Energy Firm EDP Renewables Deploys BPM System to Maintain Equipment, Collect Unstructured Data

by   |   January 25, 2013 3:44 pm   |   0 Comments

EDP Renewals implemented a BPM system to analyze data about wind mill equipment such as the Blue Canyon Wind Farm in Oklahoma, above. EDP Renewals photo.

EDP Renewables implemented a BPM system to analyze data about wind mill equipment such as the Blue Canyon Wind Farm in Oklahoma, above. EDP Renewals photo.

When your business is scattered to the four winds, keeping track of routine tasks can be a management headache.  That was the challenge facing EDP Renewables North America, with an ongoing need to efficiently schedule and perform both routine maintenance and repair of wind turbines across the U.S.

And, they also needed to incorporate real-time data, the potential for social and mobile platforms – and manage IT changes themselves – while improving collaboration and knowledge retention across the enterprise. After considering a range of options, the company, part of the Portugal-based energy giant EDP Group, created a BPM-based management system that accesses and updates existing data sources and has the capability of adding unstructured text documents, both to prioritize maintenance jobs and enable field workers to create a repository of information about equipment conditions.

A key challenge the system is designed to address is the company’s dispersed labor force. Stephan Blasilli, who runs the BPM project for EDP Renewables, the goal was to better manage 28 wind farms all across the US with a team of just 15 engineers and 100 field technicians. “Taking advantage of the wind’s potential power can be complex; spreadsheets and emails won’t do it, there’s not enough discipline in those tools and they’re only as good as the people using them,” says Blasilli.

Related Stories

Analysts: BPM can make big data analytics easier to adopt.

Read more»

Utility projects applies predictive analytics to slice of Pacific Northwest power grid.

Read more»

Visualizations highlight data for improving manufacturer’s procurement process.

Read more»

Also known as Horizon Wind Energy in the United States, EDP Renewables develops, constructs, owns and operates wind farms throughout North America. Based in Houston, with over 20 offices, the company has developed more than 3,400 megawatts (MW) and operates over 2,800 MW of wind farms. With over 300 employees, EDPR counts on economic efficiency and acceleration of technological developments that make renewable energy generation reliable and competitive for much of its long-term growth.

That translates into a need to get beyond paper and pencil or Excel-type management solutions.  To take the next step, EDP Renewables determined to build a customized system around the Appian BPM platform.

Like most wind energy companies, EDP Renewables has assets in very remote locations dispersed across the United States. Modern technology enables companies to accurately forecast wind speeds and electricity generation potential. But it is still challenging to integrate that knowledge with repair and maintenance activities, and to communicate information across a geographically-dispersed organization, all while dealing with the challenge of working around the availability of resources. Because  wind power generation is a much more changeable process than what traditional energy generation companies face, Blasilli says the company believed it required a new type of software that could handle dynamic case management, prioritization and resolution.

Asking the Right Questions
Before the system was constructed, the design team discussed what questions they needed the system to answer at the end of a successful instance of a process. Then the team designed forms for end-users to input data.

“For our issue management system, we devised a way to rank issues in order of their seriousness, based on some user inputs about how severe the problem is,” says Blasilli. The design team worked to narrow down the list of questions to its essential qualities to make the process as painless for the user as possible. “Fewer is usually better,” says Blasilli.

From that start, EDP Renewables was able to configure the Appian BPM system for using existing data repositories to provide vital statistics on operating performance and other information relevant to end-users create a case for managing equipment in the field.

Now dubbed COBRA (a play on “COlaBoRAtion”), the BPM system “allows us to bring together what would be a disparate array of single use software systems to view data more holistically,” says Blasilli.  “Without a BPM system to structure our process tracking, we would be reliant on a mix of email for communication tracking and approval documentation, an internal shared drive for document storage, multiple databases for data tracking and analysis, and personal notebooks for solution documentation and notes,” explains Blasilli.

As currently configured, COBRA has a dashboard for each issue, which displays information relevant to the case. According to Blasilli, the dashboard serves the needs of the requestor, who can see the solution documentation; the needs of the case workers, who can see detailed information about the causes of problems, any notes posted to the case, and any solution creation documentation; and the needs of the manager, who can see information about reviews and approvals.

“The manager also has a separate dashboard for the system that allows him or her to see information about how their employees are sharing the workload of the department, the case closing rate, and other important metrics to track process performance,” he adds.

Unstructured Data in Field Workers’ Notes
The system also allows the users requesting services to connect related cases, to create a chain of linked issues.

That aspect of COBRA relates to its handling of unstructured data. Blasilli explains that for EDP Renewables, it was important to include the collaboration occurring on a case, including notes posted about a potential solution, “because we wanted to see if more communication traffic about an issue is correlated with an increase in the quality of the solution.” The properties (who, what, where, why, when) of approvals was also a significant piece of data and one that doesn’t typically fall into the realm of big data, he notes.

After accumulating data over a year, Blasilli says the company finally felt comfortable developing a feature of COBRA that helps users identify issues that may have been encountered and addressed previously. “We felt that we needed to have a critical amount of solved cases in order to adequately triage the potentially thousands of different issues that can occur on a wind farm,” he says. “The only thing worse than having no data is having insufficient data to provide the support you need, which leads to bad conclusions and user dissatisfaction,” he adds.

Samir Gulati, vice president of marketing for Appian, calls EDP Renewables’ BPM project, a process automation classic.

Gulati says one unique aspect was how the company sought to reduce process times by incorporating field data and mobile device data. “They found by capturing data faster they could speed up their process,” he notes.

The decision-making was also faster. “They studied the data from the field and developed an algorithm, which they put into the BPM rules engine that automatically prioritized tasks based on economic factors,” Gulati says. For instance, by including data on actual or predicted winds in different parts of the country, EDP Renewables could determine which repairs were most critical to maintaining and generating energy capacity.

Alan R. Earls is a business and technology writer based near Boston.

Correction, January 28, 2012. The original version of this story misspelled the last name of Stephan Blasilli. 

Tags: ,

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>