Algorithm of Weighted Full Flight Data Processing for Aircraft Health Management System

This research introduces a comprehensive methodology for employing Full-Flight Data Sensors (FFDS) to evaluate aircraft health inside an operational Health Status Infrastructure system. We formalize the acquisition chain (sensor - Data Concentrator Unit (DCU) - Digital Flight Data Acquisition Unit (DFDAU) - Flight Data Recorder) over ARINC 429/717/767 and ARINC 664 for intelligent transport system design, and detail the binary-to-physics translation (frame synchronization, Binary Number Representation, two’s-complement, scaling, and time alignment). A precision-weighted fusion model and a collaborative decision-making aggregation layer (additive/multiplicative scoring with classical criteria such as Wald, Laplace, and Hurwicz) convert channel-level evidence into a single, interpretable health index. For diagnostics, we implement a transparent pipeline-low-pass and thermal filtering, steady-regime selection, residual formation, PCA projection, and a one-class Mahalanobis decision rule. Simulations with representative scenarios show clear separation of a mid-flight compressor-efficiency drop from nominal behavior, while a constant +1% N1 bias remains below alarm thresholds, illustrating robustness. Transport-layer integrity is quantified via bit-error-rate to frame-error models.