The management of drinking water abstraction assets is a complex and extensive core activity of utilities. Well rehabilitation represents major annual investments and expenses to maintain the service quality. Wells are usually maintained based on condition and capacity data obtained from pumping tests (water levels, discharge rates) and CCTV inspections (ageing, corrosion, etc.). A decrease below 80% of initial capacity or a bad physical condition are seen as triggers for maintenance. Scheduling of monitoring activities is typically based on long-term practical experience of waterworks staff. Over the years, regular intervals were established, enabling time-based maintenance of single wells. Utilities are lacking methodologies to prioritize the operation and maintenance and could benefit from the increasing use of sensors and loggers leading to higher amounts of available data at the right time and location. Also, solutions must be viable and attractive to field staff and lead to a reduction in work load, instead of adding complexity to the process.
Limitations of current practices
Well and groundwater data are collected by various stakeholders (water utility, environmental authority). They describe the maintenance, operation and condition of the assets and the aquifer and are usually available in several databases or in paper form. Field work is still often recorded offline and manually integrated in the company databases. Utilities are lacking solutions to support the efficient and secure data collection on-site. They might consider using digital devices and applications to improve the efficiency of the procedures but remain concerned with cyber-security issues linked to critical infrastructure.
Efficient data integration in the utilities’ and authorities’ data management systems is the key to allow further data analysis such as semi-automated well condition assessment and groundwater level monitoring. The solution will facilitate the use of different sources of data for field works and automatically validate and transfer the collected data to the utility database(s). Having such data readily available will reduce OPEX by accelerating the maintenance procedures and focusing activities on wells with highest maintenance needs. It will improve decision making by using available data and machine learning algorithms for automatic well condition assessment and the elaboration of cost-efficient maintenance plans to efficiently allocate long-term investments for well rehabilitation and replacement. This tool will be developed as a modularized solution with open source components, building on existing tools for long term management sewer and drinking water networks.
The solution is deployed in Berlin to improve the data collection and management of drinking water wells. It is implemented at the city scale (about 650 groundwater wells) with a focus on wells at risk (shallow, unconfined, bank filtrate wells).
VRAGMENTS – Tino Breddin : email@example.com
VRAGMENTS – Ronny Esterluss : firstname.lastname@example.org
VRAGMENTS – Stephan Gensch : email@example.com