Sensing machine elements offer the potential to upgrade conventional machine elements by extending their primary function to be able to measure a quantity of interest at its point of origin in a technical system, the so-called in situ measurement. To ensure the functionality of these next generation machine elements, special attention must be paid to uncertainty in terms of modelling to be able to correctly evaluate the provided signal and obtain reliable information. Consequently, this contribution describes an approach to classify uncertainty in sensing technology, especially in SMEs, based on the amount of available information, which can be used as a point of departure to reduce the impact of occurring uncertainty to improve the robustness of the obtained signal. Starting from the understanding of uncertainty and the corresponding classification scheme as well as its linkage to robust design from the Collaborative Research Centre 805, a quantitative model is presented to determine the impact of uncertainty on a measuring signal. The applicability of the proposed approach is demonstrated using the example of a sensing timing belt by taking into account the uncertainty from the SME itself and also from the surrounding technical system.