Dynamic Virtual Infrastructure Benchmarking - DynaVIBe

The analyses of case studies are a well known instrument to identify problems and interactions of processes. In the field of urban water management the methodology is suited to evaluate new technologies, strategies or measures. However, data availability is limited and hence, new technologies, strategies or measures can only be tested on a limited number of case studies. The expenditure of time to build a detailed model of the water infrastructure of a midsize city can be quantified with several man-months. Moreover the results determined by the case studies are limited to specific boundary conditions and system properties and can hardly be generalized or transferred onto other boundary conditions. At the Unit of Environmental Engineering (IUT) at the University of Innsbruck recently a tool was developed for an algorithmic generation of virtual case studies for water supply and urban drainage systems (VIBe - Virtual Infrastructure Benchmarking). Via computer aided methods (agent based models, cellular automata approach etc) virtual water infrastructure systems and subsequent the input data for the subject-specific simulation software of such systems (EPANET, SWMM etc.) are generated. Following, the results of the simulation runs can be benchmarked respectively analysed stochastically. Thereby the understanding of real world systems can be increased and the potential of measures applied on infrastructure can be evaluated. While the approach is successful and also appreciated by the scientific community it allows investigating stationary systems only. The project described, denoted as DynaVIBe (Dynamic Virtual Infrastructure Benchmarking) is based on this existing efforts. However, it is not only an enhancement of VIBe but rather adds a complete new aspect to it which is fundamental for the further developments, fully new and innovative. In DynaVIBe it is intended to include the temporal dynamics to all processes, which represent real world urban development concerning technical and socio-economic aspects. Therefore the temporal change of infrastructure can be modelled. These development algorithms will take into account change processes in urban structures and their interactions with infrastructures. These changes can be initiated by population growth or decrease, changes of connected areas to the sewer system and the influence of new legal standards. This integrated approach in which furthermore socio-economic aspects as well as interactions over time are considered is a scientific novelty. DynaVIBe can also be applied to real world case studies in order to stochastically investigate e.g. development scenarios or climate change adaption strategies. It is expected that DynaVIBe allows indentifying optimal mechanisms regarding change and control of water infrastructure. The cooperation with the Monash University in Melbourne Australia targets at this aspect.

The analyses of case studies are a well-known instrument to identify problems and interactions of processes. In the field of urban water management the methodology is suited to evaluate new technologies, strategies or measures. However, data availability is limited and hence, new technologies, strategies or measures can only be tested on a limited number of case studies. The expenditure of time to build a detailed model of the water infrastructure of a midsize city can be quantified with several man-months. Moreover the results determined by the case studies are limited to specific boundary conditions and system properties and can hardly be generalized or transferred onto other boundary conditions. In line with the project "DynaVIBe" the Unit of Environmental Engineering (IUT) at the University of Innsbruck developed a tool for an algorithmic generation of virtual case studies for water supply and urban drainage systems including whole urban structures. Beside the generation and simulation of infrastructure systems on a spatial scale, also the temporal dimension within all simulations was established. Thus, it is possible to map the dynamics of urban water infrastructures into computing models and simulations. The implementation of the developed methods and algorithms is realized within the open source software "DynaMind". The strength of the algorithms is the flexibility to use input data at different level of details. Virtual case studies can be generated by using only few input parameters up to using detailed data sets (generating semi-virtual case studies). The usage of input data sets with different levels of detail can explicitly control the validity of model simulation results so that various research questions in the field of urban water management can be investigated. Due to the high flexibility of the generation algorithms also surrogate data can be used as input. This approach was applied on street network data sets which are of good quality and free accessible (e.g. OpenStreetMap). Further, the history of an urban infrastructure network can be recreated by extracting street network data out of historical orthophotos. It was proved that the DynaVIBe approach is also capable to be applied on real case studies for finding optimal mechanisms for developing and controlling urban water infrastructure systems.