Better information for better energy performance
Change how information and users interact to achieve double-digit percentage energy performance improvement
By Stan DeVries
Most energy performance programs are not sufficiently successful. This is what government agencies such as the U.S. Environmental Protection Agency Star and United Nations Industrial Development Organization (UNIDO), and market analysts such as Aberdeen Research and others report. Many manufacturing corporations have focused on mechanical projects, and some have installed additional reports and displays. And yet the common requirement for additional double-digit percentage improvements in energy performance cannot be achieved by implementing more of these activities. But there is hope. Aberdeen Research has shown a strong correlation between good operations performance (high availability, high utilization, and high yield) and good energy performance. UNIDO and Aberdeen Research point to the establishment of an energy culture as a component of success. An energy culture is similar to quality and safety cultures—it is a “new normal” and “how we do things,” not just a series of projects at one point in the company’s history.
The need for better energy information
Energy performance changes in patterns are difficult for organizations to process without a better approach. This performance can change too slowly, too quickly, and often in complex patterns. Add to this the different needs throughout the organization. As one client’s engineer put it: “One user’s signal is another user’s noise.” The “solution” must effectively present appropriate timely and accurate energy management information to the different layers in the organization, in financial terms, which are compatible with and linked to their financial system. While several decision-making theories exist, two common elements are time and resolving decision conflicts. Users must have enough time to observe the performance, orient the information, seek approvals for recommended decisions, and then implement the required actions. While this might seem to be obvious, the consequence within a corporation with multiple sites presents a challenge for many.
Consider the following types of changes that result in significantly different requirements for energy information among various users throughout the corporation:
- Changes that normally take years, such as equipment capacities and efficiency ranges.
- Changes that normally take weeks or months, such as equipment configurations and their maintenance. Depending upon the diversity of energy equipment at a site, some equipment can be used to handle the swings in demand better than others, and in some cases equipment connection changes produce much better energy performance.
- Changes that normally take minutes or hours, such as timing of planned and unplanned shutdowns and startups, and the variability of operating the equipment. Energy-intensive operations can incur significant costs when equipment must be restarted or started too early.
In addition to this variety, energy performance measurement can be a challenge. There is a bad joke that the perfect energy performance is achieved when the site is shut down. How should energy performance be measured? The answer depends upon the user, but most often it needs to be expressed in the context of produced product for a given part of the corporation over a specific time period. This accounts for availability, utilization, and yield, especially for organizations that use Overall Equipment Effectiveness calculations as a key performance indicator.
The above examples of different speeds of change and different contexts justify a very different approach to popular information management strategies in corporations. The contrast might be summarized as follows:
- The traditional approach with the “four any’s”—with some exceptions, any information, at any time, to any user, in any context.
- Many energy performance reports with the “four one’s”—with some exceptions, only one set of information (often site summaries), at only one time (often monthly), for very few users (often regulatory and key stakeholder reports), in only one context (overall energy cost for overall site business performance, often throughput).
- The recommended approach with the “four right’s”—with some exceptions, a multiple simultaneous set of only the right information (sufficient detail and accuracy), at the right time (sometimes ahead of real time), to the right users (many), in the right context (often for major equipment, areas of a site, production lines, etc.)
Why is a different information management approach needed? Two common reasons: to overcome the inherent delays in preventing or effectively minimizing energy performance problems, and to address the variety of energy performance decisions. When a user receives a criticism or a question about energy performance at a site during the previous month, it often is not actionable; special studies can identify causes, but many of the opportunities have been missed, and the data to understand the cause has to be reconstructed. Some sites do not know the price for exporting electrical power until two weeks after the opportunity. Many sites have fuel costs, which change more than once an hour, but live with information systems that only allow monthly adjustments of those costs.
Better information management
Consider the following three examples of better information management for much better energy performance.
Real-time energy management by the entire operations organization: The first example focuses on helping users at all levels of management throughout the operations areas of the corporation to learn how to manage the built-in conflicts of key targets and their measures using the “four right’s” approach. This is based on innovation: a convergence of industrial psychology and game theory. This might not seem to be “technology,” but when applied to information systems, it is. The idea is to change from users browsing summary data “later,” to all users interacting with real-time changes in targets and measures. It might sound “low tech,” but it is advanced. The basic concept was developed by Dr. Thomas Vollmann, who was a professor at Boston University in the U.S. and IMD (business school) in Lausanne, Switzerland. He developed a game theory for organizations, which is simple, yet very effective. Users learn how to “play the game”—adapt to changes in targets (including how to change the targets for others) and balance conflicting measures.
