Some Thoughts on Minimizing Cost of Arc-Furnace Transformers

A series of failures of large arc-furnace transformers have caused users and manufacturers to reevaluate their traditional view of transformer reliability. The costs of outages are higher because arcfurnaces now produce a larger fraction of all steel in the Western world than in the past. The purpose of this paper is to present a model for minimizing the costs of failure including down time, caused by a sudden breakdown of a large piece of equipment, which could be replaced with another unit performing identical function and called ``sliding standby.´´ Models include replacement of the failed unit with a i) standby, ii) leasing, and iii) repair program; the time value of money; costs associated with failure; taxes; variables suggested by Jorgenson and Radner; variables suggested by steel industry experts. Since the useful life of an arc- furnace transformer is rarely known in advance with a degree of certainty, the reliability is analyzed. An empirical distribution of failure times is based on a sample taken from experts in the steel industry.