Process Data Historian Improves Business
Historian software helps collect and analyze data to improve manufacturing processes and decision-making. That capability can help rewrite the future history of any business, but is especially useful in the competitive market of supplying terephthalic acid (TPA), polyester resins, and polyester staple fibers to the western hemisphere.
Historian software helps collect and analyze data to improve manufacturing processes and decision-making. That capability can help rewrite the future history of any business, but is especially useful in the competitive market of supplying terephthalic acid (TPA), polyester resins, and polyester staple fibers to the western hemisphere. Doing so requires a firm understanding of capabilities and potential. DAK Americas (Charlotte, NC) has gained this advantage by incorporating historian software to track past business trends, as well as bring manufacturing intelligence into the process of improving safety and product quality, while reducing production costs.
Making extensive use of a 1980s vintage Aspen Technology Setcim system to guide and optimize the entire facility, DAK Americas' Cooper River, SC, plant is a prime example of how to exploit a process data historian to use manufacturing intelligence from the engineering level to improve the business as a whole.
The company has been successful in the blending of technology, engineering, and business largely because of its adherence to the DMAIC (Define-Measure-Analyze-Improve-Control) business improvement concept. The historian in use at the Cooper River site serves as the technological anchor to the "define," "analyze," and "improve" components of the concept for the facility.
Years ago, DAK Americas' personnel took the time to "define" what data were needed and how the data could improve operations. Those definitions have changed with time due to increasing requests for additional field and lab data to improve the ability to analyze the process and provide more detailed production and financial accounting. Today, most of the plant's process and lab data finds its way into Setcim, and users regularly ask IT to find ways to import more data from other areas of the plant not currently connected to Setcim.
One key to success at the Cooper River site is that only appropriate data from appropriate units is tied into the historian. For the purpose of tracking quality and performance, Cooper River's historian connects to:
Staples and resins continuous process;
Draw machines; and
Interfaces between the historian and the process give DAK Americas a solid basis for analyzing and improving plant performance. With the standard Setcim GCS user interface, the company implemented much of the functionality that is also easily created with today's more modern historians, such as displaying ad-hoc and preconfigured trends, reports, alarms, live process data, and process graphics. GCS is also used for accepting user input for navigation, data entry, as an output to the process interfaces, or as an output to a Setcim application.
DAK Americas' historian delivers raw data and calculated information to plant operations and enterprise users, although this process is not as seamless and transparent without the benefit of newer Internet communication technologies, such as XML and COM/DCOM. Data are often presented in both graphic and table formats for an easier user overview of the processes.
Beyond these basics, DAK Americas' Setcim platform has provided application support to allow for implementing more process improvements than many process historian users currently implement with state-of-the-art technology.
The company uses a number of independent applications (typically written in SQL, VMS, or Fortran) running as separate processes on the VAX server in combination with the historian. Setcim can execute these applications on a scheduled basis, or on demand through activation of an event, such as an alarm or user request. DAK Americas has made full use of this functionality to run analytical and control applications, and to import and export data to Setcim from custom data sources. Combined with access to the right data and system interfaces, DAK Americas has made good use of its historian by encouraging plant-floor operators to perform much of the analysis formerly done by engineers or technicians.
Plant operators using Setcim to correlate process conditions and suggest changes have been able to pinpoint many improvements for the control of DAK Americas' process. Giving plant operators access to a historian is one area that much of industry is overlooking. The tendency is to use the data historian only at the technical and engineering levels. At the Cooper River site, it is not unusual for an operator to call and make an engineer aware of a problem before it becomes serious enough to have an impact on the process.
With the right data and the right analysis programmed into the Setcim system, much of the process improvement is automated and occurs in realtime. The examples below highlight some ways DAK Americas extracts value from its data historian, beyond data collection.
Closed-Loop Control : Statistical calculations and comparisons an engineer would use in reviewing laboratory test results have been programmed as an application that runs on the Setcim system, thereby automating the process and allowing more real-time use of laboratory data. Referred to as a closed-loop-control system, this system allows Setcim to automatically make process changes based on laboratory analysis. Safeguards are built in to this application to limit acceptable values of laboratory results and the range of output the application can request.
