This paper presents an approach aiming to anticipate possible transformations of the overall built fabric of a town, analysed at building level. The research is applied to the city of Nice. A grid with 400m X 400m cells is applied over the commune's area. The field of study is made up of 163 cells concentrating half of the built-up area. The possibilities of transformation of buildings within a cell are estimated starting from hypotheses relating to four groups of variable: the structure of the built fabric, attractiveness linked to an infrastructure programme planned in the cell or nearby, the town's policy (urban renovation and climate plan) and residents' sensitivity to measures and practices in favour of sustainable development. The method consists in estimating the potential of change of the built fabric in each cell by means of fuzzy logic. We use fuzzy Inference Systems to address the problem of future spatial changes. Data are either quantitative or qualitative, in view of the fact that it is often not possible to describe the environmental and social properties (e.g. architectural quality, social acceptance…) using objective indicators. Thereby, the spatial knowledge is described by semantic variables often affected by a high degree of inaccuracy and uncertainty, making fuzzy logic a suitable tool for expressing and processing this information. The potential of change is assessed according to three scenarios: i) with no specific public policy in favour of sustainable development, ii) with an aggressive public policy against energy wasting in homes, iii) integrating the tendency of inhabitants to innovate and their level of sensitivity to levers set out by public policies. The cells' potentials are mapped. Some of them are selected for carrying out simulations of plausible transformations of the built fabric, and visualising in 3D the changed aspects of façades in certain streets.