The purpose of this work is to investigate the removal of phosphate in a fixed-bed column, using a layered double hydroxide (CaMgAl-LDH) as adsorbent, applying experimental observation, mathematical simulation and statistical analysis. Also, an experimental design was carried out in order to evaluate the influence of the operational conditions applied in the column on its adsorption efficiency. At a natural pH of 7.5, three independent parameters were considered: initial phosphate concentration (15, 30, and 60 mg L-1); feed flow rate (2.5, 5, and 10 mL min−1); and solid adsorbent mass (0.2, 0.5, and 1 g). As response variables the saturation time and the concentration of saturation in the column were evaluated. For the study of the fixed-bed column performance, through its breakthrough curves, a mathematical simulation, applying Bayesian statistics, was built in order to estimate its parameters, these being: the Peclet number (Pe), the equilibrium Langmuir constant (KL) and the kinetic coefficient (KS). To illustrate the model that was used to describe the mass transport through the fixed-bed, differential partial equations were obtained by the mass balance of the column. The results obtained in this research demonstrated that the Peclet number, for all operational conditions, suggests that the dispersion and the advection forces present in the column are approximately equal to each other. Also, the Langmuir isotherm represented well the adsorption equilibrium in all operational conditions and the kinetic coefficient ranged between 4.52 and 0.81 mL mg−1 min−1, these values shows that the kinetic effect has little interference in the breakthrough curves evaluated in the present work. Then, the estimated parameters were validated by the experimental design applied. Therefore, it was shown that the technique used to estimate the parameters and the mass balance model applied in this work, to describe the adsorption process, was robust, since the model was able to predict the breakthrough curves. In this way, this approach can be used to design columns based on the desired operational conditions, as well as for scaleup studies.