A new type of sensor (BioFET) is becoming advantageous over already existing ones. BioFETs are biologically sensitive field-effect transistors which are equipped with a biofunctionalized surface that replaces the gate structure of conventional MOSFET design. This surface enables the device to be sensitive to special bonds in an aqueous solution that contains an analyte. The analyte specifically binds to the biofunctionalized surface layer and changes the charge distribution and surface potential, which in turn influence the conductivity of the transducer. This little change in conductivity offers a label-free detection within minutes. Hence a wide variety of applications is possible, and therefore BioFETs are becoming very important in the biomedical field. Over the last few years the first sensors based on conventional silicon structures and silicon nanowires have been developed. BioFETs with the ability of detecting single-stranded DNA and tumor markers have been produced. In spite of such successes, the theoretical framework to enable an understanding of the experiments in a quantitative manner is still missing. This is due to the complexity of the system, which needs a consistent analysis of the semiconductor, the surface chemistry, and the transport mechanisms in the aqueous solution. As mentioned above, the functioning of the device is due to the charge distribution and potential on the biofunctionalized surface layer. Thus it is very important to understand the physics behind it and to describe it with an adequate multi-physics model. The semiconductor is described via a drift-diffusion-equation, while the liquid obeys the Poisson-Nernst-Planck equation. So a self-consistent simulation of the whole device is necessary to quantitatively understand and engineer these novel devices.