The field of neutrino oscillation physics has yielded numerous significant results in recent decades. The Japan-based long-baseline accelerator neutrino oscillation Tokai-to-Kamioka (T2K) experiment, renowned for its contributions to accelerator neutrino oscillation studies, is undergoing several upgrades to further advance high-impact measurements. Among the newly constructed subdetectors of the T2K off-axis near detector ND280 is the Super Fine-Grained Detector (SuperFGD), which was installed in 2023. It is a high-resolution particle tracker that is made up of approximately two million plastic scintillating cubes. Its novel design allows the detector to measure neutrino interactions with high precision, an essential key to reducing uncertainties of current and future neutrino experiments.

This thesis includes a brief summary of the SuperFGD design and the related calibration tasks. In particular, the charge and pedestal calibration for the Multi-Pixel Photon Counters (MPPCs) are described in detail. A neural network based on a Transformer architecture developed for general voxelised 3D detectors was utilised to increase calibration tolerance.

Finally, this thesis presents the development of a Standard Model three-neutrino oscillation global fit study within the GAMBIT software framework, where the incorporation of the likelihood functions of the NO$\nu$A, MINOS, and KamLAND experiments are detailed. Their validation results demonstrate good agreement with the experimental counterparts, but the limited public data and experiment information negatively affect the accuracy of the implementations. A preliminary scan, which included eight neutrino oscillation experiments that covered a wide range of measurement technologies, was also performed. In general, the early results are consistent with the existing neutrino global fits, and some noticeable deviations are discussed.