MAISY^® Utility Customer Databases, Forecasting & Analysis Models

Utility Customer Data Resources for Electric Utilities and Technology Companies

MAISY Databases provide energy use and hourly kW loads for more than 7 million
residential and commercial utility customers

MAISY^® Utility Customer Energy Use and Hourly Loads Databases:
- Provide granular, block-level energy use and hourly load data for actual, individual utility customers (all personal identifiers removed).
- Support energy customer market assessments and energy technology product development.
- Provide input to AI-assisted agent-based and utility customer digital twin forecasting models.
- Support enhanced grid visibility including EV and weather impact analysis, customer growth/dwelling unit churn, smart technology design, DSM technology and program analysis and targeted DER deployment.
Unlike engineering-based static datasets available from EIA and NREL, MAISY^® captures localized, dynamic, hourly energy use reflecting dwelling unit, appliances, demographics, and weather for 7+ million actual U.S. households.

GIM Block-Level Grid Risk Assessment with Targeted DSM, DER, and VPP Solutions

The Excel-Based Smart Grid Research Consortium (SGRC) Gird Impact Model
provides utilities with ZIP and neighborhood hourly load forecasts, impacts of
EV, electrification, weather extremes and demand management

The Grid Impact Model (GIM) is a block-level electric distribution planning and screening platform that evaluates how:

EV adoption, housing growth, demolition/rebuilds, electrification, and extreme weather affect local grid hourly loads, peaks and infrastructure risk
Demand side management (DSM) programs can modify loads and peaks and moderate risk
DSM, EV and home battery storage can support a virtual power plant (VPP) strategy

GIM models end-use energy consumption and hourly kW loads for a sample of actual individual utility customers (all personal identifiers removed) who reside in each block. This customer data has been curated by us, and enhanced by AI-assisted integration with over a dozen supporting datasets—including 15-minute end-use metered data.

While customer records reflect real utility customers, no identities or personally identifiable information are used or available, and no customer contact is required to implement the model for utilities.

The Excel-based GIM applies a customer (as opposed to asset) digital twin analysis to simulate real-world localized grid behavior. Intuitive tools for scenario testing, flexible inputs, and real-time visualization, make complex hourly load modeling and analysis both accessible and actionable.

Dashboard summaries and interactive filtering provide block-specific results for all single family customers and for customer segments defined by income, demographics, or housing attributes. The model also supports extraction of individual representative customer records containing both baseline and forecast 8760 hourly load profiles, allowing users to explore individual customer contributions to segment results.

For utilities without detailed feeder or transformer power-flow models, the GIM provides a fast, block-level way to evaluate future grid impacts of electrification, load growth, extreme-weather coincidence and potential utility demand side program impacts. GIM identifies potential local constraints and performs rapid non-wires alternative (NWA) screening without requiring complex engineering software. It also produces baseline and scenario-specific 8760 hourly load profiles that can be used directly in future engineering studies or shared with consultants when needed.

For utilities with detailed feeder and transformer models, GIM serves as a pre-populated, analytics-driven front end that identifies the most critical locations, scenarios, and customer segments before circuit modeling begins. In addition, GIM provides ready-to-use baseline and scenario-specific 8760 hourly load profiles, enabling more accurate and efficient power-flow and time-series studies. This upstream scenario intelligence reduces engineering workload while improving study precision.

A quick intro is provided by the following video and online presentations viewer:

1-minute Executive Overview Video
5-minute Online Viewer or pdf presentation of an example EV impact analysis

Forecasts are available for current year, 2030 and 2035 single family customer bases with service area, ZIP, block group, neighborhood, and customer detail. Output detail includes:

Electricity use and 8,760 hourly kW loads
Future EV ownership and 8,760 hourly load charging impacts
Impacts of increased population and dwelling unit demolition/rebuilds
Extreme weather impacts
DSM, DER, and VPP program impacts

Download the no-hassle working GIM Model demo to explore unique distribution planning and analysis capabilities.

The GIM Resources Page provides links to GIM resources and examples of distribution system analysis applications.

The Grid Impact Model is the third generation Jackson Associates energy forecasting and analysis model. The Smart Grid Investment Model was developed and implemented for two dozen electric utilities to provide business case analysis of AMI and related smart grid investments while CEDMS and REDMS agent-based models have provided end-use electricity use and DSM analysis for more than forty electric utility, power pool, state, and federal government clients.

...
Grid Impact Model (GIM)

A critical new imperative for many utilities is forecasting the upsurge in small-area grid loads caused by rapidly growing EV ownership, electrification and weather extremes. Unanticipated grid overloading can cause low voltage, flickering lights, reduced transformer lifetimes and even blown transformers, not to mention customer complaints/ dissatisfaction.

Forecasting these neighborhood-level loads requires an entirely new set of tools. Digital twin modeling, a new forecasting technology that is increasingly being used to model and analyze electric utility distribution equipment, can be applied to model individual utility customer hourly loads providing new insights into small area grid threats.

Customer (as opposed to asset) digital twin modeling applies information on actual anonymous utility customers to identify likely changes in distribution grid loads and results of load-shifting utility programs. While customer records reflect real utility customers, no identities or personally identifiable information are used or available, and no customer contact is required to implement the model for utilities. Models are updated periodically to refine customer ownership, loads and utility program model relationships based on actual household information.

