Spanning Tree-Based Algorithm for Generating Water Distribution Network Sets by Using Street Network Data Sets

Developing a water distribution infrastructure is a challenging task. Therefore, the presence of layout and design algorithms is crucial. In this work a water supply network layout generation algorithm is investigated, which is able to generate huge sets of case-unspecific water supply networks with different characteristics controlled by three parameters. As input as few as possible and easily accessible data sets are used. The algorithm is applied to an Alpine case study with 130,000 inhabitants. The generated water supply models are used as input for systematically testing a water supply network design algorithm. Results show that the investigated water supply network generation algorithm is able to create realistic networks with different characteristics, and they can be used as a benchmark set for newly developed water supply network design algorithms (e.g., pipe sizing) to identify weak points in the strategy, robustness, and runtime of the algorithm.