From knowIT

There's little margin for error when you're in the business of selling electrons. After all, if an electron traveled around the world instead of bouncing around the nucleus of an atom, it would circle the earth some 8.3 times in one second. Since there's no time to react, electricity providers must do all they can to prevent system failures.

Arizona Public Service (APS) has developed award-winning innovations in this area. During 2008, the utility earned the Edison Electric Institute's Edison Award -- the highest industry recognition an electric utility can earn -- for one such breakthrough.

Ostensibly, the engineering that earned this honor was a diagnostic tool to detect pending problems in transformers, the electric-grid equipment that raise or reduce power from one voltage to another. But contributing to that development were top-tier process-improvement methods and a dedicated team that teaches other utility departments how to get business-process management (BPM) rolling.

Shifting focus

Called the Center for Process Excellence (CPE), the APS process-improvement group is the focus of an instructional case study written by Raghu Santanam, an associate professor of Information Systems at the W. P. Carey School of Business. Santanam credits the APS team with creating the right conditions for process orientation to become part of the utility's corporate culture.

You're not going to create a process-based organization overnight, he says. It takes time and effort, and you have to create a structure in which it can happen.

What Santanam admires about APS is that the utility's Chief Information Officer (CIO) first built an infrastructure -- the CPE -- then built competence within that organization. Finally, the process team learned how internal consulting would work within the different business units of the utility. Now they have a coherent story and a number of examples where this process orientation added business value, he says.

It all started with the utility's Information Systems (IS) Department, and that's where the CPE still resides. Why did IS managers want to shift to a process-oriented focus? When people have a problem, they come looking for some whiz-bang application to fix it, says Randy Sorensen, APS CPE manager and a man who admits he's sought a few problem-solving gizmos in his past, too.

If you don't first look at the underlying business process and get that squared away, slamming technology on top of your problem is going to give you the same results -- maybe quicker, he adds. The most important thing to do is understand the underlying business process, then look at how technology can accelerate it.

It was this type of thinking that prompted the utility's former CIO to roll out a process-improvement initiative in his own department nearly a decade ago. It worked so well, he kept the process-improvement team together and began offering the service to other business units within APS.

According to Santanam, this visionary CIO realized that the functional business model had run its course in corporate America. Businesses that wanted to thrive had to switch to a process-oriented approach.

What's the difference between a functional and process-oriented business model? In a functional business model, business is organized, managed and optimized within departmental -- and functional -- business units. Process orientation means you don't look at departmental boundaries, Santanam explains. You look at the business from end to end -- from the moment the customer demands something from you to the time you deliver the product or service.

This cross-functional approach is preferable, he says, because when you're optimizing work within one function only, it could be detrimental to the business as a whole.

Santanam points to an example of this in mortgage sales. If the person selling loans is measured based on how many loans he gets approved, not how many loans end up in default, he won't care if loans are good or bad.

Another example Santanam recalls occurred in a computer-selling operation. The business configured servers for customers, which involved putting together all the pieces, such as memory, processors, disk drives and related technology. To capture customer specifications correctly, the manufacturing department created an order-processing system. But, the sales team wouldn't use it, because salespeople were rewarded on sales, not correctly configured servers. Consequently, errors continued to plague the production process until the company started rewarding salespeople for both sales and order accuracy.

Beyond boundaries

Getting companies to look beyond departmental boundaries is a key part of the CPE's mission and success.

As noted earlier, the team started as a departmental initiative in the IS group. The CIO brought in a consultant to teach business-process improvement 101, says the CPE's Sorensen. Employees learned process modeling, how to facilitate a process-modeling session, how to capture issues from a session -- basic skills.

The team that learned these skills are the same people who later constituted the CPE, the infrastructure that Santanam mentions. This team continued building on its initial training, which is another thing Santanam finds exemplary.

The Center's staffers have close collaboration with other companies, they add to their knowledge through continuous training, and they share their own lessons by participating in conferences as speakers or by publishing articles on their work, he explains. All of these interactions help people broaden their horizons. An organization like the CPE has to be both outwardly and inwardly focused.

Santanam also points out that the group borrows methodologies from literature, mixing together principles from several BPM philosophies to best engage their own clients. Plus, the CPE has standardized its own methodology of interaction.

This methodology, Sorensen explains, starts with a meeting with the project sponsor to come up with a project scope. Then you form a good process-modeling team with the right subject-matter experts in the room, he says.

