Developments in simulation techniques and algorithms combined with increased computational capacity have made it possible to simulate the dynamic behavior of complex machinery with a high degree of accuracy even in real time. In this chapter, the concept of the virtual sensor is explained: first on a general level, and then in more detail and by introducing a few case studies deploying different simulation tools. Combining actual measurements and the simulation model enables the use of simulation methods to estimate the dynamic behaviors of a machine in situations where measurements are not possible or economically feasible. There are two main categories of virtual sensors. The first, suitable for low complexity cases, is based on analytical models. The second is based on detailed numerical models. It is suitable for more complex cases. Sensor measurement can be further divided into online and offline measurements, the difference relating to measurement pace. From the business perspective, the virtual sensor can potentially provide better insight into a machine and its operation, and therefore bring added value that original equipment manufacturers can offer to end-users. Moreover, virtual sensors can lead to a reduction in the number of physical sensors needed, thereby reducing cost.