A small and low-cost re-entry vehicle can be a good means for doing hypersonic research, testing new heat-resistant materials, and qualifying newly developed subsystems in a realistic environment. To establish the optimal vehicle shape a response-surface methodology using design-of-experiments techniques is proposed. With these techniques the effects of changing several geometric design parameters in an ‘all-at-the-same-time’ approach can be studied, instead of the more traditional ‘one-at-a-time’ approach. Each of the design iterations includes an aerodynamic analysis based on the Modified Newtonian method and a threedegrees-of-freedom trajectory analysis. Generating response surfaces for each of the performance indices and optimising them with a multi-objective optimisation method, a set of geometric parameters is found that gives the best alternative for each of the performance indices. Two fundamentally different vehicle shapes are considered, i.e., one based on a trapezoidal cross section and a sharp, watercooled nose, for an increased lift-to-drag ratio, and one being a blunted bi-cone that is simple to manufacture, has good stability properties and good potentials for various aerodynamic and material experiments. The developed methodology leads to significant insight in the design space and provides sub-optimal vehicle shapes at a limited computational cost. It may serve as a good starting point for more detailed analysis of a sub-region of the original design space.