Digital twins are tools originally developed in the world of industry to study objects or technical systems through simulation in order to reduce the cost of experimentation. Indeed, at first sight, this approach seems promising in the field of urban planning, especially since full-scale experimentation is generally impossible. Therefore, when considering scenarios for future development, it would be interesting to explore their dynamic behaviour by confronting current results with real data. Digital twins can be used in the context of urbanism to support planners, but also for participatory planning, discussions with local politicians, etc. (Yamu, 2023). Here, we consider an approach that aims to support planners in the context of an analysis of an existing urban pattern with a view to future development. It is based on a zoning approach that links in a coherent way a wide range of scales, ranging from the metropolitan scale down to that of urban blocks, which is not currently available in existing digital twin models of cities. The simulator is primarily aimed at developers, but it can also be used in the context of resident participation. Reality is taken into account through the footprint of buildings, their height, networks, non-buildable areas and all types of amenities. It allows the design of development scenarios from the metropolitan scale down to the scale of urban blocks, the effects of which can then be simulated by the tools integrated or interfaced with a digital twin. This approach to semantising space uses spatial modelling based on a system of interlocking scales inspired by multifractal geometry. It allows us to distinguish between areas of active development and areas to be preserved. The latter are the urban spaces crossed by green corridors used for recreational activities, but also for peri-urban agriculture, for the ventilation of urbanised areas and for the maintenance of an interconnected system of natural areas to preserve biodiversity. The zoning is considered through an approach that refers to the theory of central places, which distinguishes different levels of services according to their importance and frequency for the resident population. In addition, the different zones are characterised, at different scales, by measures of accessibility to amenities as well as to different types of leisure spaces. The parameterisation is based on an exploitation of urban planning documents of master plans and allows the calculation of the satisfaction level of the inhabitants according to the importance of the amenities. This satisfaction is calculated for each zone and aggregated into a general satisfaction level. Moreover, it is possible to distribute housing according to the same logic of interlocking scales, promoting a diversity of densities at all scales, while respecting the penetration of green and blue networks in the urban blocks. Hence, the model allows on the one hand to reproduce the spatial organisation of an existing city while considering the needs of residents. For this reason, it can be considered as a digital twin of the city. The approach is structural and in this sense static, as the goal is not considering, for example, daily mobilities or the impact of climate. But the concept allows evaluating the positive or negative impact of new urbanistic projects as the development of new residential areas by adding dwellings or by adding diverse types of amenities. Shrinking cities can be considered, as well. Thus, potential future development can be evaluated in the spirit of long-term dynamics. The algorithmic formalisation is rather transparent, thus avoiding a ‘'black box’ effect.
Read full abstract