A Novel Configuration-Driven Data Mining Framework for Health and Usage Monitoring Systems

Health and Usage Monitoring Systems (HUMS) gather a large amount of real-time data from various aircrafts. Such data can be utilized to support Condition Based Maintenance (CBM) paradigm, which analyzes (i.e., mine) data to facilitate proper maintenance actions if and when necessary. Methods and tools for automating such CBM systems are, therefore, key to efficiently identify, and report, the components on which maintenance is required. This paper presents a novel framework for building Automated Condition Based Maintenance Checking Systems (ACBMCS). The proposed framework is based on a layered architecture that combines two prominent techniques, namely XML-based metadata storage, and dynamic code generation and execution. These techniques help significantly cut down the cost of software maintenance, by allowing system engineers to add new functionality, analyze new data sets, and perform, for instance, complex exceedance/BIT operations, without the need to write new, or modify existing, executable source codes. We envision this framework as a huge concept, due to the fact that any software release is typically very expensive to test and deploy.