Abstract
This thesis presents a new procedure for setting the target levels of reliability performance across service regions of an electric utility. Conventionally, the management board of utility companies sets corporate targets on reliability indices, such as the system average interruption duration index (SAIDI), the system average interruption frequency index (SAIFI) etc., in terms of the SAIDI or SAIFI reduction, and then dispatches these corporate targets to the regional offices commonly at the same percentage of reduction as set for the corporate targets. Such an even distribution of corporate targets among regional offices is unfair in particular for the offices that have reached a good record through their managerial effort. To overcome this difficulty, a new procedure based on the analytic hierarchy process (AHP) is designed, which accounts for the disparity among regions on a variety of factors including the geographic, load, circuit, customers, maintenance work and expansion budget conditions, by: (1) measuring the regional status of each factor, and (2) weighing the relative impact among factors on the regional SAIDI or SAIFI. Because the weighted impact of each disparity factor is evaluated individually, any disparity factors which have clearly distorted the benchmarking in setting regional performance targets can be segmented effectively from the evaluation. The regional disparity model and the weight acquisition procedure, designed by following the AHP, have been tested in the field on dispatching Taipower’s corporate targets to her 6 transmission regions and 22 distribution districts. The disparity model, weighing procedure and field test results as well as the effectiveness and limitation of designed procedure are presented. The proposed procedure, after slight modification, can be extended to setting the target levels on reliability performance across utilities, and can serve as the basis for benchmarking the reliability performance among utilities.