Component Markov Chains for Composable Fault Tree Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current safety assessment methods for complex systems, such as aerospace and automotive, face limitations in expressing temporal sequences of events and composability, particularly with Markov chains, which are essential for reliability and safety analysis.
Innovation Solution
A component concept for Markov chains is introduced, allowing the creation of safety artifacts associated with system development elements, enabling modular and compositional safety analysis by assigning component Markov chains to each component with input and output failure modes for propagating failures, and transforming them into fault tree elements for qualitative analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If Markov chains are used for safety and reliability analysis of complex systems, then temporal sequences of events can be expressed and fault tolerance capabilities can be analyzed, but the model size experiences exponential explosion and composability is lost
Solution Approach 1:
The patent segments the system into multiple components, each with its own Markov chain model. Instead of creating one large exponential model, the system divides the analysis into manageable component-level models that can be independently analyzed and then composed. This segmentation breaks the exponential explosion by localizing the state space to individual components rather than the entire system.
Solution Approach 2:
The patent implements nesting by placing component-level Markov chains within a hierarchical structure where they are composed into system-level models. The component Markov chains are nested within the overall system architecture, allowing detailed component behavior to be embedded within the broader system context without creating a single exponential model.
2Reliability
If traditional Fault Tree_analysis is used for safety assessment, then failure modes and their causes can be identified, but temporal sequences of events cannot be expressed
Solution Approach 1:
The patent merges the strengths of Fault Tree Analysis (FTA) and Markov Chain analysis into a hybrid approach. The system combines the failure mode identification capability of FTA with the temporal sequence expression capability of Markov chains, creating a unified methodology that can both identify failure modes and model their temporal evolution. This merging allows the system to leverage the complementary advantages of both techniques.
3Ease of manufacture
If component Markov chains are created for each technical component, then modular and compositional safety analysis is enabled and Markov chains can be reused, but the integration complexity with fault trees increases
Solution Approach 1:
The patent creates universal component Markov chain models that can serve multiple functions: they can be used independently for component-level analysis, composed into system-level models, and integrated with fault tree structures. This multi-functionality allows the same component models to be reused across different analysis contexts, reducing overall complexity despite the enhanced integration capabilities.
Data Source
AI summary
A method for modelling technical systems having a plurality of technical components, including the step of assigning a component Markov chain to each component having a Markov chain for representing various states of the respective component, at least one input one failure mode for externally triggering a transition from one state of the Markov chain into another state of the Markov chain, and at least one output failure mode to each Markov chain for propagating failures to other components, is provided.