This paper is interested in model updating problems which consists in identifying optimal values of model parameters by comparing the model outputs with the experimental outputs. Such a problem generally yields a challenging multivariate inverse problem to be solved in high dimension. The high-dimensionality requires the use of a global optimization algorithm in order the explore efficiently the parameters space. In this paper we propose an alternative algorithm which allows each model parameters to be identified separately and sequentially by solving separated univariate inverse problems. For each parameter, a devoted inverse problem is introduced by identifying an output which is sensitive to this parameter only, the sensitivity being quantified using Sobol indices. The proposed method is illustrated through a three-storey structure for which experimental measurements are collected.