Laser-driven inertial confinement fusion (ICF) experiments generate massive multi-modal diagnostic data (optical/X-ray images, waveforms, arrays, etc.), whose manual processing leads to severe inefficiencies, data silos, and reproducibility challenges. We present CPED, an end-to-end framework that transforms ICF data curation through standardized pipelines. CPED integrates diverse diagnostics (laser, X-ray imaging, radiation flux, backscatter, neutron measurements, etc.) into queryable repositories. By enforcing strict Standard Operating Procedures (SOPs) codified in institutional standards, CPED ensures operator-agnostic reproducibility and eliminates ad-hoc scripting. Implementation at our facility demonstrates >10x faster processing and has handled hundreds of experimental shots, producing FAIR (findable, accessible, interoperable, reusable) datasets. These datasets have directly enabled statistical analyses of laser performance, hotspot morphology, and hohlraum scaling in our studies. As a replicable blueprint for data infrastructure, CPED transforms manual workflows into systematic data governance. The framework now serves as a core component of our facility’s digital ecosystem, laying the foundation for future AI-driven research.
 

ICF data governance,automated processing pipelines,scientific reproducibility