to such new systems. And when data is migrated to such systems, integration
and techniques to derive information are at often overlooked. Better use of data
needs to be made by bringing datasets together, particularly in the development
and use of metadata models. Essentially, data that is in many silos needs to be
located and pulled together and (sometimes) missing data accounted for before
sophisticated analysis can be achieved. Most large utilities today have an EAM
(enterprise asset management) or CMMS (computerised maintenance management
system) in place for network operational processes. In the future, utilities will
be able to move beyond time-based to condition-based maintenance, so by adopting
the ability to understand the effective age of their assets and then forecasting
potential failures, they will be able to identify and schedule improvements in life
extension maintenance activities as well as strategically plan for their replacement
in their long-term asset plan.
With advances in data manipulation (such as the use of metadata and
format interoperability) and analysis systems, in particular the integration of GIS
information with data mining methodologies, it is now possible to explore
relationships between data in increasingly sophisticated ways. If data is available,
is of good enough quality and can be linked and associated with other data, it can,
in principle, be used for a multitude of applications such as business analytics,
problem ‘hotspot’ mapping, operational assessment and investment modelling.
1.2 Big Data and Analytics Opportunities
The availability and affordability of varying forms of sensing, smart systems, data
storage and transmission technology means water utilities are becoming able to
collect more data than ever before. This information revolution era opens up hereto
unseen possibilities in the creation of tools for future engineering application. Water
utility databases are currently growing rapidly and will continue to do so. Globally,
IBM [5] estimates that 2.5 quintillion bytes of data per day are being collected.
In fact, more than 50% of the world’s data was created last year, but less than 0.5%
was analysed or used. Experts predict that 40 zettabytes of data will be in existence
by 2020. Collecting more data doesn’t necessarily result in better information or
knowledge. But, substantial datasets do offer a potential way to tackle traditional
issues via development and application of novel data-driven analysis. Big data
has been compared to being like an iceberg where most of the value to be unlocked
is still hidden under the surface.
Machine learning (ML) relates with the study, design and development of the
algorithms that give computers the capability to learn without being explicitly
programmed. Problems usually need describing in features to use ML. Machine
learning and big data can be used to ‘learn’ how the system operates and reacts to
events. The combination of big data and machine learning will eventually result
in a breakthrough in the integration and analysis of heterogeneous data types as
is already occurring in other more developed industries. Big data analytics lies at
Data Science Trends and Opportunities for Smart Water Utilities
5
and techniques to derive information are at often overlooked. Better use of data
needs to be made by bringing datasets together, particularly in the development
and use of metadata models. Essentially, data that is in many silos needs to be
located and pulled together and (sometimes) missing data accounted for before
sophisticated analysis can be achieved. Most large utilities today have an EAM
(enterprise asset management) or CMMS (computerised maintenance management
system) in place for network operational processes. In the future, utilities will
be able to move beyond time-based to condition-based maintenance, so by adopting
the ability to understand the effective age of their assets and then forecasting
potential failures, they will be able to identify and schedule improvements in life
extension maintenance activities as well as strategically plan for their replacement
in their long-term asset plan.
With advances in data manipulation (such as the use of metadata and
format interoperability) and analysis systems, in particular the integration of GIS
information with data mining methodologies, it is now possible to explore
relationships between data in increasingly sophisticated ways. If data is available,
is of good enough quality and can be linked and associated with other data, it can,
in principle, be used for a multitude of applications such as business analytics,
problem ‘hotspot’ mapping, operational assessment and investment modelling.
1.2 Big Data and Analytics Opportunities
The availability and affordability of varying forms of sensing, smart systems, data
storage and transmission technology means water utilities are becoming able to
collect more data than ever before. This information revolution era opens up hereto
unseen possibilities in the creation of tools for future engineering application. Water
utility databases are currently growing rapidly and will continue to do so. Globally,
IBM [5] estimates that 2.5 quintillion bytes of data per day are being collected.
In fact, more than 50% of the world’s data was created last year, but less than 0.5%
was analysed or used. Experts predict that 40 zettabytes of data will be in existence
by 2020. Collecting more data doesn’t necessarily result in better information or
knowledge. But, substantial datasets do offer a potential way to tackle traditional
issues via development and application of novel data-driven analysis. Big data
has been compared to being like an iceberg where most of the value to be unlocked
is still hidden under the surface.
Machine learning (ML) relates with the study, design and development of the
algorithms that give computers the capability to learn without being explicitly
programmed. Problems usually need describing in features to use ML. Machine
learning and big data can be used to ‘learn’ how the system operates and reacts to
events. The combination of big data and machine learning will eventually result
in a breakthrough in the integration and analysis of heterogeneous data types as
is already occurring in other more developed industries. Big data analytics lies at
Data Science Trends and Opportunities for Smart Water Utilities
5