Home > Case Studies > Development of Failure Curves for Various TDT Assets

Development of Failure Curves for Various TDT Assets

Expert capabilities to assess the impact of investment scenarios for replacing or refurbishing major assets
Project objective


A large Canadian power utility wished to assess the impact of various investment scenarios for replacing and / or refurbishing their existing major assets on corporate risk factors.
To assess risk, which is a product of probability of failure and consequence of failure, requires the utility to have degradation curves for these major assets that relate asset condition to the probability of failure.
The utility hired Kinectrics to develop these curves based on the available removal statistics, condition information and input from the utility’s Subject Matter Experts (SME).


Scope of work

Kinectrics was requested to develop degradation age-based failure curves for 6 major asset categories, namely station transformers, circuit breakers, wood poles, steel structures, overhead conductors, and protection relays.
These curves were then to be used to determine a relationship between probability of failure and condition-driven estimated “effective age” (age based on asset condition as opposed to chronological age).
Solutions and work performed

Hazard density curves F(t) were first developed for station transformers and circuit breakers using removal statistics, which included number of units removed and age at removal. From F(t), hazard function or rate of failure f(t) and probability of failure POF(t) could then be derived.
The developed curves were then compared with the curves available from other industry sources, such as CIGRE, and the results presented to the utility’s SME for their comments. SME feedback was incorporated in modifying the curves developed for transformers and circuit breakers, and for developing curves for the other four asset categories.
One of the major challenges of the project was a lack of removal data, so Kinectrics and the SME had to make some assumptions. Another major challenge was that the removal curves included removals that were condition based and non-condition based, e.g. due to obsolescence. The ratio of condition based and non-condition based removals varied greatly from asset category to asset category. Since only condition-based removals impact risk, an asset-specific approach was developed in collaboration with the utility’s SME to derive condition-based curves from the removal curves.
Results and client benefits
The project results and data obtained will allow the utility to utilize the condition (status) based degradation curve in assessing impact on risk from different investment scenarios.