- Utility Transformations
- Utility Transformations
- Advanced Metering Infrastructure
- Smart Utilities Program
- Energy Efficiency
AMI Data Brings New Possibilities for Energy Efficiency Measurement and Verification: Part 1
Coauthored by Emily Cross and Peter Steele-Mosey
Utility industry stakeholders have been debating whether the proliferation of advanced metering infrastructure (AMI), also known as smart meters, will change the way energy efficiency program evaluation, measurement, and verification (EM&V) are conducted. Many utilities remain unsure about what is realistically possible. This uncertainty is compounded by the fact that new firms seem to emerge each year, claiming to provide increasingly deep insights into customers’ energy reduction potential (such as appliance-level load disaggregation and building-specific identification and targeting) using little more than consumption data from the utility.
How Can AMI Data Be Used?
In the field of EM&V, what is AMI data good for? How can it be used by utilities, regulators, and stakeholders to reduce evaluation costs, deliver more accurate and precise estimated program results, and improve the effectiveness of program delivery?
To answer these questions, it is helpful to define the two key evaluation-driven use cases for AMI data:
- Operational improvements: Early indications of program achievement provide the opportunity for course correction. Due to the continual collection of AMI data, it should be possible to quantify the impacts of changes in marketing approach and customer targeting on energy efficiency achievement more quickly than is traditionally required for program evaluation.
- Program impact evaluation: What is the best estimate of the energy and demand savings that a program delivered? This type of information is required to track utilities’ progress against mandated energy efficiency targets, to enable energy efficiency programs to be bid into energy and capacity markets as resources, and to quantify overall program cost-effectiveness.
Part 1 of this blog covers operational improvements, while part 2 will cover program impact evaluation. This topic is covered in detail in Navigant Research’s new report, Utility Strategies for Smart Meter Innovation: Energy Efficiency Measurement and Verification.
Utilities are all too familiar with the frustration of waiting for results from evaluators. Typically, a full year of data is required and the evaluation itself may take several months. This lag between implementation and assessment limits the ability of program administrators to course correct underperforming programs or understand how to tailor messaging to maximize the recruitment of high potential customers.
AMI data is collected continually, and several firms have recently come to market with prebuilt software solutions designed to quickly plug and play with this data. In theory and depending on the type of program, it should be possible to obtain ongoing updates of program performance long before the actual evaluation even begins.
These software packages have their limitations and are no substitute for a custom econometric evaluation, as they tend to be one size fits most. Additionally, the innovative approaches they employ sometimes lack the support of academic and professional literature from which econometric approaches benefit.
There is no denying, however, that these prebuilt software solutions can deliver results much more quickly than the traditional approaches. The results may not be sufficiently robust for a regulatory environment, but they may (depending on the program and the vendor) be sufficient to allow program administrators to take greater control of their programs and monitor their progress in near real-time. Program administrators would have the opportunity to make more effective use of program budgets and increase the value of their programs for their shareholders and ratepayers. They could use these software solutions for programs where simply multiplying the implementer‑reported savings by the prior year’s realization rates is not expected to be accurate.