Accelerated Reliability Assessment of Electronic Warfare Systems Using Integrated Arrhenius-Based ALT and Crow-AMSAA Modelling

Modern defence electronic systems require rapid and reliable performance assessment under severe environmental conditions, where conventional long-duration testing and field data are often impractical. This paper presents an integrated reliability assessment framework combining Arrhenius-based Accelerated Life Testing (ALT) with Crow–AMSAA (Non-Homogeneous Poisson Process) modelling to enable time-compressed evaluation of system reliability. The methodology transforms accelerated stress data into equivalent use conditions and statistically characterizes failure occurrence behaviour. The framework is demonstrated through a case study on a Radar Exciter Receiver Processor Unit (ERPU), incorporating parameter estimation, confidence interval evaluation, and residual diagnostics. The estimated Crow–AMSAA slope under use conditions (β ≈ 0.966) indicates near-constant failure intensity, while the accelerated condition (β ≈ 0.607) suggests an apparent decreasing failure trend. To provide physical insight into these observations, Failure Mode, Effects and Criticality Analysis (FMECA) is integrated to identify dominant failure modes across key subsystems, including the waveform generator, synthesizer, and radar signal processor. The analysis shows that differing sensitivities of failure mechanisms to temperature stress led to non-uniform activation under accelerated conditions, resulting in early failure clustering and influencing the estimated NHPP parameters. Residual diagnostics further reveal systematic deviations at higher equivalent mission times, indicating limitations of the single-mechanism assumption inherent in Arrhenius modelling. The results demonstrate that reliability trends derived from accelerated testing are governed by both statistical transformation and the underlying distribution of failure mechanisms. The proposed framework provides a physically informed and statistically consistent approach for early-stage reliability assessment, design comparison, and identification of potential reliability risks in complex electronic systems.