The present work proposes an approach pertaining to finite element (FE) model to identify imbalance in the rotor systems. Conventional balancing methods involve a number of runs, i.e. initial run, a few trial runs and the final run. The proposed method eliminates the time-consuming trial run as the corresponding data is gathered from the FE model. Hence, accuracy of the FE model should be checked before using it in the identification process. The method requires the influence coefficient as determined from the FE model and the measured vibration response. The usefulness of the approach to predict the imbalance has been demonstrated using numerical simulation for a rotor system. Hence measured vibration response has also been simulated using the FE model of the system. Experimental conditions have been imitated by polluting the simulated vibration response with measurement noise. The method is capable of predicting imbalance magnitude as well as its location on the rotor disc at different speeds despite noise. The demonstrated approach shows that imbalance can be determined with lesser number of runs with good accuracy which ultimately saves time in balancing. Lesser number of trial runs help in reducing time of balancing the machine which is essential for any industry.