Carl H. Lindner College of BusinessCarl H. Lindner College of BusinessUniversity of Cincinnati

Carl H. Lindner College of Business

A Dynamic Exchange

Part-time MBA student's learnings lead to substantial, immediate savings for his employer, Duke Energy.

When part-time MBA newcomer Dan Howett enrolled in a UC College of Business introductory business analytics course, he—like most students—could only imagine what the professor had in store for them.

“I wasn't sure what to expect going into the MBA program at UC—it had been twenty-plus years since my last college course,” says Dan.

One thing he certainly wasn't anticipating was a substantial and almost immediate payoff to Duke Energy Generation Services, where he is a station manager.

Duke Energy engineers often collaborate to troubleshoot various issues within their individual stations. Before the course, Dan was part of a team supporting a plant that had noticed a steady decline in thermal efficiency—the plant was slowly requiring more fuel to produce the same amount of steam and electricity as previously. The answer for the inefficiency wasn't obvious and many engineering man-hours were spent poring over data without finding a conclusive cause.

As part of his MBA program, Dan began coursework in Statistics and Decision Models for Managers, a business analytics intro class taught by Professor W. David Kelton. One of the topics covered, multiple regression analysis, is a statistical tool that can help identify which variables are the important inputs in affecting a host of different business outcomes. In the case of the Duke Energy station, the relevant outcome was efficiency.

Intrigued by the possibility of solving his station's inefficiency, Dan built a multiple regression model using relatively basic spreadsheet add-in software supplied with the class.

“I asked Professor Kelton a lot of questions during this analysis, but that's what helped me understand the tool and how to apply it at the station,” he says.

His model tracked the station's efficiency in correlation to the temperatures at various points within its steam cycle as well as temperature drops across many different heat exchangers. The regression results clearly showed the temperature drop across one particular heat exchanger as the critical contributor to the inefficiency.

Armed with this information, the station team disassembled the device—a difficult and time-consuming task—and discovered clogged tubes. Since this type of fouling cannot be detected without taking the device apart, Dan's regression model was instrumental in guiding the team to the faulty exchanger. The team tracked the cause of the fouling and corrected the issue to prevent the inefficiency from occurring again.

“The [program] experience has been amazing,” Dan shares. “Learning and applying regression analysis was just the first of many times where I learned something on a Monday or Wednesday night and applied it the next Tuesday or Thursday morning. The relevance of the course work to real life has been just phenomenal.”

Ever the business analyst, Dan made a cost-benefit calculation for the whole experience. While the exact amount saved by Duke Energy is proprietary, Dan has confidence that his employer's tuition investment was by far one of the highest financial returns of all time.