opportunities and barriers) and smart wastewater treatment (including preliminary,
primary/secondary treatment and sludge and resources recovery processes).
Online monitoring of several common parameters of water quality is used in
water distribution systems, in order to ensure its safety for drinking and sanitation.
The most common parameters are free chlorine, turbidity, and pH. A water quality
event will typically result when one or more parameter values reach abnormal
levels. Detection of water quality events before customers are affected is paramount
to prevent possible public health impacts and potential regulatory action. Several
general methods have been suggested in the past for identifying and classifying
such water quality events from multiple parameters. These methods include supervised methods such as regression or regression trees and methods that make use of
unsupervised learning such as clustering. The chapter by Brill [5] presents and
demonstrates the utilisation of radial basis function as a tool for detection and
classification of abnormal events in water quality. The methodology is based on
calibration of a radial basis function using historical true events classified by human
experts. The aim of the process is the selection of parameters that ensure zero falsenegative events. The chapter continues to describe the main method of using radial
basis function and then compares four different kernel functions, which are used for
implementing the radial basis function. The case study part of the chapter illustrates
actual analysis of real-world data (obtained from a monitoring station located in a
large city) as well as an illustrative example (data originating from a laboratory rig).
The chapter concludes with some practical advice on how kernel functions should
be selected for this task and will be of value to practitioners implementing their own
water quality alert systems.
The first five chapters of this book show a general portrait of the many implications of data science with the water industry, in particular for what regards specific
monitoring demands and also, more in general, when dealing with big datasets of
heterogeneous parameters. The potentialities and the opportunities offered by the
information and communication technologies (ICTs) for improving the management of water are globally recognised. However, ICT solutions are not well
exploited in developing countries and for solving this issue a partnership approach,
based on open innovation, is necessary. Mvulirwenande and Wehn [6] show how
ICT-focused water innovation partnerships (ICT-WIPs) can play a relevant role in
building the capabilities of developing countries to implement smart water systems.
In particular, this study demonstrates that ICT-WIPs allow a variety of stakeholders
in the water sector (such as municipalities, ministries, large utilities, and regulatory
agencies) to work together and increase the awareness about the potential of smart
water systems. At the same time, the innovation partnership approach promotes the
culture of mutual learning, thus allowing partners to strengthen each other’s
innovation competences relating to smart water systems in developing countries.
This is particularly true when, as in the case of the ICT-WIPs analysed in this study,
partners from foreign countries need the knowledge and experience of local partners (e.g., about specific problems concerning the water systems, existing solutions
and their weaknesses, possible additional local risks). Finally, the nature of the ICTfocused water innovations analysed in this study leads to the insight that fostering
xii
Preface
primary/secondary treatment and sludge and resources recovery processes).
Online monitoring of several common parameters of water quality is used in
water distribution systems, in order to ensure its safety for drinking and sanitation.
The most common parameters are free chlorine, turbidity, and pH. A water quality
event will typically result when one or more parameter values reach abnormal
levels. Detection of water quality events before customers are affected is paramount
to prevent possible public health impacts and potential regulatory action. Several
general methods have been suggested in the past for identifying and classifying
such water quality events from multiple parameters. These methods include supervised methods such as regression or regression trees and methods that make use of
unsupervised learning such as clustering. The chapter by Brill [5] presents and
demonstrates the utilisation of radial basis function as a tool for detection and
classification of abnormal events in water quality. The methodology is based on
calibration of a radial basis function using historical true events classified by human
experts. The aim of the process is the selection of parameters that ensure zero falsenegative events. The chapter continues to describe the main method of using radial
basis function and then compares four different kernel functions, which are used for
implementing the radial basis function. The case study part of the chapter illustrates
actual analysis of real-world data (obtained from a monitoring station located in a
large city) as well as an illustrative example (data originating from a laboratory rig).
The chapter concludes with some practical advice on how kernel functions should
be selected for this task and will be of value to practitioners implementing their own
water quality alert systems.
The first five chapters of this book show a general portrait of the many implications of data science with the water industry, in particular for what regards specific
monitoring demands and also, more in general, when dealing with big datasets of
heterogeneous parameters. The potentialities and the opportunities offered by the
information and communication technologies (ICTs) for improving the management of water are globally recognised. However, ICT solutions are not well
exploited in developing countries and for solving this issue a partnership approach,
based on open innovation, is necessary. Mvulirwenande and Wehn [6] show how
ICT-focused water innovation partnerships (ICT-WIPs) can play a relevant role in
building the capabilities of developing countries to implement smart water systems.
In particular, this study demonstrates that ICT-WIPs allow a variety of stakeholders
in the water sector (such as municipalities, ministries, large utilities, and regulatory
agencies) to work together and increase the awareness about the potential of smart
water systems. At the same time, the innovation partnership approach promotes the
culture of mutual learning, thus allowing partners to strengthen each other’s
innovation competences relating to smart water systems in developing countries.
This is particularly true when, as in the case of the ICT-WIPs analysed in this study,
partners from foreign countries need the knowledge and experience of local partners (e.g., about specific problems concerning the water systems, existing solutions
and their weaknesses, possible additional local risks). Finally, the nature of the ICTfocused water innovations analysed in this study leads to the insight that fostering
xii
Preface