Strong population growth and the resulting urbanisation are increasing the global consumption of material resources and energy. The construction sector is responsible for 60% of extracted raw materials and also generates 40% of energy-related CO2 emissions. In Austria, waste from the construction sector accounts for around 70% of total annual waste generation - facts that emphasise the urgent need for recycling measures. The existing building stock has great potential to serve as a raw material reservoir, but there is currently a lack of holistic knowledge about the building stock, which is the biggest obstacle to the reuse and recycling of materials and elements.

The main objective of BIMstocks is to develop a methodology for the continuous digital recording of the material composition of the building stock for the purpose of modelling the secondary raw material register and predicting recycling potentials by creating a BIM object catalogue for typical existing buildings in Vienna, generating as-built BIM models and subsequent upscaling to city level. Around 10 use cases, which cover a large part of the residential buildings typical for Vienna, are to be recorded in order to enable upscaling to city level. The final goal is to generate a GIS-based urban mining platform that embeds the information obtained from the individual use cases and forecasts the recycling potentials, material flows and waste masses. In addition, a framework is to be developed that enables the implementation of urban mining strategies. The framework will describe all individual steps and the methods used. The project thus represents the continuation of the framework developed in the SCI_BIM research project for the integral determination of geometry and material by coupling laser scanning and GPR technology for the purpose of semi-automated BIM model creation. SCI_BIM showed that the GPR technology needs further testing to a) apply to different construction types and b) build a material database that would significantly increase the efficiency of material identification.

The innovation of the project lies in the coupling of different technologies that enable scaling from component level to city level: Recording technology using GPR, the application of machine learning for the purpose of automated determination of material composition, and predictive modelling at city level in the digital urban mining platform. For the first time, the uncertainties resulting from the use case samples, the measurements and the extrapolation will also be estimated. The intended result, based on GPR images and subsequent machine learning algorithms, is the creation of a component catalogue for typical residential buildings in Vienna, which enables upscaling to city level, as well as the embedding of the components or buildings in the GIS-based urban mining platform.

The main benefit of the results generated from BIMstocks is the increase in recycling rates through the application of urban mining strategies, for which the generated public urban mining platform serves as a basis.