From there, the CPE staffers facilitate two or three modeling sessions. The first session, Sorensen notes, usually gets the team about 70 percent done with their process mapping and issues identification. The second session gets them about 95 percent done and, often, no third session is needed.

What's really remarkable about these sessions is their length. We try to be very respectful of time, says APS CPE project manager Christine Dicken. We keep the sessions down to two hours.

Dicken adds that the CPE team writes up the session outcomes, distributes them to project participants and then gives everyone a week to think about what's been done already. It helps people get thoughts together for more efficient follow-up meetings.

She also notes one of the biggest tasks the CPE tackles is being able to help others discover and map what the process really is. It only works if you create enough trust in the room that people will tell you what's wrong. You need to draw out the issues, the gaps and the opportunities that people haven't had a chance to share.

A burning platform that actually burned

Trouble often precedes an organization's foray into process improvement and the changes it brings. Business leaders sometimes call that trouble a burning platform, in reference to a story about an oil-rig worker who faced a burning platform or a leap into the ocean as his only options for action.

At APS, the process work that earned this utility an Edison Award actually started with a similar platform of sorts. It began with a fire at a key substation.

Although analysis revealed that no maintenance error caused the fire, the APS maintenance team still had to present a corrective plan to regulators. This directive started the maintenance team down its process-improvement path, which resulted in dramatic changes.

Before the process-improvement effort, more than 80 percent of the department's work orders had been reactive, or prompted by some problem, requiring attention, Santanam notes in the case study he presents to classes. After the process improvements, only 20 percent of work orders are reactive. The rest reflect preventative maintenance efforts, which vastly improved the utility's power reliability. In 2008, the number of outages per customer dropped 34 percent compared to the previous year, and the average duration of each outage was down 14 percent.

Work flowed more efficiently, too. In 2005, the maintenance team had more than 1,000 work orders open at any given time. By 2007, that number was down to 200. In 2005, the average age of a work order was 450 days. By 2007 that age had dropped to 100 days.

And, because the maintenance department was preventing outages rather than running around restoring power, the group had time to innovate, Santanam says. That's when they developed their Transformer Oil Analysis and Notification (TOAN) system.

Transformers contain cooling oil that heats up when equipment is pushed to its limits. Before TOAN, utility workers would go to transformer sites and manually test the oil for decomposition gasses, which could indicate problems.

The trouble with this approach was that it was costly and laborious, so utility workers could only perform the tests periodically. The TOAN system automates the process using on-site transformer monitors, a communications network to continuously bring back data and sophisticated algorithms to analyze those data and send alerts to utility engineers when troubles arise.

In developing the TOAN system, maintenance-team managers turned to the CPE for help. They had the technology but they needed a process to implement that technology. That process was brand new, explains Dicken.

Great expectations

Given that a single transformer can cost as much as $6 million and take months to purchase, the $40,000-per-transformer cost of TOAN seems a bargain. This is especially true when you consider that 79 percent of APS's 345-kilovolt transformers are up to 46 years old, and 66 percent of the larger 525-kilovolt transformers are between 21 and 38 years old.

With aging infrastructure to monitor, TOAN's accuracy really pays off. Traditional monitoring methods deliver an accuracy rate between 23 percent and 43 percent, depending on the method. TOAN has proven to be 93 percent accurate in its analysis of potential problems.

These stellar results reflect another reason the CPE is so highly effective. It's not a top-down initiative. The Center gets invited into projects, and the people who do the inviting are also the people who do the day-to-day work, as well as the process-improvement strategy.

According to Sorensen, the CPE's methodology allows workers themselves to discover the processes that exist and then chart out desired outcomes. It's an eye-opening experience, he says. Recalling one process-mapping session, he remembers a supervisor with 20 years of experience who turned to a co-worker and exclaimed, I didn't know we did that!

This supervisor discovered something that process-improvement evangelists like Sorensen have known all along. The knowledge capture of documenting business processes is, by itself, of value, he says. If you don't document your business processes, how can you ever improve them?

Bottom Line:

  • Business-process management (BPM) allows companies to optimize workflows throughout the organization, but it doesn't happen overnight.
  • Smart organizations create an infrastructure for BPM and build competency among employees who do the process-improvement work.
  • Arizona Public Service, an electric utility, has such an infrastructure in its Center for Process Excellence (CPE).
  • The CPE has developed a methodology for consulting with other groups within APS, and the group has chalked up remarkable successes.
  • CPE involvement helped APS earn industry recognition among utility trade groups and other professional organizations.