T2K (Tokai to Kamioka) is a long-baseline neutrino oscillation experiment. The neutrino beam is produced at J-PARC and measured by the near detector located at 280 m (ND280), and then by the far detector (Super Kamiokande) located 295 km away from the beam target. One of the main goals of T2K is to find evidence of CP symmetry violation in the lepton sector by measuring neutrino oscillations. The evaluation of systematic uncertainties related to the neutrino flux and interactions in the matter is essential to achieving this goal. One of the main sources of systematic uncertainty stems from the modeling of neutrino-nucleus interactions. To constrain these uncertainties, a fit of the near detector data is performed. However, the choice of the neutrino interaction model is susceptible to introducing a bias in the measurement of oscillation parameters. To estimate the size of the possible bias and the robustness of the error model, we perform fits of simulated data sets which are generated by varying parts of the baseline interaction model. This poster will present various alternative models considered in the analysis, along with the results of the simulated data set fits and their impact on the sensitivity to oscillation parameters.