In this work, we studied the influence of the variables temperature (T), water/glycerine ratio (R), and flow rate of the feeding water/glycerine solution (VW+G) on the non-catalysed steam reforming of glycerine. The experiments were carried out on a bench-scale equipment and the margins of the processing variables R, T, and VW+G were 0.7-33 wt wt(-1), 682-1018 degrees C, and 8.5-35.5 mL min(-1), respectively. The implementation of a Design of Experiment-Response Surface Methodology approach (DoE/RSM) allowed us to analyse the importance of each variable, as well as their interactions, in both the composition and the energetic features of the dry gas stream obtained. The temperature and the water/glycerine ratio played the principal role in determining the concentration of the main components (H-2, CO, CH4], and CO2) and the low heating value of the resulting dry gas stream. The effect of both variables was likely related with their influence on the thermodynamic equilibrium of the different reactions taking place (reforming, water-gas shift, and methanation reactions). Two variables were defined in order to evaluate the efficiency of the glycerine gasification: the steam-reforming efficiency (SRE) and the carbon gasification efficiency (CGE). On the other hand, the rate at which the energy can be supplied by the installation (LHV) was strongly affected by all the three processing variables and was mainly related with the volumetric flow rate of the dry gas stream, while the LHV played a secondary role. The predicted ranges of H-2, LHV, and LHV were 25.8-60.7%, 9.03-14.40 MJ N m(-3), and 0.47-5.26 kW, respectively. In all cases, high interactions between the processing variables were detected, putting in evidence the usefulness of the DoE/RSM approach. (C) 2017 Published by Elsevier Ltd.