Water Leakage detection

(Bojan Erker @ Mariborski vodovod &

Dušan Gleich @ University of Maribor)

+Logos Greed_website-36.png

NEED for AI:

Data from the flow sensors, placed on a water pipe, are collected using a remote system. Sensor’s data are sent over a wireless sensor network and are collected 2 times per month, because access points are not fully automated. The data are transferred into the information system and manually analyzed.



The remote sensing data can be used to detect wet water zones. By comparing water leakages and the results of the water consumption of the network the AI based reasoning for water leakage detection will be built up, which is then will be used to properly schedule the replacement of broken water pipes. Based on the gathered data, an automatic analysis of water pipe design types and/or suppliers shall be performed.



  • optimization of maintenance activities and water consumption

  • prediction of defects on a water-pipe system and reduction of water leakages

MARIBORSKI VODOVOD is the largest water supplier in Slovenia. Finding water leakeages is a problem of all water suppliers. Modern sensory system produces large amount of data, which is a challenge to process and analyze daily. In collaboration with UNIVERSITY OF MARIBOR  and use of AI technology we wish to detect water leakeage locations in the water supply system and improve maintenance.         

The goal of the experiment is to provide optimized schedule for maintenance activities by automatically detecting water leakages and consequntly optimize water supply system in a technical – economic way. Experiment is focused on industrial-manufacturing areas.