Data historian's role in plant performance expands in an energy-management era
Two men were walking into the opening session of the recently held OSIsoft user conference. One turned to the other and said, “He's supposed to spend his time solving problems, not building spreadsheets.” Inadvertently or not, these gentlemen had homed in on not just one of the conference's themes, but of OSIsoft's mission as vendor of the PI System data historian, which collects ...
Two men were walking into the opening session of the recently held OSIsoft user conference. One turned to the other and said, “He's supposed to spend his time solving problems, not building spreadsheets.”
Inadvertently or not, these gentlemen had homed in on not just one of the conference's themes, but of OSIsoft's mission as vendor of the PI System data historian, which collects data from heterogeneous systems and places it in a context meaningful for decision makers.
With the second gentleman nodding in agreement, the two were giving witness to a well-known fact: The most commonly used decision-support tool in virtually all functional areas of the manufacturing enterprise remains the Microsoft Excel spreadsheet. It attained this lofty stature despite its recognized limitations as a means to integration and collaboration.
New capabilities from Microsoft and OSIsoft allow engineers and managers to keep their spreadsheets, but rigorously managed in a shared environment and supported by tools that will allow, hopefully, overly detail-oriented managers to avoid pointed criticism.
More important, though, is knowing that the term “data historian” doesn't do justice to the pervasive role such systems are poised to play in today's most dynamic industries, nor does it hint at the big part these systems will have in initiatives ranging from demand-driven manufacturing to plant energy management. (For more on Excel use, see OSIsoft brings elements of Microsoft Business apps into PI System ).
With this expanding role for data historians in mind, some have suggested that today's most advanced systems are better referred to as “enterprise historians.”
The PI System data historian helps Microsoft manage energy use in the midst of a massive buildup of its data centers, needed to furnish Internet services to consumer customers. Eastman Kodak uses it to manage processes and reduce energy costs at The Kodak Park, its Rochester, N.Y.-based “city within a city.” And PI System is the corporate standard for data historians at bio-tech manufacturer Amgen 's six production plants.
Some distinctions made
Since at least early 2005, Alison Smith and Colin Masson of AMR Research have been calling attention to the big role data historians will play in an era where multi-plant global operations are increasingly common.
According to AMR, data historians originated as special-purpose data repositories for environments in which vast quantities of time-series or temporal data are acquired and stored at very high rates. Historians allow users to archive and retrieve years of data in a fraction of the time of a conventional relational database.
That led automation vendors to bundle data historians into DCS and other type control systems, and they are available as stand-alone solutions from application vendors as well.
At this level, says Rick Bullota, VP and CTO at automation software vendor Invensys Wonderware, “Data historians are purely a commodity.”
Yet it could be said that OSIsoft made the market for enterprise data historians, and for its 25 years of efforts the company today has annual revenues exceeding $100 million, while remaining privately held.
Even then, the term “historian” is inappropriate to the extent the systems have evolved to become a real-time data-delivery mechanism, performing mathematical operations that aggregate and transform acquired data into operational intelligence used for decision support.
Moreover, says AMR, a historian today “provides a data abstraction layer between the real-time process realm and the more ponderous world of transaction-oriented business and financial systems… [and] offers a platform for aggregating, consolidating, and recalibrating data from various systems in the production environment.”
This is the role of historians, for instance, in plant service-oriented architecture (SOA)-based software suites for production and performance management.
Scale and fidelity
In practice, PI System is a tool kit applied in environments where deep understanding of real-time data is required. While perhaps brought in for some specific purpose, companies tend to initiate a practice in it and identify more applications over time.
Amgen, the aforementioned biotechnology manufacturer of protein-based medicines with nine global production sites, recently signed an enterprise agreement to expand and standardize its use of PI System across the company.
“Tag- and server-based pricing limited adoption of PI,” says Rob Gamber, principal engineer, Amgen. “The enterprise agreement has streamlined procurement and helped avoid unnecessary cost justifications. We now have the dedicated resources of 'managed PI,' a governance structure, cross-site support, and, to the degree possible, standard architectures, configurations, and processes.”
Biotechnology manufacturing is an inherently variable process, Gamber adds, and Amgen has standardized on PI System since 2001.
“Only recently have we begun to understand the value of process data integrated with other types of information, for things like batch report generation and in-depth analysis and optimization,” he says.
Moreover, by 2006, Amgen realized it had six localized, disparate instances of PI System, with user experience and adoption rate varied from site to site.
The enterprise agreement was the means to address the situation.
That said, there were challenges that had to be overcome in standardizing the architectures, configurations, and processes involved.
“Everybody loves standards, and the best thing about standards is that there are so many to choose from,” says Gamber. Configuration includes the naming standards for tags and classes; equipment models; procedural models; and security.