MAISY Energy Use and Hourly Load Residential Databases, consisting of more than 7+ million actual individual utility customers across the US, provide the perfect foundation for customer digital twin modeling. Information for each customer includes hourly electricity use by end-use (space heating, water heating, etc.), dwelling unit data, EV ownership, commuting schedules, income, and a variety of other socio-economic variables.

The Grid Impact Model (GIM) extracts utility customer records for customers residing in a specific block area for its digital twins. Changes in customer loads resulting from increased EV ownership, increased electric appliances, weather extremes, population growth, dwelling unit demolition/rebuilds, price and utility programs are forecast to determine impacts on local area transformers and feeders.

The digital twins customers reflect a statistically valid sample of actual utility customers within each block group, providing a low-cost alternative to collecting information on every customer in the service area.

Regardless of the extent of existing utility feeder and transformer power-flow modeling resources, the Grid Impact Model provides vital intelligence needed to understand future, localized grid impacts.

For utilities without circuit models, GIM delivers rapid block-level impact assessment and non-wires alternative (NWA) screening providing exportable hourly load profiles for future engineering use.

For utilities with circuit models, GIM acts as a pre-populated, analytics front end that identifies priority locations and delivers baseline and scenario-driven 8760 hourly load shapes for more timely, accurate and efficient engineering studies.

GIM modeling and analysis ensures that any utility can more effectively analyze emerging grid risks and evaluate potential solutions quickly and consistently.

Grid Impact Models Page

MAISY Residential Utility Customer Databases

- 7+ Million actual individual household records across the US
- Geographic detail: ZIP code areas, counties, utility service areas, individual customers, etc.
- Socio-economic, dwelling unit, end-use energy use, hourly/15-minute electric loads, EV charging loads for each household
- Supporting more than 150 market analysis, forecasting, technology analysis, hourly load analysis projects for electric utilites, federal and state governments, private company clients.

Other Database Items
Summary of all MAISY Databases	Sample database applications
Applications by industry/technology	Analysis and consulting projects

Let Us Help Define Your Data Needs

Jackson Associates has worked with
more than 150 MAISY Databases clients to identify critical data items and analysis with detailed
household databases and user-friendly Excel-based forecasting models

Not sure exactly what data items you need, or which data items are best for the project at hand? Need some summary data or specific analysis results?

Let us help! – just e-mail us with your questions and/or suggest a time to discuss. We provide free consultations to help identify the most useful data/analysis for your application. We also provide free telephone support to assist with client data applications after data delivery.

MAISY Energy and Load Forecasting Models

MAISY forecasting models provide EV, electrification and weather extreme
impacts on energy use and hourly loads as well as demand management program analysis
designed to mitigate these utility grid threats.

MAISY energy models and forecasts have powered energy applications for decades. MAISY clients include fortune 100 companies, start-ups, electric utilities, US, state energy agencies and more.

MAISY AI agent-based models provide forecasts and analysis for geographic areas ranging from ZIP codes to utility service areas to states. Model output can also provide detailed household record data for users who want to drill down on specific issues. ...

MAISY AI agent-based model methology
EV ZIP/census tract saturation forecasts , household hourly loads W/WO EV charging
Smart grid, solar, battery storage, DER market analysis and peak hour impacts
Microgrid design and assessments
Residential household forecasts including household income, demographics, dwelling unit, appliance, energy use, and hourly loads data for 6+ million US households

Dwelling unit data, e.g., square feet, space heating equipment, appliances, etc.
Location data: ZIP code, county, place, metro area, 30-year degree days
Emissions data: Total, electricity, natural gas, fuel oil, propane
Annual Energy Use by fuel type (electricity, natural gas, fuel oil, propane) and end use.
8760 and 15-minute kW loads (whole building and end-use, including EVs, monthly averages)

Jackson Associates Provides Industry-Leading Data, Models and Analysis

Jackson Associates has worked with
more than 150 MAISY Databases clients to identify critical data items and analysis

Why trust Jackson Associates (JA) to help with your forecasting, analysis and data needs? The internet is filled with sites offering all kinds of information, often of dubious quality - consider the following:

We provide decision-makers with information that often informs multi-million-dollar investment decisions for some of the largest US corporations.
Our energy, hourly load data, and forecasts/analysis results and our expert witness testimony have supported electric utility and regulatory agency investment decisions in dozens of states.
MAISY data have provided the information basis for development of several US Department of Energy energy efficiency standards.
We were among the first to apply AI machine learning to integrate and validate disparate data sources. Our patented business intelligence (BI) drill-down software (US Patent 5,894,311, Computer-Based Visual Data Evaluation) has been licensed by every major BI software company including Microsoft, SAP, Oracle, and others.


Recent Updates, Notes & White Papers

Agent-Based Model Uses AI to Map Future Utility EV Distribution Challenges; Identifies ZIPs with Greatest Future EV Increases	Built on a $1B Data Asset: AI-Powered Neighborhood Load Forecasts for Utilities	Next-Gen Models: Forecasting EV Loads with Digital Twins. Forecasting New Neighborhood Loads Requires New Tools