Congress Resources: Papers, posters and presentations

< Return to abstract list

Frontline Information Technology

Congress: 2015
Author(s): Elizabeth Burlon, Rohit Banerji

Keyword(s): Sub-theme 10: Management of water resources,
AbstractIntroduction Water utility companies across the UK and Europe are faced with the challenge of improving business performance by being more proactive. Reactive regimes typical of the today's ways of working come with avoidable costs and service quality failures. While most water companies have had the real time data to act more proactively for many years, it is only recently with the advent of Big Data technologies, that the opportunity to put it into action has become affordable. Currently, water utility operations are largely reactive. Assets are run until an issue is reported through a remotely monitor alarm, or, in the case of sewage pumping stations, a pollution event. As monitoring and modelling technology becomes more sophisticated and companies are able to gather more frequent and better quality data about their assets and networks, operators are presented with an opportunity to use this data to drive insights which allow proactive interventions. Industry leaders are questioning how smart metering will change the face of retail utilities , however this study proposes that there is additional opportunity in using data from network sources to intervene before or soon after failure by exploiting the survival window on asset failure -- reacting faster through the use of Big Data will save valuable time to mitigate consequences and impact to the dynamic system as well as customers served by the network. Methods The hypothesis was tested through the design and implementation of a near-real time monitoring and modelling software at a major UK water utility to monitor the performance of wastewater catchment network and it's receiving sewage treatment works. The team worked with a variety of roles in the business involved in the management of the catchment to understand their requirements for a dynamic monitoring system and how it would allow them to work in a new, more proactive way. The requirements gathered formed the basis of the design of the new software. Data was integrated into a single Big Data platform from disparate systems which were not designed to interact. The use of unstructured data with static contextual data points gave the ability to turn near real time data points into insights. Insights in this context are perishable -- the risk of a pollution event from a sewage pumping station or the level of sewage in a main is can give the operators time to act proactively, but only up to the point of failure. The data platform is the foundation of a geospatial monitoring dashboard, showing the location and assets with their status. Users are able to click into the assets to display real time status, static consequence of failure information and collaborative messages from other users. The team iterated the design of the software with the roles which are able to take advantage of the insights to act proactively at the same time as collaboratively designing new ways of working to act on insights. This approach ensured that the information and insights displayed made a difference to the ability of the operators to manage the network dynamically. Discussion The objective of the project was to test the hypothesis that Big Data is a good way to exploit the survival window of asset failure and consequence. The project revealed this approach enables operators to act proactively to asset failure and it allows operators to manage a waste water network more dynamically than without near-real time data. Visibility of near-real time asset condition and predictive failure alerts gave the central control team a broad set of opportunity windows to intervene prior to failure. Waste water networks controllers were able to send technicians to fix failing pumps within a window of hours prior to a predicted pollution event. The software interpreted information about the rate of rise in the level in the wet well compares to the rate at which the working pumps can pump out waste water to predict the time to pollution, giving the controller information about how much time they had to call and technician make the necessary fix to bring up the rate of water being pumped out or to have the pumping station contents taken away in a tanker to another catchment with the capacity to treat the waste water. The contextual data on the consequence of failures allowed the controllers to effectively prioritise work to prevent failure in times of high risk. Dynamic management of the catchment using insights from the Big Data platform allowed operators to prepare for wet weather when the risk is higher. The status of a waste water catchment during dry weather indicates the readiness of the catchment for wet weather. Before the software was developed, catchment operators had limited access to information about the status of the network during dry weather and therefore limited opportunity to prepare for high risk situations. Surfacing information in the form of actionable insights on a user-friendly interface allowed them to make decisions about strategic interventions which prepare the network and the receiving works to be able to process the load demand in wet weather. Conclusion This project revealed that there is opportunity for the use of Big Data platforms in transforming the way water utility wholesale operations exploit the convergence of operational technology and information technology. It offers a low cost route to integrate data without integrating systems, an approach that works well for water companies that must work with a legacy of disparate software systems. The use of near-real time monitoring information interpreted dynamically into insights is an exciting development, unlocking potential to exploit the survival window of asset failure and prevent costly reactive work and safeguard against consequences to customers. Will the traditional utility business model survive smart metering?, Utility Week, Accessed: 28 Oct 2014 []
2011 IWRA - International Water Resources Association - - Admin