The decision making process related to urban areas synthetizes a lot of data taking into account numerous parameters. The indicators produced from simulations or data mining on 3D city models (3DCM) can be a great help into these decision-making processes. The problem is to acquire and process the 3DCM in order to produce these indicators at city scale. The EnVIE project proposes a framework to perform the acquisition of 3DCM and then to produce 3D or 2D metadata on these models. As several city councils have been interviewed in the scope of the project to describe their real needs, the computed metadata are directly usable into the decision-making process. The main contribution of this framework is to merge existing techniques to build a pipeline allowing to compute lighting and wind simulation, and to extract socio-economic metadata with data-mining techniques. This on-going work builds a link between indicators at the building scale and 2D indicators at larger scale from GIS, and therefore is a step towards an indicator production process at full-scale range.