The Water Utility of the Future – Vision of the Future

The Utility of the Future Series

In this six-part series, Travis Smith, Joseph Dryer and Zachary Barkjohn look at the social, environmental, and economic pressures facing utilities. They will share their vision of the future using current technology, integration, and methods to break from a vicious financial cycle to become a sustainable utility.

In part two, the authors look at the interdependencies of smart water and address why utilities should take a holistic approach to the entire water cycle.

By combining sensors through the smart utility network with advanced analytics technology, utilities can integrate their data to capture metrics around usage patterns, hydraulics, water quality, sustainability and asset management. Building utility intelligence is about addressing solutions – the right combination of smart technology paired with enhanced analytics can result in success and sustainability.

Utility intelligence is the result of a holistic approach to the entire water cycle and all of the operational and asset parameters (see figures 1, 2, and 3), similar to Integrated Water Resources Management. By having an interrelated combination of modular solutions and practices to empower the utility to make their jobs easier and more effective while also being adaptive, flexible, and responsive to the changing needs. These solutions comprise of data, integration, policies and business practices, and services. 

At the essence of utility intelligence is the reliable, long-term capture of the right data at the right time. Utilities need flexible data models for volume, variety, veracity, and velocity. Far too often the capture of data is too little, too much, inaccurate, latent, and not correlated. This renders paralysis or worse poor return on investment with little confidence in the actions to be taken or the actions leading to inadequate or underachieving results.

With the coming of age of artificial intelligence (AI) and machine learning, a flexible model to increase data volume by locations or intervals, coupled with prioritization on information (such as alarms) can change the usability, as well as the asset life and expense of sensors, communications devices, and data storage. The ability to configure sensors and alarm set point data intervals remotely allows utilities to adapt to changes and prioritize issues. By correlating multiple sources and measurement parameters, utilities can create useful information that can improve the validity of the issue(s) and allow them to charge forth with confidence. 

The second key aspect is the integration of data silos. Utilities are saddled with a labyrinth of systems, firewalls, premise devices, cloud access, duplication of records, conflicting records making the isolated functions difficult to use. In order to improve the usability of the various Geographical Information Systems (GIS), Enterprise Asset Management (EAM), Work Order Management (WOM), Computerized Maintenance Management System (CMMS), Meter Data Management (MDM), Customer Information Systems (CIS), portals, Enterprise Resource Planning (ERP), enterprise data management, Hydraulic Models, and supervisory control and data acquisition (SCADA) systems – there needs to be a plan of relating the data, the system of records, and integration to make all the data impactful.

Each of the systems have specific functions and data that are trying to increase their scope. One way to improve the usability is to review the data and system integration into groups, metadata (weather, parcels, locations), customer data (usage, service records, accounts), operational data (SCADA, Work Orders, MDM), asset data (type, age, maintenance, etc.), planning data (modeling, historical), and storage. It typically becomes clear which system is the best for the function and system of record based upon these groups. Integration with systems should then be evaluated based upon latency and methods such as flat files, restful API, or other exchange methods. The often overlooked part is the configuration and alarm data that can clear the confusion as it reduces the volume and lends focus to potential issues.  

As all of this technology becomes deployed, the utility must adjust business practices, policies, and human resources to reap the benefits. For example, remote service control at the meter with valves is becoming increasingly popular and offers alternative means, such as reduced flow. This technology requires updates to policies and business practices around shut offs restoration, reduction and accounts. Likewise with the adoption of analytics, and data integration the staffing needs at the utility may need to change to include experience in analytics, and data science to capture a solid return on investment.

The workload at a utility often fluctuates due to planned expansions/upgrades, emergency issues, and changes all while balancing customer service, operations and maintenance needs. Budgets are typically less flexible than the changing nature of operations or maintenance needs, a utility may want to consider services as a way to fill in gaps, or change capital expenses. There are a variety of new models on the market for changing what was typically a capital improvement to services, such as pumping as service, water as service, network/software as a service, not to mention a variety of innovative financial models and professional services for engineering, consultation, business reviews, and data integration or customization. 

The solutions must be holistic. Utilities are involved in all aspects of the water cycle – treatment, transport, sustainability, usage, and the assets, so their solutions cannot be isolated to one function. For instance, a typical strategy for non-revenue water is to deploy district metering. Along with that method is zone control which may change the water age patterns and thus the flushing needs to adjust. Likewise, the repair of leaks have great efficiency benefits, but may have unintended consequences like a new burst or water quality issues; as the leaks provided stress relief for water hammer and flushing. One can think of the interdependencies of hydraulics, quality, sustainability, assets and usage as a spider web. If tugged on in one area, it impacts the others (see figure 4).

The data should be presented in a way that is consumable and easily actionable, to increase usability and effectiveness. The usage and value propositions should span across all users and multiple departments such as: operations, customer service, accounting, engineering, management, planning, and third parties such as consultants. Categorization of the interfaces and integration will help with usability. The following describes various types of interfaces as shown in figures 5-7.

• Dashboards – what is the pulse of today and what’s going on in the process
• Heat maps – concentration of data points, what’s coming and the scale of magnitude
• Trends – data over time with context and aggregate metadata
• Alerts – signals that warn when events may happen and allow the user to monitor them to catch use of data
• Models – predictive information to help with planning, emergency management and sustainability of future events

To increase the effectiveness and usability the solutions need to be integrated down to the field, use of technology to make data actionable and tangible – all the information needs to be integrated to capture the return of investment (see figure 7).

There is no reason to limit the scope to a single portion of the water cycle. Too often source water, treatment, distribution, collection, reclamation, reuse, storm water are isolated from one another. Implementation of a smart utility network can be applied across the water cycle (see figure 12) to provide new insights into the overall watershed management efficiency and sustainability – creating utility intelligence. This will also provide better utilization of data systems, and communications infrastructure to lower costs. 

The approach and methodology to bring all of this together is a time-honored engineering one. Starting with measurement, progressing to monitoring, then analyzing, followed by improvements, and lastly maintaining control (see figure 8). This approach can be iterated over time for continuous improvement for any issue.