Fleet Reliability Monitoring for Bad Actor Part Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current monitoring systems are inadequate for efficiently identifying and addressing anomalies in Maintenance and Repair Operations (MRO) of asset fleets, leading to increased sustainment costs due to the operation of 'bad actor' assets or parts, which are not effectively detected by existing Statistical Process Control (SPC) methods.
Innovation Solution
A system and method for monitoring reliability data of asset fleets, utilizing computational processing of reliability data to identify 'bad actor' parts and assets by employing automated modeling analytics and decision support analytics, which includes data cleanup, probability distribution modeling, and Statistical Process Control (SPC) rules to flag and prioritize parts at higher risk of failure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Statistical Process Control (SPC) methods are used to monitor asset fleets, then anomaly detection capability is improved, but the ability to identify 'bad actor' parts and assets is insufficient
Solution Approach 1:
The patent segments the asset fleet into individual assets and parts, tracking each with unique identifiers. This segmentation enables the system to identify specific 'bad actor' parts and assets by analyzing their individual reliability data separately from the fleet average, rather than treating the fleet as a homogeneous group.
Solution Approach 2:
The system performs preliminary actions by collecting and storing reliability data for each part and asset over time, establishing baseline performance metrics before anomalies occur. This preliminary data accumulation enables the SPC rules to detect deviations and identify bad actors more accurately when anomalies occur.
2Difficulty of detecting and measuring
If conventional monitoring systems are used, then basic anomaly detection is achieved, but sustainment costs increase due to undetected bad actor assets
Solution Approach 1:
The patent implements feedback by continuously monitoring reliability data and applying SPC rules to generate alerts when parts or assets deviate from expected performance patterns. This feedback loop enables timely identification of bad actor parts and assets, allowing operators to take corrective actions that reduce sustainment costs by preventing further failures and inefficiencies.
Solution Approach 2:
The system replaces conventional mechanical monitoring approaches with computational processing of reliability data. By using automated data collection, storage, and analysis systems, the patent enables more sophisticated anomaly detection and bad actor identification without requiring physical inspection or manual monitoring of each asset.
3Productivity
If SPC methods are applied to asset fleets, then process control is improved, but the complexity of implementing effective monitoring increases
Solution Approach 1:
The patent applies universal SPC rules that can be used across different asset types and industries. The same core monitoring framework and statistical methods are adapted to various contexts by adjusting parameters and data collection methods, reducing the need for completely separate monitoring systems for different asset classes.
Solution Approach 2:
The system manages complexity by changing parameters such as the significance level (alpha) and control limits based on specific asset characteristics and operational requirements. This allows the monitoring system to be customized for different applications without fundamentally altering the core SPC methodology, balancing effectiveness with implementation simplicity.
Data Source
AI summary
A computer-based monitoring system and monitoring method implemented in computer software for analyzing the reliability data collected in the maintenance and repair operations in a fleet of assets with parts of the same type over a period of time. The reliability data include reliability event data, such as failures, repairs, and replacements of the parts. The reliability data further include asset usage data, such as usage time, or usage missions, or usage mileage, or usage flight hours, or such. The monitoring system analyses historical reliability data to build a reliability model. This model is further used in reliability SPC algorithms detecting bad actor assets and bad actor parts that show consistently worse reliability than normal parts and assets.