Within this framework, the chapter by Mounce [1] investigates the opportunities
offered by data science in the context of smart water utilities. He discusses the role
played by digitalisation for smart water networks, analysing several aspects of IoT,
artificial intelligence, cloud computing, blockchain, and other new technologies.
The chapter explores relevant issues connected with data science applications to the
water industry. The first obstacle to the adoption of digital technologies is related to
the extraction of useful information from big datasets, being that the water industry
is generally considered as ‘data-rich but information poor’. Thus, collected data are
typically underused and data analytics are still not generally perceived as valuable,
in the road map to more efficient networks based on information and knowledge.
Currently, available computing power permits the implementation of data-driven
modelling and deep learning techniques for prediction and classification purposes.
The author foresees strong possibilities offered by deep artificial neural networks,
based on their excellent unsupervised feature extraction capabilities. Big datasets
concerning water quality generated by multiparametric sensor systems are given as
an example of relatively undeveloped sectors for data analytics, offering opportunities for further developments, which are discussed also in other chapters of this
book [3, 5]. Finally, the chapter provides reference and overview of case studies,
demonstrating the kind of applications that are candidates to be more commonplace
in the near future.
As pressure increases on water resources, there is a growing emphasis for water
service providers to minimise the loss from leakage. Optimal sensor placement in
water distribution systems for leak/burst detection and localisation is a wellestablished and very productive research field. Its primary focus is to minimise
the cost of a proposed sensor network infrastructure while maximising the capability to detect and localise leaks and bursts through the analysis of the collected data.
Romano [2] provides a systematic review of previous work covering relevant
articles published over the last decade aiming at rationalising the work carried
out in this field. The chapter presents a synthesis and analysis of the relevant
published works to: (1) provide insight and awareness of differing arguments,
theories, and approaches; (2) highlight their capabilities and limitations; (3) identify
the state of the art in their development. The chapter also provides insight and
awareness of differing approaches that have been proposed to tackle specific issues
encountered by researchers when developing their proposed techniques such as
model and measurement uncertainties. Trends and gaps in the current research and
future research directions are identified and discussed, and a number of considerations to promote further developments in this important field of research are
presented. This comprehensive chapter can serve as a useful reference resource
for researchers and practitioners involved in sensor network design for leak/burst
detection and localisation methodologies and in the development/adoption of these
techniques.
During recent decades, the role of data as a vital resource that enhances decisionmaking and which supports efficient systems operation has become evident, with a
growing number of water supply companies viewing data as a key organisational
aspect that has to be properly managed, instead of an operational side-product. At
x
Preface
offered by data science in the context of smart water utilities. He discusses the role
played by digitalisation for smart water networks, analysing several aspects of IoT,
artificial intelligence, cloud computing, blockchain, and other new technologies.
The chapter explores relevant issues connected with data science applications to the
water industry. The first obstacle to the adoption of digital technologies is related to
the extraction of useful information from big datasets, being that the water industry
is generally considered as ‘data-rich but information poor’. Thus, collected data are
typically underused and data analytics are still not generally perceived as valuable,
in the road map to more efficient networks based on information and knowledge.
Currently, available computing power permits the implementation of data-driven
modelling and deep learning techniques for prediction and classification purposes.
The author foresees strong possibilities offered by deep artificial neural networks,
based on their excellent unsupervised feature extraction capabilities. Big datasets
concerning water quality generated by multiparametric sensor systems are given as
an example of relatively undeveloped sectors for data analytics, offering opportunities for further developments, which are discussed also in other chapters of this
book [3, 5]. Finally, the chapter provides reference and overview of case studies,
demonstrating the kind of applications that are candidates to be more commonplace
in the near future.
As pressure increases on water resources, there is a growing emphasis for water
service providers to minimise the loss from leakage. Optimal sensor placement in
water distribution systems for leak/burst detection and localisation is a wellestablished and very productive research field. Its primary focus is to minimise
the cost of a proposed sensor network infrastructure while maximising the capability to detect and localise leaks and bursts through the analysis of the collected data.
Romano [2] provides a systematic review of previous work covering relevant
articles published over the last decade aiming at rationalising the work carried
out in this field. The chapter presents a synthesis and analysis of the relevant
published works to: (1) provide insight and awareness of differing arguments,
theories, and approaches; (2) highlight their capabilities and limitations; (3) identify
the state of the art in their development. The chapter also provides insight and
awareness of differing approaches that have been proposed to tackle specific issues
encountered by researchers when developing their proposed techniques such as
model and measurement uncertainties. Trends and gaps in the current research and
future research directions are identified and discussed, and a number of considerations to promote further developments in this important field of research are
presented. This comprehensive chapter can serve as a useful reference resource
for researchers and practitioners involved in sensor network design for leak/burst
detection and localisation methodologies and in the development/adoption of these
techniques.
During recent decades, the role of data as a vital resource that enhances decisionmaking and which supports efficient systems operation has become evident, with a
growing number of water supply companies viewing data as a key organisational
aspect that has to be properly managed, instead of an operational side-product. At
x
Preface