How the study was done
For its analysis of power company reliability, The Washington Post made use of the two most-widely accepted industry standards.
The System Average Interruption Frequency Index (SAIFI) measures the average number of sustained outages that customers experience in a year. A sustained outage is generally defined as one lasting more than five minutes. SAIFI is important because it represents the probability of a service outage.
In 2009, Pepco reported a systemwide average of 1.74 interruptions. The average was 1.06 for the District and 2.06 for Maryland.
In surveys, most utility companies reported substantially less-frequent outages than Pepco. A survey of large metropolitan utility companies by the Salt River Project, an Arizona utility, reported that the average number of interruptions in 2009 was 0.9. A broader survey of more than 100 electric utilities conducted in 2009 by the nonprofit Institute of Electrical and Electronics Engineers (IEEE), reported an average of 1.18.
The other metric used by The Post is the System Average Interruption Duration Index (SAIDI). It measures the average length of sustained outages per customer.
In 2009, Pepco reported that the systemwide average length of a sustained outage was 3.07 hours. The average was 2.35 hours in the District and 3.42 hours in Maryland.
Most other utilities reported shorter average outages. The Salt River survey reported an average duration of 1.2 hours. The broader IEEE survey reported an average of 2.23 hours.
Most data used in the newspaper's study excluded major interruptions, including those associated with large storms, to provide a more accurate picture of daily performance.
The Post collected data from the government agencies that regulate power companies, from trade associations and from the companies themselves. Although disclosure requirements vary by state, most jurisdictions require electricity companies to compile statistics on the frequency and length of outages.
- Joe Stephens and Mary Pat Flaherty