Tacoma Power formulates a risk-based life-cycle strategy to support asset management decisions.
When Tacoma Power's T&D group switched from an informal to a formal asset management program in 2009, it immediately saw a return on its time and resources investment. A systematic approach to evaluating and prioritizing equipment life-cycle expenditures ensures the utility's efficiency, reliability and employee safety.
Tacoma Power has instituted a process to support sound maintenance and replacement decisions in a repeatable, defensible, objective and systematic way. In partnership with BIS Consulting, the utility developed several risk-based economic life-cycle models. The model helps the utility balance the costs due to risk of failure of equipment as it ages equipment against the benefit of delaying spending. The approach is founded on three core asset management principles:
- Customer focus
- Least life-cycle cost
- Use of hard data in a transparent way.
Tacoma Power's T&D group is already using the refined approach to support its planning, budgeting and reporting functions.
A risk-based life-cycle strategy improves and justifies Tacoma Power's replacement and maintenance strategies, streamlines its budget and business case analysis, and helps the utility prioritize its equipment replacement and intervention evaluation process. Employees are influenced by the risk cost of all assets and can make better asset inventory, purchasing, and compliance and regulation decisions.
The asset management model is designed to minimize life-cycle costs, including repair, cleanup and replacement costs, as well as the implicit costs to electric customers from outages and other consequences of failure. The decision to replace or maintain an asset is based on balancing the cost of the project against the expected benefits from avoided risk. Postponing recommended interventions will result in lower spending and increased risk.
The life-cycle model is part of an overall asset management program that takes a comprehensive approach to spending justification. Analyses make use of engineering-level data, such as condition assessment and consequences of failure, as well as customer-level drivers, such as avoiding outages and minimizing total cost. The life-cycle model starts by prioritizing individual asset replacement or rehabilitation within each asset class. It also produces streams of risk, capital and maintenance cost for any scenario, which feed into business cases and other decision-support modules. This type of analysis provides both short-term and long-term benefits, which are reflected in reduced or optimized expenditures and improved asset performance.
Strategy for Asset Management
Tacoma Power's strategy for managing aging infrastructure includes a step-by-step process to combine the health and age of each asset, which indicates failure probability. Repair costs coupled with customer and system impacts from the equipment failure reflects the consequence of failure. Together, these define the current and future risks the asset poses to the system. The process also includes measures to constantly review and refine the inputs to ensure a high-quality, defensible and usable program.
The process begins with asset classification — identifying equipment classes and sorting them based on criteria such as long lead time, high replacement cost, whether there is an existing testing or data collection program, whether data is easy to collect and whether it will be easy to analyze more data.
Then, for each equipment class, various condition criteria are valued and weighted to determine its health assessment. The weighting of each health criterion should reflect its portion of the asset's probability of failure.
Health and Risk
The equipment's health criteria should be tangible, recordable, objective and repeatable for future assessments. Some typical condition criteria include age; inspections, monitoring or diagnostic testing; industry knowledge; equipment loading; and number of operations. Each individual piece of equipment should then be ranked by the established criteria based on the equipment's actual test or inspection results. The health assessment is used as an indication of failure probability. Historically, Tacoma Power has based this mainly on age, but it has recently recognized the condition or performance of equipment as a greater indication of potential failure.
To determine an asset's overall risk, the failure probability must be coupled with a set of criteria that measures its consequence and impact of failure. This consequence of failure is based on economic, social and environmental impacts. Tacoma Power has several considerations for the impact of failure:
- Number and type of customers affected
- Duration of outages based on system configuration
- Lead time and cost of repair
- Overall system impact.
The customer impact must include the initial interruption as well as the duration of the outage, that is, system average interruption frequency index (SAIFI) and system average interruption duration index (SAIDI) effects.
Tacoma Power estimates consequence costs by modeling multiple failure scenarios for each asset class. For example, a circuit breaker may fail to open, fail to close or fail catastrophically. Each scenario has different implications in terms of how much it will cost to repair and how many customers will be affected; furthermore, these effects will vary not only by asset class but also for each individual asset.
The utility establishes an industry standard or utility standard failure probability curve for each equipment class based on the manufacturers' predictions and historical incidents. The projected failure probability for an individual asset, then, is a function of the standard curve for its class and an adjustment based on its health.
Tacoma Power has found continuous improvement is an important part of its asset management tool. The utility has improved its data compiling process, which is important to ensure ease of repeatability. It considers changes to data inputs for accuracy or output verification. The T&D group also performs alternative analyses for maintenance cycles and replacements, as well as other asset interventions that might defer replacement, such as pole stubbing and cable injection, to streamline with other T&D projects. This helps to levelize replacement costs, optimize expenditures, and incorporate new products and technology. Re-evaluation and refinement must be included as the program is used to ensure accuracy and incorporate changes.
This strategy is reflected in the T&D group's transition from age-based maintenance of some equipment to condition-based coupled with time-based maintenance. By analyzing trends in equipment replacement and maintenance, the testing, inspection and intervention cycles can be adjusted to mitigate a pending failure.
Substation Transformer Project
Using the new approach, the T&D group first assessed its distribution station power transformers. The first step was creating a transformer workgroup made up of the asset management team and internal subject-matter experts, with support from BIS Consulting.