Accounting : The Setcim system allows for hourly, daily, and monthly statistical accounting for easier evaluations of process conditions over extended periods of time. The site can analyze a production run based on daily averages of process conditions during the time of the run, instead of trying to manually interpret individual data points over the same time frame. This reduces the time needed for data analysis and developing standard operating conditions for future production runs of the same product. Statistical data available for these time frames include minimum value, maximum value, total, average, and standard deviation. Using the accounting package, the plant also can track actual production for final inventory controls and ordering of raw materials.
Billing : Metering and billing is a key issue for many industries. Having a historical record of real-time billing meter readings gives the company the ability to immediately identify and correct problems with meters and measurement systems; verify and/or contest supplier billing to ensure it pays only for what is provided and bills customers accurately for what is delivered; and identify and quantify usage cycles—allowing development of programs for energy conservation and lowering production costs.
Human-machine interface : Availability of historian client interfaces and extensive connectivity to process data have allowed DAK Americas to use the system as an HMI for certain processes added at Cooper River over the years. (Since DAK Americas' Setcim implementation features role-based security, plant control functionality is still realized separately from general user access to the system.) One of the latest applications allows for alarms to be automatically adjusted based on process setpoints. This allows for more than one layer of deviation alarms, such that the first alarm can be used as a warning for off-standard conditions and an additional alarm can be used to identify disposition/waste conditions.
Operator training : By having all process data available at the floor level, operators are better able to learn process interactions.
Transitions challenges : Product transition and product wheel manipulation are viewed as two of the most advantageous challenges to overcome. DAK Americas has made improvements simply by having almost all process and lab data available on the same historian and applications, thereby allowing engineers to make reliable decisions. The company's engineering group has reduced transition time as much as 30% between products.
While its current Setcim system works very well for DAK Americas' needs, looming Setcim and VAX obsolescence concerns are driving the need for change. DAK Americas is currently evaluating different modern historians to determine which available platform(s) best meet its current and future process improvement needs at the plant and corporate levels. While the company has come a long way with legacy technology aiding its commitment, vision, and planning, there is much to be gained as the company moves to a state-of-the-art historian architecture. Using modern historian software will allow DAK Americas to more tightly couple plant performance and business decisions by:
Enhancing communication of process data across different levels of the organization, at any desired level of detail;
Communicating data across manufacturing sites to maximize resources and provide a common platform to make decisions;
Providing technology to codify operator actions with standard procedures and practices;
Simplifying future improvements and additions to the system and application architecture while maintaining high reliability and availability;
Leveraging Microsoft Windows .Net technologies and standards as well as other Web-based technologies; and
Using standard reporting and calculation engines.
As DAK Americas' experience proves, historians, even old ones, provide an excellent opportunity for companies to bring the define, analyze, and improve aspects of DMAIC in-line with automated measurement and control to realize sizeable improvements to manufacturing processes across the board.
Greg Huff is project manager for the DAK Americas Quality Systems Initiative which includes process historian standardization and Cooper River Setcim migration; Harold Mynster, the primary plant DCS/Setcim support position for the Cooper River site, also manages the distributed control systems and develops additional applications for the Setcim system; Mike Sadler is a regional manager with Ready Engineering Corp. (
Extract value from projects
Common hurdles that companies often face when extracting value from information-based projects include:
Isolating and quantifying specific financial benefits;
Inconsistent tag naming;
Inconsistent instrumentation and data available from plant to plant;
Limited application development expertise: Those needing the data don't know how to get it; and those with access to data don't know who needs it. Ability to convert raw process data to meaningful and readily useable information is where actual value of a process data historian resides, and few make the most of its available potential;
Data integrity: Making sure that data are accurate and consistently recorded; and
Data traceability: Making sure that raw data used today to make a decision is also available for future reference and use.