Investigation of the 2.5 Beta Methodology

The IEEE Standard 1366-2003-2.5 Beta Methodology was developed with the intent of providing a reasonable way to define a major event day with respect to distribution reliability performance. According to this methodology, it is valid only if a utility's reliability data completely follows the log-normal distribution, particularly with respect to the tails. This letter's desire is to show that this is not the case for data sets provided by 12 utilities and may not be the case with other utilities. Problems arise when the right tail of utility data sets do not fit the log-normal distribution. The threshold defined by the IEEE 1366-2003-2.5 Beta Methodology will vary since it is dependent on a utility's reliability data from the previous five years. As a result, catastrophic events, reflected by a large daily SAIDI value, will cause an unsuitable increase in the threshold for future years.