This work-in-progress paper proposes an approach to quantify process information in the context of sensor selection in chemical plants. The method is based on approximating the global state uncertainty via Monte Carlo simulation. In contrast to most existing approaches for sensor selection, this promises insusceptibility against dependent non-Gaussian uncertainties in steady-state or dynamic processes with the possibility to integrate data-driven models into the set of state equations. First results demonstrate that the approach can find the Pareto optimal sensor configuration for an in-silico continuous stirred tank reactor (CSTR). Additionally, the method's further development for applications in real-world scenarios is discussed.