Traditionally, vibration analysis in operating conditions (on the road or on a bench) had to be combined with experimental modal analysis in controlled laboratory conditions in order to understand the modal behaviour of the structure. This requires additional measurements, costs and time. However, in many applications, the real operating conditions may differ significantly from those applied during the modal test and hence the vibration modes from the modal test might not be representative for the active modes in operation conditions. The need for a capability of doing a modal analysis on data from operating conditions is obvious.Over the last years, several modal parameter estimation techniques have been proposed and studied for modal parameter extraction from output-only data. Each method needs to make a number of assumptions and has some limitations. In this paper, the capabilities and limitations of the Least Squares Complex Exponential and two subspace techniques, referred to as the Balanced Realization (BR) and the Canonical Variate Analysis, are evaluated for its applicability to industrial cases.