editor’s note: This article was adapted from a piece that appeared in a 2020 eBook developed by Water Finance & Management sponsored by Mueller. Click here to download the complete eBook.
The water utility sector is frequently taking the temperature on innovation. When it comes to drinking water, there are important matters of public health at stake. So, with all the new advancements flooding the marketplace, it’s interesting to examine how developments in distribution system management are being adopted and applied. What better time to examine this than now, when utilities – like so many other industries – have been put to the test in the face of a crisis that has affected them from both an operational and economic standpoint.
The COVID-19 pandemic is the most recent crisis to challenge utilities. Many utility workers have been forced to work remotely, while other essential personnel who must remain in the office, at treatment plants or in the field, have had to take new precautions with regards to social distancing. Moving forward, most, if not all utility managers, will be thinking about the longer-term issues as they continue to grapple with the current crisis. Utilities must improve their workforce and infrastructure resiliency in order to reduce their vulnerability to recurrence of a pandemic or a similar crisis.
Technology: A Critical Investment
The COVID pandemic is yet another reminder of the critical role that drinking water and wastewater systems play in protecting public health and safety, and supporting the social and economic well-being of communities. In fact, readily-available digital technologies that did not exist a decade ago can help utilities address their major pain points and drive significant economic and environmental improvements right now. Remote monitoring and control of processes can help to ensure critical infrastructure remains in service when staff are working remotely.
Writing for Water Finance & Management in 2020, former utility leaders and industry experts George Hawkins (former DC Water CEO and general manager) and Andy Kricun (former CEO and chief engineer of Camden County Municipal Utilities Authority in New Jersey) identified three important themes that would arise as areas of opportunity as a result of the coronavirus pandemic. Hawkins and Kricun note that resilience, workforce and affordability will be key areas in which utilities can and should focus on now and in the future.
Kenji Takeuchi, senior vice president of technology solutions for Mueller Water Products, has a similar perspective and says COVID brings technology resilience in the water sector to the forefront.
“Having the ability to do remote workforce management, remote monitoring through sensors and through wireless platforms shows that technology can help mitigate these types of risks and pain points,” he says. “Because of the crisis, we’re starting to see the use case and the value of having the technology information without having to physically be there, become top of mind for water utilities,” noting that industry studies on digital technology back up the trend.
One of those studies came from Bluefield Research in January 2020, which reported that over the next decade, annual capital expenditures for digital water solutions in the United States and Canada will rise from $5.4 billion (USD) in 2019 to $10.8 billion by 2030. The report suggests that the projections set the stage for more advanced monitoring and management of critical infrastructure.
According to Bluefield, the forecast is an encouraging sign that more and more utilities are recognizing the potential for digital technologies to transform their operations and mitigate the risk of failure during a crisis event, while providing financial resilience.
Day-to-Day Utility Operations
While the COVID-19 pandemic has intensified the discussion around the importance of resilience, emergency planning and taking advantage of the most advanced systems to help manage water operations, it’s important to remember these challenges did not just arise. Rather, utilities are tasked with these operational challenges every day – many of which have financial implications for both the utility.
“For a municipality or any one of our customers, it’s all about how [technology] can help drive a better focus on utility spending, a return on that investment, and how it can help them run more efficiently and spend money in smarter ways,” Takeuchi says. He explains that digital products drive efficiency in specific areas like addressing aging infrastructure of pipes and valves in a water network, and helps utility personnel prioritize repairs, replacements and how the workforce is utilized.
One piece of technology that in many ways exemplifies this on the water distribution side is the PipeRank™ virtual condition assessment which enables water utilities to get visibility into pipe failures before they happen. Historically, desktop condition assessment technologies relied on engineering judgment and linear or survival models that assumed degradation factors for pipe failure prediction. The PipeRank machine learning technology leverages on-site specific historical failure data and identifies causal factors for failures that are unique to each pipe network. The end result is an accurate prioritization of which pipes will likely break next. The utility can then focus their pipe repair program accordingly and avoid or minimize costly emergency repairs.
The figure above shows a comparison among prior failures, age and the PipeRank machine learning model for a utility in the Southwest United States. In this example, each model was conducted using available pipe construction, historical failures, environmental information, and was used to prioritize all the pipe segments from most to least likely to fail. The red dots in the diagram represent actual failures in the year following the modelling analysis.
The PipeRank model correctly placed 77 percent of the actual pipe failures in the top 5 percent of pipe segments most likely to fail. This compares favorably to 44 percent for the Age Based model and 16 percent for the Prior Failures model. The ability to define and predict which pipes will fail offers exciting opportunities to manage pipe networks more efficiently with better levels of service.
Artificial Intelligence and Where R&D Could be Focused
According to Bluefield’s January 2020 report, the fastest growing technology segment in the water industry is in Information Management. The report notes that the accelerated growth is driven by the increasing flow of data from all utility departments (e.g. customer service, operations, finance) that needs to be collected, organized and leveraged for analysis.
Larger utilities are also turning to artificial intelligence to address network and customer challenges including water loss, aging infrastructure and unforeseen climate events. Bluefield’s forecast finds artificial intelligence (AI) technologies represent significant market opportunity with a $6.3 billion projected investment by 2030.
Takeuchi says AI is an area of opportunity for the water utility sector, which he thinks is just scratching the surface on AI/machine learning potential.
“The very foundation of what Mueller is investing heavily in is a software and data platform that allows us to give machine learning insights,” he says, referencing Mueller’s new Sentryx™ Water Intelligence platform. Sentryx has the ability to collect and display data and give insights on sensor-reported data throughout the distribution including pressure, flow, leak alerts and water quality data like chlorine levels, pH levels and temperature.
Takeuchi reiterates that it’s all about prioritization. With Sentryx, he explains, utilities can prioritize assets, workforce and detect and predict events like leaks and monitor the condition of infrastructure.
“A lot of the use cases for AI in the water industry are about understanding the network – so we can understand many things like the pressure, condition of the pipes, or for demand/usage planning,” he says.
“To create an AI approach to managing your water network, you have to be able to model your water network as accurately as possible, in a hydraulic model or digital twin. It’s almost impossible to have a perfect model based on having an infinite number of sensors in the network. So, AI could be helpful in helping [utilities] estimate, predict and gain insights about what is going on in the water network.
For the generally risk-averse water utility sector, Takeuchi predicts adoption will soon accelerate at a faster pace.
“Compared to where we’re going to be, and compared to the acceleration we’re going to see, we’re just at the beginning. I expect there to be an exponential acceptance beyond where we’re at now,” he says, pointing to other industries that have allowed technology to advance further.
“Whether it’s the power industry or even oil and gas, you can see that the technology has advanced further and the value [those industries] are getting is clear,” he adds. “There’s no doubt that same value we’re seeing in other industries will apply to the water industry.”
Andrew Farr is the managing editor of Water Finance & Management, published by Benjamin Media in the greater Cleveland, Ohio, area. He has covered the water sector in North America for nine years and also covers the North American trenchless construction industry for sister publication Trenchless Technology.
