project

Data mining techniques for water utilities

Exploratory research into data mining

Expert(s):
Peter van Thienen PhD, Dirk Vries PhD MSc, Joost van Summeren PhD, Erwin Vonk MSc, Martin Korevaar MSc PhD, Henk-Jan van Alphen MSc

  • Start date
    20180101
  • End date
    20190331
  • Principal
    Bedrijfstakonderzoek
  • collaborating partners
    Waterbedrijf Groningen, WLN, Oasen, PWN, Phinion

The promises of data mining techniques have long been very inviting – and for the water utilities as well. In this research we will lay out the most important techniques and produce an overview of the extent to which they are already being implemented, both in and outside the water sector. We will also collect the experiences of and points highlighted by practitioners, both in and outside the water sector.

In this project’s second phase, three implementation pilots of data mining techniques will be carried out at water utilities, with the aim of determining the key success/failure factors in the implementation of data mining at water utilities.

Overview of techniques and applications

There have been numerous recent developments in the field of big data mining, which present techniques for the extraction of knowledge from databases. These developments create new possibilities for the implementation of these techniques in the water sector as well. In this project an extensive literature study will be carried out as a basis for an overview of the available data mining techniques and approaches. This includes an examination of the extent to which these techniques are already being implemented in water utilities or elsewhere. The techniques or datasets that hold promise for new applications will also be identified.

Experience of practitioners

Interviews will be conducted with practitioners, both in and outside the water sector, to help identify the biggest challenges in the implementation of data mining techniques.

Success/failure factors

Lastly, three implementation pilots of data mining techniques will be carried out at Waterbedrijf Groningen (jointly with WLN), Oasen and PWN (jointly with Phinion). The aim is to determine the key success/failure factors in the implementation of data mining at water utilities.