Aircraft Maintenance Decision Aid Using Failure Message Graphs

Solution Overview

Problem

Current aircraft maintenance policies often defer curative actions, increasing the risk of flight delays or cancellations due to unforeseen equipment degradation, as they lack effective predictive tools for anticipating critical failures.

Innovation Solution

A decision aid method for aircraft maintenance that uses a directed graph to analyze failure messages from past flights to predict the probability of future warning messages, employing random walks and prior learning to identify combinations of failure messages likely to generate warnings, thereby providing operators with proactive maintenance insights.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If curative actions are deferred to successive checks, then maintenance productivity is improved, but aircraft reliability deteriorates due to increased risk of flight delays or cancellations

Engineering Contradiction:
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system performs preliminary analysis of failure message combinations and calculates probability of warning message occurrence before the actual maintenance check. By predicting which equipment is likely to fail based on historical patterns, the system enables maintenance operators to prioritize inspections and perform curative actions proactively, rather than merely reactively during scheduled checks.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors failure messages in real-time and provides feedback to maintenance operators about the probability of warning messages occurring. This feedback loop allows dynamic adjustment of maintenance priorities and enables operators to make informed decisions about whether to defer or immediately address potential failures, balancing productivity and reliability.

Inventive Principle:
Principle #23Feedback

2Reliability

If predictive analysis of failure messages is implemented, then aircraft reliability is improved, but device complexity increases due to graph processing and random walk algorithms

Engineering Contradiction:
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system introduces an intermediary computational layer that processes raw failure messages through directed graphs and random walk algorithms to generate probability predictions. This intermediary layer acts as a mediator between the complex data processing requirements and the simple binary decision of whether to perform maintenance, encapsulating the complexity within a manageable predictive model that outputs clear actionable insights.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system replaces traditional mechanical inspection methods with computational analysis of failure message patterns. Instead of physically examining each component during scheduled checks, the system uses algorithmic processing of textual failure messages to predict failures, substituting computational complexity for physical inspection complexity.

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

3Measurement precision

If detailed failure message combinations are analyzed, then measurement precision of failure prediction is improved, but loss of time increases due to processing extensive message sets

Engineering Contradiction:
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system segments the analysis by organizing failure messages into a directed graph structure where nodes represent warning messages and arcs represent transitions based on failure message combinations. This segmentation allows the complex processing of extensive message sets to be divided into manageable graph traversals, improving both precision through detailed pattern analysis and efficiency through structured processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system changes the parameter of analysis from examining individual failure messages to analyzing combinations of messages represented as paths in a directed graph. By transforming the data representation and using random walk probabilities, the system achieves higher prediction precision while managing processing time through mathematical optimization of the graph traversal.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS9767413B2Method and computer program for the maintenance aid of aircraft equipment
Publication Date: 2017.09.19
  • US9767413B2 patent drawing
  • US9767413B2 patent drawing
  • US9767413B2 patent drawing

AI summary

According to a first aspect, the subject matter disclosed herein relates to a decision aid method for maintenance operations of aircraft equipment, the aircraft including a maintenance system arranged to make a record of equipment failure messages and of equipment failure warning messages emitted during a flight, the method comprising extracting at least one combination of failure messages corresponding to failure messages recorded by the maintenance system during a flight and a set of previous flights; determining, as a function of the at least one extracted combination, a probability of occurrence of at least one warning message possibly accompanied by one or more failure messages.