One of the most interesting emerging applications of PI System is managing energy usage in the massive data centers that companies like Google, Yahoo!, and Microsoft are building to deliver Internet services.
Jeff O'Reilly, a senior program manager with Microsoft Global Foundation Services, says Microsoft entered this world with its acquisition of Hotmail in 1997. Today there are 300 million Hotmail users, involving 500 million unique visitors in a month, and surprising usage spikes when new services are introduced or integrated with the system.
Having gone from management of a few hundred servers involving kilowatts of power to thousands of servers and megawatts of power, the management task has changed significantly.
“To start it was about capacity management—space-based, and with little concern for the environment,” says O'Reilly. “Today we're monitoring a range of interrelated parameters including load, usage, environment, HVAC system, and temperature and humidity.”
The result is the PI System for a data center can involve as many as 400,000 tags, with expectations that this will quickly rise to more than one million tags.
The Kodak Park, located in Rochester, N.Y., often is referred to as a “city within a city.” The site covers 1,300 acres, and includes two utility power plants, two water and wastewater treatment plants, and 150 buildings housing 11,000 employees.
Kodak Park utility power plants have enormous generation output and demand requirements, including two million pounds per hour steam load and a 125 MW electric load. The site also has 600 electric distribution meters, 600 additional nonelectric distribution meters, and many generation site meters.
The utilities systems were operated and monitored by disparate automation and distributed control systems.
According to Stephen O'Keefe, senior systems analyst, worldwide information systems, Eastman Kodak was using OSIsoft PI System to manage some of its production systems when it expanded its use as a means to reduced utility costs, optimization of generating assets, and the consolidation of utilities data from many different legacy systems into a common historian accessible through a Web browser.
Besides PI System, the solution includes the RtPM Business Package for SAP Enterprise Portal (RtPortal iViews); and OSIsoft smart clients. Use of RtPortal iViews allows users to link and correlate PI and production data, displaying it in the corporate SAP NetWeaver Enterprise Portal.
“This has led to tremendous benefits,” says O'Keefe. “Enhanced metering delivered actionable information, and we're able to diagnose anomalies in power usage that have been unexpected—like spikes following from too many fan, pumps, and other type motors being started at one time.”
By putting energy-related information on the portal, says O'Keefe, “We have a way to better understand energy usage over time, and a way to gauge whether agreed-upon conservation steps are being followed. This can lead to meetings with operators to ask them to change their behavior.”
The overall impact of the system is that Eastman Kodak was able, “with confidence,” to shut down one of the two power plants, leading to savings in the millions of dollars and payback in 24 months.
According to O'Keefe, Eastman Kodak continues to expand its use of PI System, incorporating automation and analytic capabilities as it finds new applications in areas such as power purchasing and water utility.
Moreover, “It's a fun application, real-time, and much different than enterprise systems,” O'Keefe concludes.
It seems clear that the possibility of real-time performance management and energy management holds huge opportunity for OSIsoft, and that the challenge will be in meeting demand for what could still be seen as a leading-edge technology. If doubts remain, however, consider that the short list of OSIsoft's major business partners include Microsoft, SAP, Rockwell Automation, and Emerson Process Management. That tells you something of what those situated at the convergence of manufacturing and information technology think about the matter.
Kevin Parker can be reached at firstname.lastname@example.org
OSIsoft brings elements of Microsoft Business apps into PI System
Speaking at the recent OSIsoft user conference, Chris Colyer, WW director for plant operations strategies with Microsoft , said, “Think about how much Excel is used. The Microsoft Office Server System and Office Business Applications [OBA] make it that much more usable because we're putting Excel inside the application. You don't have to move to Microsoft Office to get to it.”
At the conference, OSIsoft announced availability of new releases of RtWebParts and PI DataLink for Excel Services as OBA tools. Also, as part of the OBA platform, OSIsoft is adopting Microsoft Silverlight technology within RtWebParts to offer users more advanced visualization and flexibility in daily business operations.
The PI Visuals and PI Analytics layer, integrated with the 2007 Microsoft Office System, brings operational data and technologies together in real-time, easy-to-understand dashboard formats, supporting real-time collaboration and intelligent actions.
PI DataLink for Excel Services allows applying PI real-time analytic functions using the Microsoft Office Excel 2007 Web part while integrating PI DataLink functions into the RtWebParts environment. This functionality gives users the ability to validate and lock down analyses on the server.
“OSIsoft quickly embraced the benefits that Office Business Applications offer,” said Colyer. “It created products that bring the mountains of data generated by plant applications to life in the familiar Office environment.”
Office Business Applications connect line-of-business systems such as ERP or CRM with the people using them through the Microsoft Office interface.