the process of reconfiguring the urban fabric aims to step beyond anthropocene preconceptions by merging human and machine  intelligence in order to create novel urban compositions which are designed purely as a reaction to its contextual climatic parameters. the output generates often unexpected, even radical urban proposals seen as futile inspiration to the planning process.

each site that the workflow is applied to is automatically analysed with the help of opensource climate data, realtime satellite imagery and socioeconomic statistics . this dataset becomes the backbone for all contextual suggestions by the system.

in order to guide the urban_GAN network into contextually reactive  outcomes, the context is encapsuled as a three-dimensional datascape which aims to summarize the key social, functional and environmental parameters of the site. to iterate through the solution space of a site's urban adaptability potential, different informations of the site are prioritized in their relevance to the current environmental disbalance and this is then fed as input into the algorithm. the output data is automatically translated into synthetic satellite images to illustrate the options.