The mandatory usage of fuses within the electrical distribution system (EDS) has a strong impact on the cross section of the connected wires and thus on the weight and required amount of material, especially of copper. Currently, circuit breakers and, in most cases, thermal fuses are used due to their low cost and robust design. However, for future EDS architectures, the implementation of fuses using solid state switches is being considered for several reasons, e.g. resetability, diagnosis, smaller tolerances, and reduced dependancies on ambient temperature, arcing, or pulse sequences. In addition, there can be advantages on the system level, especially for higher voltages or in combination with other switch-based functionalities such as pulse width modulation, power multiplexing, or substitution of relays. Until now, smart solid state switches only emulate the behavior of thermal fuses and are therefore based on a fixed time-current data set. The new approach presented in this paper, however, uses a thermal model of the wire and the electrical current profile to calculate the wire temperature and thus allows a precise determination of an electrical failure. As the primary function of a fuse is the protection of the wire, such virtual temperature monitoring provides the most direct and reliable information possible. This thus minimizes the probability of faulty switching and simultaneously assures an exact and robust detection of a critical situation, e.g. a short circuit. Results include higher reliability, a reduction of wire cross-section and decreased copper usage. Another advantage is that further improvements are possible if additional data about the wire, e.g. the ambient conditions or the wire location, can be made available. The paper describes the theory, the method and its requirements, and the basic algorithm, as well as initial simulation and measurement results.