Diagnostic testing of stochastic circuits

The desire of circuit owners to prioritize their underground cable rehabilitation expenditures has spawned a variety of diagnostic testing tools. The goal of these diagnostic tests is to separate the "at risk" cable population into the subsets of those that are likely to provide reliable service for some time and those that are likely to fail first - the "good" from the "bad." A decade of field failure data, covering over 18,000 km of cable installed over the 3 decades between 1969 and 1999 examined in this paper, provide statistically significant insights into the stochastic nature of cable reliability. A low cost method to prioritize circuit rehabilitation efforts is described. The method provides the foundation for the economic modeling and optimization of various rehabilitation strategies.