Aircraft Engine Maintenance Forecasting via Failure Model Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current maintenance forecasting for aircraft engines lacks precision due to attributing the same average intervention level to all causes of failure, failing to account for engine specifics and history, leading to incomplete maintenance and potential re-breakdowns.
Innovation Solution
A method that selects a relevant failure model based on engine age and associated parameters, such as operation duration and life potential of parts, to determine a specific maintenance intervention level, using decision rules and statistical models derived from fleet data to predict maintenance needs with greater accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the same average intervention level is assigned to all causes of failure based on Weibull statistical failure curves, then maintenance planning is simplified and standardized, but precision in forecasting maintenance operations deteriorates and engine specificities are not taken into account
Solution Approach 1:
The patent segments the maintenance forecasting approach by dividing causes of failure into different categories (e.g., wear-related failures, fatigue-related failures, corrosion-related failures) and assigning different Weibull distribution parameters to each category. This allows tailored intervention levels for each failure type while maintaining overall system organization and simplicity.
Solution Approach 2:
The patent applies local quality by considering engine-specific parameters such as operating hours, flight cycles, environmental conditions, and historical maintenance data to adjust the intervention level for each specific engine and failure cause combination. This enables precision in forecasting while maintaining standardized procedures through parameterized adjustments.
2Productivity
If average intervention levels are used for all engines, then resource allocation is simplified, but reliability deteriorates due to incomplete maintenance and potential re-breakdowns
Solution Approach 1:
The patent changes parameters by using different Weibull distribution parameters (shape parameter β and scale parameter η) for different failure modes and engine conditions. The shape parameter indicates the failure pattern (infant mortality, random failure, wear-out) while the scale parameter indicates the characteristic life, allowing dynamic adjustment of maintenance timing and intensity to optimize both efficiency and reliability.
Solution Approach 2:
The patent introduces dynamics by continuously updating the Weibull parameters and intervention levels based on actual engine performance data, maintenance history, and changing operational conditions. This dynamic adjustment ensures that maintenance plans remain optimized for reliability while adapting to real-world variations in engine behavior.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method and a system for forecasting maintenance operations for a common aircraft engine, comprising: processing means for checking against a set of failure models (Ml, Mn) adapted to said common engine in order to select a relevant failure model (Mi) with a failure age (T0) defining the age of said engine at the time of failure; processing means for associating decision rules (R), relating to the level of intervention with respect to said common engine, with said relevant failure model (Mi) on the basis of a set of parameters (PI, p2, Pi) relating to said common engine; and processing means for determining, on the basis of said decision rules, the required level of the maintenance intervention to be performed on said common engine.