. The method dynamically adjusts by detecting loops during back-tracing and re-evaluating the fault tree until convergence is achieved, allowing accurate analysis of systems with closed loops that static methods cannot handle
2Extent of automation
If automated fault tree composition is used for multi-component systems, then the analysis can be performed automatically, but loops develop during composition that cannot be prevented by traditional methods
Solution Approach 1:
The patent incorporates feedback mechanisms where the analysis results from one iteration inform the next. During back-tracing, detected loops provide feedback that triggers re-evaluation of the fault tree. This feedback loop allows the automated system to handle complexity by learning from each iteration and converging to an accurate solution
Solution Approach 2:
The patent performs preliminary loop detection through back-tracing before final analysis completion. By identifying potential loops early in the automated composition process and addressing them through iterative refinement, the method prevents loop-related errors from propagating through the analysis, enabling reliable automated handling of complex multi-component systems
3Productivity
If existing loop removal techniques are applied, then some loops can be broken analytically, but the results become inexact and optimistic, and the failure grows with the number of loops
Solution Approach 1:
The patent uses dynamic iterative refinement where the fault tree analysis is repeatedly evaluated and adjusted. Rather than applying static loop removal that degrades accuracy, the method dynamically converges to an exact solution by detecting loops through back-tracing and re-evaluating the Boolean expressions until the analysis stabilizes, ensuring accurate failure probability even in systems with multiple loops
Solution Approach 2:
The patent replaces traditional mechanical loop-breaking approaches with a computational back-tracing method. Instead of analytically breaking loops which introduces approximation errors, the method uses systematic back-tracing through the fault tree structure to detect and handle loops exactly, substituting approximate mechanical methods with precise computational analysis
4Device complexity
If Boolean logic is used for failure propagation models, then the logic is simple and manageable, but Boolean logic cannot contain loops in general
Solution Approach 1:
The patent segments the Boolean logic evaluation into iterative passes, with loop detection and handling as distinct segments. By separating the Boolean evaluation from loop management through structured back-tracing, the method maintains the simplicity of Boolean logic while adding loop handling capability through a systematic multi-phase process
Solution Approach 2:
The patent introduces back-tracing as an intermediary mechanism between Boolean logic evaluation and loop detection. This intermediary process systematically traces failure propagation paths through the Boolean fault tree, identifying loops without requiring changes to the underlying Boolean logic, thus preserving logical simplicity while enabling loop handling
Data Source
AI summary
A computer-implemented method and device for resolving closed loops in automatic fault tree analysis of a multi-component system is provided. Also provided is a method for resolving closed loops in automatic fault tree analysis of a multi-component system, the closed loops corresponding, for example, to closed-loop control circuitry of a multi-component device. The closed loops are first identified in a top-down approach within failure propagation paths. Next, the loops are resolved by setting each loop interconnection to Boolean TRUE, adjusting the fault tree in a specific way and finally setting each loop interconnection to Boolean FALSE. Embodiments of the invention are relevant for analyzing safety-critical systems. However, the present concepts are not limited to these applications and may be applied to general use cases where fault tree analysis is applicable. The proposed solution advantageously provides a method that features linear complexity.