Transfer of Science or Research, Cooperation agreements

Within a 3-year project, the city of Lausanne will further develop and use the innovative tool SEMAplus for forecasting deterioration of the sewer and water networks. The software solution which helps to maintain the water systems for future generations has been developed by the Berlin Centre of Competence for Water (KWB) and Berliner Wasserbetriebe. The former is a Berlin-based non-profit research organisation providing innovative, practical solutions to water management challenges, while the latter is the largest municipal water supply and wastewater disposal company in Germany. SEMAplus is currently used in Berlin for over 10,000 km of sewer network, to optimise the annual financial planning of sewer rehabilitation.

Ageing water and sewer networks require enormous investments for rehabilitation, which represents a substantial share of municipal budgets. The machine learning based software solution SEMAplus provides rapid and accurate information to localise urgent rehabilitation needs as well as a foundation for long-term investment planning. To accomplish this, SEMAplus requires only a limited amount of inspection (visual inspection of the sewer pipes by CCTV cameras), technical (age, material, diameter, etc.) and environmental (traffic load, groundwater level) data.

In Lausanne, the existing modules of SEMAplus will be implemented and completed to address specific needs. The main objective is to provide a tool which integrates asset management from the collection of data in the field (CCTV inspection of sewer or localisation of pipe failures) and prioritises pipe replacement. SEMAplus will be extended through a new module that automatically rates the conditions of the pipes based on the visual sewer inspection reports by CCTV cameras. The goal is to obtain a single condition score for each pipe that will be used with sewer characteristics and relevant environmental factors to simulate the failure probability of individual sewer pipes. The result is a list of all inspected and uninspected sewer pipes, ranked and sorted by their immediate need for rehabilitation.

New machine learning techniques will also be explored to improve the accuracy of our predictions. Modelling accuracy is of high relevance here, as results can be used immediately to decide on local construction. A specific module will also be developed for prioritising the rehabilitation needs of pipes according to the risk of failure. This analysis is performed in close collaboration with experts from the French research institutes Institut National de la Recherche Agronomique (INRAE) and Institut National Des Sciences Appliquées Lyon (INSA). It will allow additional impact or vulnerability criteria to prioritise rehabilitation investment (e.g. under high-traffic streets or in resource protection areas) to be considered.

The existing tool for the asset management of the drinking water network of the city of Lausanne, based on the same philosophy, will additionally be upgraded and integrated into SEMAplus.

Finally, at a strategic level, SEMAplus will be used to simulate the evolution of the network’s condition over the next decade. The goal is to visualise the deterioration or improvement of the network’s condition, depending on the annual investments for sewer rehabilitation and the type of rehabilitation techniques used (e.g. more repair, renovation or replacement). Considering the price of the investments, this will give new leverage to the City of Lausanne to justify the relevance of the proposed investments and convince municipalities to invest purposefully in sewer rehabilitation.

In the end, for the city of Lausanne, the purpose of using and developing SEMAplus is to have an integrated tool for the asset management of sewer and water networks.

More information on SEMAplus is available here: https://www.kompetenz-wasser.de/en/newsroom/news/semaplus-lausanne

---

Wissenschaftlicher Ansprechpartner:
Dr. Nicolas Caradot