I remember the day Ameren decided to break with tradition, the day we started to move away from the traditional way investment decisions were made in the design, construction and maintenance of our system. Charlie Schukai, our senior vice president, challenged us to come up with a new method to rank the value of our proposed distribution projects.

Our planning strategy is considered radical by some. You see, we've moved away from the traditional criteria-based planning. Instead, we are embracing a system that weighs the cost of every system enhancement against its impact on system reliability. It is dollars spent against reliability gained. Today, Ameren is watching every dollar; we have to. Our budgets are tight. Some of our service territories are even facing rate freezes.

So What Are We Doing?

Right now, our system designers are charged with evaluating the reliability benefit of every component added to the system. We are rethinking our old habit of providing redundant feeders to all shopping centers and subdivisions with underground service. Similarly, we are moving to case-by-case determinations of whether we install redundant supply and redundant transformation at our distribution substations. These steps might seem like heresy to traditional utility planners, but here is our reasoning: We've come to the conclusion that two redundant facilities are not always of equal value.

Evaluating Redundant Facilities

First, we valuate our ability to respond to a failed component in the absence of redundant facilities. If the lights go out, how long will it be before we can restore power? What portion of the load can be diverted? How long will it take to locate, isolate and repair a faulted cable? How long will it take to install a portable substation transformer? Our analysis shows that portable capacity is significantly less expensive to provide than fixed in-place capacity. Our company maintains portable substation transformers with capacity up to 35 MVA. We have processes and equipment in place so that if any distribution class substation transformer fails, we can provide replacement capacity within 24 to 36 hours.

Second, we determine the percentage of time in a year that redundant capacity would be unavailable. Taking a look at Ameren's typical mix of residential, commercial and small industrial load, 25% of our typical summer peak-load demand exists on the system for less than 5% of the year. For example, a distribution feeder has a summer peak of 10 MVA. For 95% of hours out of the entire year the load on the feeder will be at 7.5 MVA or below.

A Little Closer to Home

I decided to apply our redundant strategy in my personal life when I was considering backup generation at home. I recently moved from a subdivision with underground service to a neighborhood served from a single-phase, 12-kV overhead circuit. Realizing we were susceptible to major ice storms — like the one that hit nearby Kansas City several years ago — that could easily put my house in the dark for days, I priced generators and the electrical work necessary to provide electricity to a few lights and the fan to my gas furnace. After balancing the cost ($2000) against the likelihood of a major ice storm, I decided the benefit was not worth the cost.

Objectivity Spoken Here

To design, build and operate a reliable and economic distribution system requires an objective cost/benefit analysis. It also requires an open mind, admittedly a difficult task. We found it difficult to let go of old paradigms, but we are finally able to accept that redundant capacity is not the only way to address single-component failures during system peak-load periods.

We realize that if we are to embark on a reliability-based investment strategy, we must have accurate outage data. We are working hard to get our arms around the components that drive failure probability so we can positively impact our outage-probability and outage-duration numbers.

A Higher View

Ameren also has moved to installing underground residential distribution in our Missouri service territory, which will reduce outages while eliminating tree trimming. We require commercial and residential developers to install conduit systems for all distribution facilities. The customer gets underground service, while Ameren reduces its tree-trimming and pole maintenance costs.

If We Lived in a Perfect World

Customers want 100% reliability. Investors want giant-sized returns with minimal risk. Our utilities want facilities with infinite life. As we move to balance conflicting wants and needs, we have to return to the concept of value, of providing a reliable product at a reasonable price.

As Ameren takes a fresh look at the decision tree for investments in distribution, we've come to realize that traditional life-cycle costing provides significantly different results than those obtained using reliability benefit analysis. As a company, we have come to accept that we will not provide redundant supply for some loads. This is hard for some of us to swallow, but we must make hard choices if we're to find a real cost/benefit balance in serving our customers.

Bart P. Angeli is a supervising engineer at Ameren Services in St. Louis, Missouri, U.S. BAngeli@ameren.com