Utility analytics: applying new techniques for smart network and customer management
Caroline Conway

Digitization is opening up a host of new opportunities for the energy, environment and utilities industry to streamline productivity and efficiency, enhance customer engagement and develop new service models. This is imperative to anticipate and manage a changing environment like demanding digital customers, stringent regulations, distributed and variable energy sources and emerging energy management technologies.

The industry’s transformation to a next generation digital model will be driven by a combination of cutting edge analytics and technologies to explore new applications of data.  On the operations side, analytics can help utilities more accurately map supply to demand, shift to a predictive/preventive maintenance and outage management model and automate real-time controls to enhance network safety, reliability and resilience. On the customer side, utilities will be able to design more personalized customer engagements based on individual usage insights rather than broad consumption patterns.

This level of analysis for both operations and customers will be critical to sustain a robust level of service in the still evolving marketplace for decentralized and distributed energy resources. Here, we look at three examples across the utilities value chain where companies can combine advanced analytics and new technologies to drive more granular management of the total utilities system.


  1. Dynamic production forecasting and risk analytics
  2. Network digitization and optimization across providers
  3. Partnership-led approach to enable smarter demand-side energy management


Dynamic production forecasting and risk analytics

Forecasting demand and production is one of the most significant and complex activities for the utilities industry, be it energy producers, waste processors or water providers. This is already complex due to the interaction of environmental, seasonal, and behavioral variables that can impact both demand and supply. And it will only become more complex as new forms of customer demand like electronics usage patterns, internet of things (IoT) developments, and electric vehicles come into play alongside new supply side patterns like distributed energy and more highly variable renewable forms of energy. Finally, infrastructure limitations create tight demands and require more accuracy in forecasting and planning than ever before.

Advanced neural network forecasting that incorporates traditional and new data variables is becoming a reality and can enhance utilities’ efforts to plan available supply and accurately meet demand in a highly variable environment.  This modeling is a form of artificial intelligence that can work with limited or very diverse variables to build up forecast accuracy over time.  It is self-learning, meaning that it can incorporate new variables and adjust forecasts quickly to get better and better at accurately predicting demand.  As part of this modeling, existing data like localized weather data and its implications for both supply and demand can also be incorporated in more precise ways.  Taking this approach, utilities can begin with existing data and over time work up to more sophisticated and accurate forecasts and planning.

Network digitization and optimization across providers

Infrastructure modernization is a major focus for energy, environment and utility companies around the world.  To justify this heavy capital investment, companies will have to demonstrate incremental value from a data-driven approach and make smart decisions about investing in lowest-cost, highest-impact technologies to deliver more value-adding information.

On the analytics, side, the application of advanced techniques in the right decision-making and process context can open up more opportunities for fast and effective operational choices on optimization of network capacity, modeling of peak flows and correction of bottlenecks. The most important aspect of this activity is being able to get to action quickly.  By evaluating the decision-making process, organization, and priority level of different types of issues, it is possible to automate certain analytical procedures through algorithms and design a relevant “control tower” for key decision makers to assess total performance and top priority issues so they can choose the right path more accurately, quickly and consistently.  This can enhance network reliability, security and cost management as well as create a robust foundation for deploying more sophisticated techniques and technologies.

On the data side, there are several interventions that can start to be made with limited capital investments thanks to new technologies coming on the market.  Two of these are blockchain databases and internet of things (IoT).  Blockchain has emerged out of the financial industry, and interesting applications are being discovered across other industries as well.  For energy, environment and utility companies, the main application is being able to share supply and demand data quickly across multiple players in distributed systems at a low cost and with the right data protections in place.  IoT presents new opportunities to build on existing network monitoring and data capabilities to get down to the asset level at a competitive cost.  Historically, sub-metering and asset-level monitoring has been a fairly costly investment.  But as monitoring technology becomes more commoditized and the right infrastructure for data capture and analysis comes into play, the investment begins to look more manageable.

By pursuing incremental improvements in both analytics and technology and prioritizing investments based on their level of impact, energy, environment and utility companies have an opportunity to move faster on infrastructure modernization and gain value along the way.

Partnership-led approach to enable smarter demand-side energy management

Many utilities are limited in their ability to accurately assess customer energy, water, and other utility usage.  However, the proliferation of connected IoT home appliances with integrated smart energy management solutions opens up a new source of energy usage insights for utility providers. Rather than rely on aggregate consumption trends providers now have the opportunity to develop insights on specific customer and household profiles real time.

Utilities can start by establishing collaborative partnerships with smart home utility management solution providers to collect usage data at a much more granular level than they can today. These detailed data sets can help utilities build more accurate and insightful profiles of energy usage over daily, seasonal and annual time frames as well as begin to segment customers into more detailed demographic profiles across their service areas. This in turn can enable development of more relevant pricing and service offerings as well as feed into demand forecasting models to more accurately map supply to demand. Additionally, this data can be leveraged to create new incentives that empower customers to manage energy usage more effectively and participate in demand side management at higher levels with more impact on grid performance.

Once utilities can access more information on their customers, it opens up many opportunities to offer better service, prices and packages, manage demand and supply, and extend demand side management programs for grid performance.  The developments in IoT and customer energy management make this a ripe area for development.


Digitization and more effective analytics are becoming critical for utilities as we gear up for grid modernization and more dynamic supply and demand.  Rather than waiting for large, long-term investments, utilities will be well served by experimenting with incremental analytics and technology. Dynamic forecasting, network optimization and creative demand-side energy management are just a few areas where data and technology can be applied in new ways to capture value more quickly. Experimenting with these and other techniques will be essential to manage multiple transitions quickly and at a low cost.