In recent years, with the Civil Aviation Administration promoting the shift from scheduled maintenance to precision maintenance, predictive maintenance technology has gradually become the core innovation direction in the civil aircraft maintenance industry. Recently, Shandong Airlines Engineering Technology Company, leveraging the successful practice of the APU oil quantity monitoring project, has established a full lifecycle APU health management system, providing a benchmark case of "data-enabled maintenance" for the industry and driving the transformation of aviation maintenance from "post-event handling" to "full-time domain health management".

It is understood that traditional APU lubricating oil management relies on fixed-cycle inspections every 15 days, which presents a dual contradiction of "over-inspection" and "lagging maintenance". This not only wastes a lot of maintenance resources but may also lead to unplanned downtime due to the failure to detect hidden dangers in a timely manner. To address this pain point, Shandong Airlines Engineering Technology Company has innovatively introduced predictive maintenance logic. By collecting lubricating oil status telegram data in real-time through relevant systems, it accurately detects the time nodes when the lubricating oil volume drops to a threshold value. Through intelligent push task instructions, it achieves "on-demand maintenance" instead of "regular inspections".

Data bears witness to effectiveness. From November 2024 to March 2025, the system successfully pushed 288 precise maintenance tasks, reducing ineffective work orders by 81.8% compared to traditional methods and effectively freeing up frontline maintenance resources. Simultaneously, by analyzing data such as lubricant addition volume and consumption rate, the system can automatically identify abnormal consumption patterns and cross-verify them with sensor statuses. To date, it has preemptively detected 10 sensor failures and 7 aircraft with abnormal lubricant consumption, boosting fault diagnosis efficiency by 60% and successfully preventing multiple unplanned groundings and associated risks.

In addition, the project innovatively introduces the "soft time limit" management concept, conducts reliability assessments on key components of the APU, dynamically adjusts maintenance plans, and achieves precise control through "one machine, one strategy". As a "predictive maintenance pilot demonstration project" of Shandong Airlines, the successful experience of APU oil quantity monitoring is being applied to the entire fleet. In the future, Shandong Airlines will integrate QAR/DAR data streams, introduce machine learning algorithms, and incorporate new technologies such as VR/AR intelligent detection and digital twins to build a closed-loop system of "perception-analysis-decision-verification", continuously enhancing the level of intelligent maintenance, driving maintenance capabilities to leap from "compliance and standard-meeting" to "smart leadership", and injecting stronger momentum into the safe operation of civil aviation.