Real-Time Multimodal Defect Detection and MES Feedback on Edge Devices via Bayesian Fusion and Causal Adaptation

Conventional single-modality or loosely coupled inspection systems often degrade under partial occlusion, transient sensor dropout, or illumination drift, and their feedback to Manufacturing Execution Systems (MES) can be delayed or non-deterministic.We present <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">D3Lite-MES</i>, a real-time multimodal defect detection framework that (i) aligns heterogeneous sensors at both hardware and software levels, (ii) fuses them via a Bayesian–Causal Weighted Fusion (BCWF) module, and (iii) adapts model parameters on edge devices through a lightweight causal adaptation routine. Deployed on commodity edge hardware, the system sustains line-takt operation and writes verified defect events back to MES with bounded latency. On three production lines covering <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">[product families anonymized]</i>, D3Lite-MES attains a mean F1 of 98.2% and [email protected] of 97.6%, reducing false alarms by 98.2% compared with lightweight baselines, at an end-to-end latency of 34 ms and average power of 8.9W. Ablations show that BCWF contributes +1.7% F1 under partial observations, while causal adaptation yields +1.7% under illumination drift. The MES feedback loop decreases manual triage time by 34% and shortens recovery from minor stops by 34%.