A data-envelopment-analysis inspired method for reliability target setting

This research proposes a method of reliability target setting for electric power distribution systems. Data Envelopment Analysis (DEA), normally used to evaluate work-unit efficiency, is applied to determine a target of reliability indices. In this paper, the target of System Average Interruption Frequency Index (SAIFI) is achieved by the proposed method. SAIFI data used in this research are provided by the Provincial Electricity Authority of Thailand (PEA). The data include SAIFI records for the past four years from 151 sub-offices. The Variable Return to Scale (VRS) DEA model is employed to identify so-called benchmark sub-offices, and form the frontier. From the frontier, the SAIFI target for each sub-office can then be calculated.