How to Use the Utility Reliability App


The Utility Reliability App presents information on CAIDI, SAIDI, and SAIFI reliability metrics as reported in documents from the Energy Information Administration (EIA) from 2013 and updated annually. This includes both FERC-regulated and non-regulated utilities across the United States. Results can be narrowed by state, ownership structure, and year. 

  • Industry: Electric
  • Sub-Industry: All
  • Data Timeline: Annual
  • Data Sources: EIA Form 861
  • Geography: US States

Use-Case Overview

The Utility Reliability App creates visualizations of the data from the EIA to easily compare metrics on any utility across the United States. The app enables users to assess performance in order to determine how reliably an electric utility can deliver and maintain electric services, and to benchmark individual company performance against any other utility in the country.

Key Features

Utility Reliability Table

This pivot table provides a consolidated, filterable set of reliability metrics in a single view. 

The data in this view can be filtered on the following dimensions:

  • State
  • Company
  • Ownership Structure
  • Year

Clicking on a company name in the table (while the year filter is cleared of selections) will result in a line chart showing the various reliability metrics over time for the selected company. The various metrics and distinctions between “With MED” and “Without MED” are explained in the appendix of this document. 

Data Visualizations
Map Visualization 

This app is intentionally condensed to a single screen, providing easy access to all of the reliability data in the HData database. The filters at the top of the screen enable you to adjust which information is shown on the map at the bottom of the screen.

By default, the map averages all companies in each state. Using the button filters to the right of the graphic, you can select additional factors such as the Measurement Type, Major Event Day, and Subtract LOS to further calibrate the reliability metrics shown on the map. 

NOTE: The button filters DO NOT impact the data in the Utility Reliability Table above. 

Figure 1: Average by State
Exporting Visuals and Data

There are several ways to export data and visualizations from the app. Use the Export button () at the top-right corner of the app to generate a PDF or PowerPoint file for download.  

To export raw data, move your cursor over the card/table and another Export () button will appear at the top-right of the table. The data will export into a csv file in a long-table format. 

An additional feature is the expand details icon on each card. Hovering your cursor over a table will show the Expand Details icon (). Once you click on the Expand Details icon, a screen will pop up giving you a few options to interact with that particular card. You can click on the Export button at the top-right of the pop-up to export the visualization either in a visual format or the data behind the visualization via CSV or Excel. Clicking the table icon at the top-right of the pop-up will show you the data feeding into the visualization.

Additional Questions? Let Us Know

Hopefully this article gave you all the information you needed to be successful with the Utility Reliability App. If you still need help, please reach out to us at



Dataset Definitions:

There are three main metrics that this app visualizes. Each metric is defined and contextualized below, along with the other factors that may be included to calibrate the metrics.

  • CAIDI: Customer Average Interruption Duration Index
      • Definition: the total minutes of customer interruption divided by the total number of customers interrupted
      • Context: describes the average time it takes to restore services

  • SAIDI: System Average Interruption Duration Index
      • Definition: calculated by multiplying the average duration of customer interruptions by their total number, then dividing that by the total number of customers in the system
      • Context: describes the total duration of the average customer interruption
  • SAIFI: System Average Interruption Frequency Index
    • Definition: calculated by dividing the total number of customer interrupted by the total number of customers in the system
    • Context: this metric is used to assess how often a customer experiences an interruption
Additional Metrics:

There are several additional metrics included in the analysis that calibrate CAIDI, SAIDI, and SAIFI based on customer needs. These are described below:

  1. With and Without MED

Major Event Day (MED) describes an occasion when SAIDI, CAIDI, or SAIFI exceed a specific threshold. It’s indicative of particularly long outages depending on which of the metrics it is included in. Excluding MED can normalize data to remove outliers, but it is important to understand that including or excluding MED depends on the research question and the scope of the analysis to ensure it is being used properly and with integrity. If a number is the same when reported with or without MED, this means there were no major events for the year.

  1. Minus MED

Minus MED only impacts the CAIDI metric. It is used in a similar way as “With and Without MED” described above, but Minus MED is tied to CAIDI's particular method of calculation.

  1. Subtract LOS: 

LOS stands for loss of supply. Toggling the filter to subtract LOS means you are filtering out outage events where high-voltage transmission lines were unable to supply power. This metric is only available for utilities that report reliability metrics based on the IEEE standard (explained below).

Measurement Standards Reported to the EIA

There are two measurement standards that are reported to the EIA: 

  1. “IEEE” which stands for Institute of Electrical and Electronic Engineers 
  2. “Other” which includes all other standards in a combined group

IEEE maintains strict standards by which to measure CAIDI, SAIDI, and SAIFI. These standards may differ from other regional or international standards. Most utilities follow this IEEE standard and it is the FERC's preferred standard. You will find a complete document on the methodology of the standard here

The "other" category as reported to the EIA also uses CAIDI, SAIDI, and SAIFI but may use slightly different thresholds to create those metrics. 

You can filter between these methodologies in a drop-down filter in the driver at the top of the screen.

More on Reliability Metrics from the EIA