The transformer workgroup developed health indexing designed to use diagnostic data from Doble oil and dissolved gas analysis testing, supervisory control and data acquisition (SCADA) loading information and equipment age. The T&D group established rating values and weighting percentages within each set of test criteria based on its importance in indicating the transformer's health.
For example, the transformer experts participating in the transformer workgroup said Doble tests were the highest indicator of a transformer's overall health. Age and operating statistics were deemed the lowest indicators of its health. During the formulation of the health index criteria, subject-matter experts identified some tests as unnecessary or at least not relevant to the health relative to end of life. On the other hand, they identified furan oil testing, which shows degradation of paper insulation, as a significant factor in determining a transformer's health. As a result, Tacoma Power is currently looking into using furan testing in its on-site oil lab or at an off-site lab.
During the substation transformer project, the T&D group performed a criticality assessment to develop a measure of the cost of failure of a transformer. It included a questionnaire the workgroup and senior managers filled out regarding the perceived willingness of residential, commercial and industrial customers to pay to avoid an outage and their willingness to accept outages.
The values estimated from the questionnaire and subsequent discussions allowed the utility to assign a monetary value to customer outages for inclusion in its economic models. The T&D group estimated the number and types of customers who would be affected by each type of transformer failure as well as the expected duration of the outage, which depends mainly on station configuration. Those costs were added to the direct cost of repair or replacement to determine the total expected cost of that particular failure scenario.
Tacoma Power gathered transformer attributes — such as location, system configuration, type and number of customers, backup capabilities and estimated restoration times — and discussed them in the context of forced, normal and catastrophic incidents, as shown in a failure scenario tree diagram.
BIS Consulting provided a failure probability curve based on data and experience with other distribution utilities. Tacoma Power compared the rates of failure projected by this curve to the actual historical failure rates experienced recently to verify the curve is a reasonable representation of its experience. There was no statistically significant difference.
Tacoma Power's substation engineering group supplied replacement costs based on current manufacturer prices with respect to transformer megavolt-ampere size and labor to install and connect the equipment. The utility's operations and planning group helped to determine system redundancy and estimated outage duration times. If any repair costs or additional costs are known at the time of analysis, they can be inserted by the user.
Health, consequence cost, failure probability and replacement cost were integrated into a single risk-based life-cycle model using the Advantage methodology developed by BIS Consulting, which has extensive experience in building models of this type. The resulting tool was built in an Excel workbook with an underlying database that can be integrated into Tacoma Power's IT infrastructure. The tool calculates optimal replacement timing for each distribution substation transformer.
The population chart shows eight two-year spending periods, as well as the levelized spending over that period, along with general graphics such as risk and age distributions. The chart also lists transformers with their age, health and life totals, as well as their economic statistics such as benefit/cost ratio of replacement and remaining economic value. Any transformer can be selected from here and loaded into the analytical model for sensitivity as an easy way to summarize the business case for replacement.
Then the results section includes the benefit/cost ratio, which is calculated based on the cost of delaying replacement for one year. The benefit/cost ratio does not justify replacement, but it is useful for prioritizing when resources are limited. Because consistent measures of risk are used for all assets, the benefit/cost ratio can be used to prioritize replacement not only within an asset class but across classes.
The concept, process and calculations to prioritize the substation transformers were accomplished in a reusable, adaptable program. The program has already shown benefits. The T&D group reprioritized its replacement order, which led to a couple of budgetary request changes for the 2011/2012 spending periods.
As Tacoma Power's T&D group assesses more equipment classes and develops more programs, it can prioritize several equipment classes to optimize spending. The team is working on a multiple-asset decision-making tool that incorporates the analysis of several asset classes into one overview program that prioritizes optimal replacement of all equipment in a specified location. The program calculates the cost of salvaged life for equipment that may be replaced sooner than recommended or risk cost for equipment that may be postponed for replacement beyond its end of life. The equipment risk costs are evaluated in the program to find the lowest-combined overall replacement time, resulting in overall life-cycle cost savings.
For example, if a substation transformer needs to be replaced, perhaps it is optimal to replace the high-voltage circuit breaker and substation getaway cables at the same time. Or, perhaps it is better to wait four years before replacing all assets, balancing the early replacement of some assets with the delayed replacement of others.
Ultimately, Tacoma Power's T&D asset management group would like to automate the data collection and data processing function of the asset management programs, optimizing SAP utilization and systematically improving the program results. Many T&D employees now use the asset management program modules as decision-making tools for budgetary replacement and maintenance expenditures. Ideally, the programs will become common use for all applicable project justifications.
Rachel Allen (email@example.com) is the transmission and distribution supervisor for asset management at Tacoma Power. She joined Tacoma Power in 2002 as a system planning and operations engineer and spent eight years performing system planning studies, justifying budget requests, performing long-term horizon planning and ensuring the utility's regional stability. Allen earned her BSEE degree from Gonzaga University. She previously worked at Avista Utilities for 10 years in generation, transmission, distribution and operations engineering.
Darin Johnson (firstname.lastname@example.org) is the president and director of the asset management practice at BIS Consulting, LLC. His experience includes risk analysis, capital planning and life-cycle cost analysis for electric transmission and distribution, water/wastewater, and hydro and thermal generation facilities. This work addresses the full range of asset management program development, from framework and strategic planning through implementation of decision-support methodologies and business processes to justify and prioritize replacement of aging assets and other spending programs.
BIS Consulting www.advantagebis.com
Tacoma Power www.mytpu.org/tacomapower