Transferability to other propulsion types and aircraft models

Investigate the transferability of the LiteInception-based fault diagnosis framework, developed and validated on the NGAFID Cessna 172 dataset, to other propulsion types such as turboprop and turbofan engines and to different aircraft models.

Background

All experiments in the paper are conducted on the NGAFID dataset from Cessna 172 aircraft. The authors acknowledge that this single-aircraft validation limits generalizability.

They explicitly state that establishing whether the results and framework transfer to other propulsion types and aircraft models remains to be investigated, identifying a key path for future research and deployment.

References

First, dataset singularity. The experiments are validated solely on the NGAFID dataset from a single aircraft type (Cessna 172), and the transferability to other propulsion types (turboprop, turbofan engines) and aircraft models remains to be investigated.

— LiteInception: A Lightweight and Interpretable Deep Learning Framework for General Aviation Fault Diagnosis  (2604.01725 - Wei et al., 2 Apr 2026) in Section 5.6 (Limitations and Future Work)