This paper presents a new sequential method to calibrate a material model in the presence of significant material variability. Material variability is handled in the mixed-effects framework. In this approach, material variability is described by a probability distribution calibrated by maximizing a likelihood. Yet, when the number of model parameters or the computational time of a single run of the models increases (for multiaxial models for instance), the maximization of the likelihood of mixed-effects is more difficult to perform. Furthermore, the parameters do not have the same influence on the material model depending on the nature of the test. The proposed procedure enables to calibrate the model on multiple experiments. It relies on the definition of a sequence of calibration subproblems. Associated to the relevant experimental data, each subproblem allows to calibrate the joint distribution of a subset of the model parameters. The maximization of the attached likelihood is eased as the number of unknown parameters is reduced compared to full problem. The subproblems are solved sequentially. To ensure consistency of the global process, the research space for the distribution parameters already estimated with a previous calibration is restricted to a trust region. The proposed calibration process is applied to an orthotropic elastic model with laminates made from T700GC/M21 base ply material. The ability of the procedure to sequentially estimate the model parameters distribution is investigated. Its capability to ensure consistency throughout the calibration process is discussed. Results show that the proposed procedure is a promising methodology to handle the calibration of complex material models in the mixed-effects framework.