Vulnerability is the degree to which human and environmental systems are likely to experience harm due to a perturbation or stress. Vulnerability assessment under climate change needs a huge amount of simulations. The set of simulations needed depends on the uncertainties hypothesis. In this study, we propose an approach for vulnerability assessment with different types of design of experiments. We build up model of these designs, to identify a common pattern seen as metamodel, which models can be conformed to. In order to build up dedicated distribution platform for vulnerability analysis, we use a Model Driven Engineering approach.