To better make the sensor system of sluice floor voiding monitoring on soft foundation to achieve the best balance between costs and performance index, at the same time to solve the problem of traditional effective independence method and damage sensitivity method that cannot consider both modal linear independence and damage sensitivity, a multi-objective approach combined with effective independence method and damage sensitivity method is proposed to optimize the sensor arrangement.To solve the problem of modal data redundancy, the node similarity function is introduced. The influence of the sluice floor voiding on the modal parameters is explored from the aspect of hydrodynamics, and the modal flexibility is determined as the voiding sensitive characteristic quantity. Taking the maximum 2-norm of the Fisher information matrix as the goal, the maximum off-diagonal element of the MAC matrix is used as the evaluation index. The optimal arrangement of measuring points is then realized, and the results of the measuring points have been applied to the voiding inversion of the sluice floor. The results show that the meanrelative error of the recognition accuracy after the sensor optimization is decreased from 28.3%, 15.8%, 39.3% to 1%, 1.9%, 15.9%, respectively, indicating that the reliability of the method is verified. Keywords: sluice on soft foundation; floor voiding; optimal sensor arrangement; effective independence method; damage sensitivity method; modal flexibility 〖FL