The heterogeneous continuous flow hydrogenation is pivotal in chemical research and production, yet its reaction optimization has historically been intricate and labor-intensive. This study introduces a heterogeneous continuous flow hydrogenation system specifically designed for the preparation of 2-amino-3-methylbenzoic acid (AMA), employing FTIR inline analysis coupled with an artificial neural network model for monitoring. We explored two distinct reaction optimization strategies: multi-objective Bayesian optimization (MOBO) and intrinsic kinetic modeling, executed in parallel to optimize the reaction conditions. Remarkably, the MOBO approach achieved an optimal AMA yield of 99% and a productivity of 0.64 g/hour within a limited number of iterations. Conversely, despite requiring extensive experimental data collection and equation fitting, the intrinsic kinetic modeling approach yielded a similar optimal AMA yield but a higher productivity of 1.13 g/hour, attributed to increased catalyst usage. Our findings indicate that while MOBO offers a more efficient route with fewer required experiments, kinetic modeling provides deeper insights into the reaction optimization landscape but is limited by its assumptions.