Application of Random Forest to Aircraft Engine Fault Diagnosis

Aircraft engine fault diagnosis plays a critical role in modern, cost-effective condition-based maintenance strategy in aircraft industry. Due to several inherent characteristics associated with aircraft engines, accurately diagnosing aircraft engine faults is a challenging classification problem. As a result, aircraft engine fault diagnosis has been an active research topic attracting tremendous research interests in machine learning community. In this paper, random forest classifier, a recently emerged machine learning technique, is applied to aircraft engine fault diagnosis in an attempt to achieve more accurate and reliable classification performance. Our primary objective is to evaluate effectiveness of random forest classifier on aircraft engine fault diagnosis. By designing a real-world aircraft engine fault diagnostic system, this paper investigates design details of random forest classifier and evaluates its performance. In this paper, we also make some efforts on investigating strategies for improving random forest performance specifically for aircraft engine fault diagnosis problem