data quality and/or increased data collection and even online instrumentation.
However, regulation in the UK is encouraging this with companies in the future
seeking a basis for their asset planning with the analysis of data from, or directly
related to, their operations. Where good quality data is available, accessible and
well maintained, the ability of the distribution engineer to monitor, evaluate
and make good decisions with regard to the operation and maintenance of the
network is greatly enhanced. However this is rare, and there is a need to aid the
decision-making process to make the best possible use of the data that is available.
Datasets currently maintained by water companies include historic and updated
asset records, discrete water quality sampling and associated laboratory analysis
and continuous/online (typically hydraulic only) data collection from an ever
increasing telemetry footprint. Companies also keep records of customer contacts
of various types (such as complaints about leakage or water discoloration); however
such data is very variable in nature and information richness as it depends on
non-standardised customer behaviour. Good procedures for formatting and using
that data should then be implemented by WSPs. In order, the key data used in the UK
to inform decision-making processes are hydraulic meter data, customer contacts,
water quality sample data and analysis and network data such as asset records,
burst records and pipe samples. Currently interventions are often responsive to
customer contacts leading to a reactive management that does not necessarily
deal with the underlying issues. An interim position currently exists, where water
companies have substantial databases but lack the connectivity and deployed
methods to extract the full potential benefit. There is a wealth of data at works/
reservoir outlets but currently less intelligence on what is happening in the
distribution network. Customers should not be acting as ‘surrogate telemetry points’,
which is often the case (e.g. to help with leakage detection or water quality issues).
Monitoring and control systems play a role in the daily operation and maintenance
of water supply and sewerage facilities.
One of the main issues in the water industry is the lack of linking together of
data as required for identifying/solving problems occurring at system interfaces.
Current practice is for the collection and maintenance of data in disparate corporate
systems, severely limiting the potential for deeper understanding. Utility databases
containing data in an unprocessed format do not lend themselves to analysis to
establish relationships or trends. The databases are generally on separate platforms,
in differing formats, with non-uniform IDs and contain many unpopulated or
utility-specific fields. Often there has been too much focus on technology and
IT infrastructure and not enough on improving data quality and data integration
and making best use of existing corporate data resources. An example of this has
been the UK water industry gradual move to enterprise resource planning solutions
(with platforms such as SAP and MAXIMO) whereby all of the water utilities’
business information is coordinated into a single environment and similarly to
enterprise infrastructure for management of real-time data and events (platforms
such as OSI-PI). Large implementations of these platforms can take (many) years
for large companies, and customisation to allow practical use can substantially
increase implementation times [4]. Hence, many legacy systems are not yet migrated
4
S. R. Mounce
However, regulation in the UK is encouraging this with companies in the future
seeking a basis for their asset planning with the analysis of data from, or directly
related to, their operations. Where good quality data is available, accessible and
well maintained, the ability of the distribution engineer to monitor, evaluate
and make good decisions with regard to the operation and maintenance of the
network is greatly enhanced. However this is rare, and there is a need to aid the
decision-making process to make the best possible use of the data that is available.
Datasets currently maintained by water companies include historic and updated
asset records, discrete water quality sampling and associated laboratory analysis
and continuous/online (typically hydraulic only) data collection from an ever
increasing telemetry footprint. Companies also keep records of customer contacts
of various types (such as complaints about leakage or water discoloration); however
such data is very variable in nature and information richness as it depends on
non-standardised customer behaviour. Good procedures for formatting and using
that data should then be implemented by WSPs. In order, the key data used in the UK
to inform decision-making processes are hydraulic meter data, customer contacts,
water quality sample data and analysis and network data such as asset records,
burst records and pipe samples. Currently interventions are often responsive to
customer contacts leading to a reactive management that does not necessarily
deal with the underlying issues. An interim position currently exists, where water
companies have substantial databases but lack the connectivity and deployed
methods to extract the full potential benefit. There is a wealth of data at works/
reservoir outlets but currently less intelligence on what is happening in the
distribution network. Customers should not be acting as ‘surrogate telemetry points’,
which is often the case (e.g. to help with leakage detection or water quality issues).
Monitoring and control systems play a role in the daily operation and maintenance
of water supply and sewerage facilities.
One of the main issues in the water industry is the lack of linking together of
data as required for identifying/solving problems occurring at system interfaces.
Current practice is for the collection and maintenance of data in disparate corporate
systems, severely limiting the potential for deeper understanding. Utility databases
containing data in an unprocessed format do not lend themselves to analysis to
establish relationships or trends. The databases are generally on separate platforms,
in differing formats, with non-uniform IDs and contain many unpopulated or
utility-specific fields. Often there has been too much focus on technology and
IT infrastructure and not enough on improving data quality and data integration
and making best use of existing corporate data resources. An example of this has
been the UK water industry gradual move to enterprise resource planning solutions
(with platforms such as SAP and MAXIMO) whereby all of the water utilities’
business information is coordinated into a single environment and similarly to
enterprise infrastructure for management of real-time data and events (platforms
such as OSI-PI). Large implementations of these platforms can take (many) years
for large companies, and customisation to allow practical use can substantially
increase implementation times [4]. Hence, many legacy systems are not yet migrated
4
S. R. Mounce