New Framework for Adoption of Distributed Energy Resources Quantifies Uncertainty, Ensures Validity Across Grid Structures

The rapid growth of distributed energy resources (DERs), such as rooftop solar, is changing how electric utilities plan for the future. In a new study, researchers propose a framework for DER adoption predictions that quantifies uncertainty and ensures validity across hierarchical grid structures.

The study, by researchers at Carnegie Mellon University, is published in the Annals of Applied Statistics.

“As more households and businesses adopt these technologies, utilities must decide where and when to invest in grid upgrades, how to prepare for changing electricity demand, and how to maintain reliable service for customers,” explains Shixiang (Woody) Zhu, assistant professor of data analytics at Carnegie Mellon’s Heinz College, who coauthored the study.

“These decisions are challenging because DER adoption does not happen evenly across communities, and small differences in local adoption patterns can have major implications for circuits, substations, and long-term infrastructure planning.”

The increasing adoption of DERs presents both opportunities and challenges for managing electric grids. Accurately predicting DER adoption is critical for planning infrastructure proactively, but the inherent uncertainty and spatial disparity of DER growth complicate traditional approaches to forecasting, and the hierarchical structure of distribution grids requires that predictions satisfy statistical guarantees at the circuit and substation levels.

In this study, researchers developed a data-driven forecasting approach that helps utilities better understand not only where DER adoption is likely to grow but also how much uncertainty surrounds these forecasts. Rather than producing a single projection, the method provides decision makers with a range of plausible adoption scenarios across different parts of the distribution grid, allowing utilities, regulators, and planners to make more informed decisions about grid investments, capacity planning, and resilience strategies under uncertainty.

Using customer-level solar installation data from Indianapolis, Indiana, the study showed that the approach can provide more reliable and actionable forecasts than existing methods. For utility planners, this means better visibility into which circuits or substations may experience faster DER growth and where proactive infrastructure planning may be needed. For regulators and other stakeholders, the framework offers a more transparent way to evaluate long-term investment needs and prepare for the evolving role of customer-owned energy resources.

“Our project has already demonstrated real-world impact,” says Wenbin Zhou, a PhD student in machine learning and public policy at Carnegie Mellon’s Heinz College, who coauthored the study. “Our approach was adopted and incorporated into an Indiana utility’s 2025 integrated resource plan, where it helped inform long-term planning for distributed energy adoption and grid infrastructure needs.”

The project also won second place in the 2026 Innovative Applications in Analytics Award at the INFORMS Analytics+ Conference, recognizing its contribution to analytics-driven decision making in the energy sector.

The study was partially supported by an NSF grant and the award money provided by AES.

###

Summarized from an article in the Annals of Applied Statistics, "Hierarchical Probabilistic Conformal Prediction for Distributed Energy Resources Adoption," by Zhou, W (Carnegie Mellon University), and Zhu, S (Carnegie Mellon University). Copyright 2026 The Authors. All rights reserved.

About Heinz College of Information Systems and Public Policy
The Heinz College of Information Systems and Public Policy is home to two internationally recognized schools: the School of Information Systems and Management and the School of Public Policy and Management. Heinz College leads at the intersection of people, policy, and technology, with expertise in analytics, artificial intelligence, arts & entertainment, cybersecurity, health care, and public policy. The college offers top-ranked undergraduate, graduate, and executive education certificates in these areas. Our programs are ranked #1 in Information Systems, #1 in Information and Technology Management, #8 in Public Policy Analysis, and #1 in Cybersecurity by U.S. News & World Report. For more information, visit www.heinz.cmu.edu.