Vollmann’s theory simplifies the management decision process to a triangle of three attributes: the strategy for each manager in the organization and the senior operators; the measures they use; and the “actions,” which include planning. The “measures” attribute is also the targets for their staff. These change as frequently as is needed. This approach is thorough: This is not some ERP table of numbers or some spreadsheet and chart. All users, from the top to near the bottom of the organization, use the same method. Its result is significant, and it has been in use for 17 years. There are multiple measures, but Invensys’ experience has shown that the maximum is four. Other data can be treated as constraints.
In the example of implementation for a portion of energy management, the lowest layer of the asset hierarchy is a set of major pieces of equipment or refinery units within areas. Real-time finance and energy efficiency calculations are performed at this level of detail. Visually, this may seem to be boring, but the results are exciting. The human factors approach agrees with the modern expert on business intelligence and presentations, Dan Roam, author of The Back of the Napkin.
The results are significant. The blue diamonds show a baseline during the first month of assessing performance across a fleet of sites and all operating shifts. Over a 2.5:1 range of energy throughput, the unit cost of produced energy in the utilities area has a wide range. The cost variability is even high at high throughputs. The magenta squares show improvement after the first month of using the information management approach. Cost variability is still high, but the average is close to the minimum. The yellow triangles show improvement after the second month. The difference in performance is dramatic. All levels of management and the senior operators work as a team, and they have learned how to balance the naturally conflicting targets. All this, and with double-digit percentage energy performance improvement.
Right-time energy management by the organization: The second example focuses on helping users at all levels of management throughout supply chain, operations, technical services, and marketing to learn how to adapt their work for energy performance using the “four right’s” approach. This example focuses on “right time” and “right context—consistent with the “four right’s”—for management and technical services personnel who need to make decisions based on analysis of performance patterns.
While the first example deals mainly with targets and measures, this example deals mainly with metering and events. Energy measurement is the foundation, and many implementations lack a consistent, traceable mechanism for managing metering information. Add to this challenge the major contributor to energy performance variability is starting or restarting equipment, which often arises from how production lines are scheduled and how production lines recover from upsets. While many information installations have some type of information hierarchy, they often lack an appropriate, traceable implementation for managing these types of information structures.
Instead of traditional information approaches, which attempt to aggregate data and then hope for users to find event patterns, this information approach automatically captures and structures event information as close to the data sources as possible. Configurable templates of meter and event objects make this approach trustworthy.
A template for configuring behaviors for managing energy metering information is enforced so all energy measurements throughout the sites and the corporation are consistent. Offering a screen for viewing real-time execution of the current event and for viewing previous events is also enforced so all energy events, from the smallest to the largest, throughout the sites and the corporation are consistent. Automating the management of this information greatly improves the usefulness and trust in energy information for managers, technical services, and supply chain professionals and others who help operations to continually improve energy and business performance. Without this level of consistency, detail, and trust, users cannot make effective energy management decisions across the corporation or among a group of sites.
Real-time management of complex energy targets: The third example focuses on helping users in a continuous process—a petroleum refinery—to learn how to manage a complex set of targets and measures, which changes throughout the day, using employees with less experience.
In this example, the complexity of the energy management problem is daunting. There are many simultaneous targets and constraints, and these change within the day—such as periods of high prices for purchasing electricity. The “optimization” word can be challenging for many because it can refer to a variety of techniques, and some of the implementations have not been sustainable by the average technical services team in most sites. Too often, the available technology is not sufficiently configurable and is not sufficiently robust (withstand input errors and optimization calculation problems). And “optimization” is a combination of technologies—advanced regulatory control, model predictive control, and closed-loop first principles online modeling with workflows.
One refinery user summarizes their energy utilities optimization goal to be the meeting of steam, fuel, and electrical needs of a site at minimum cost, taking into consideration the dynamic process, environmental, and operational constraints:
Energy Utilities Cost = Fossil Fuel (Coal, Oil, Gas byproducts, LPG) + Electricity (Imported or Exported)
To achieve this in almost all operating conditions, expertise must be incorporated into the “solution,” such as responding to changes in weather (the onset of rain is a prominent example). This third example is more than facilitating organization learning (first example) or the systematic automation of measurements and events (second example); this requires automating knowledge of conditions—in a configurable way that can be maintained by a broader team with less experience and less specialization.
One challenge in implementing “optimization” is the equipment efficiencies are significantly non-linear, and the “curves” move, especially with weather changes.
The “Steam turbine efficiency and electric power” charts show how steam turbines change in efficiency and conversion effectiveness. A key requirement of trustworthy information management is to automate the adaptation of these curves in ways that can be consistent and maintained by average technical services personnel. The result is a more frequent, more accurate, and more consistent response to changes in conditions (see chart, pg 24).
This is evidence of managing complexity. An experienced senior operator can implement this without the “solution,” but others would not be as consistent. The results diagram shows how the “solution” guides teams through a planned change in conditions. A coal-fired boiler must be shut down, while the utilities area must deal with the changing demands and optimal targets of delivering steam and co-generating electricity. This is “real-time optimization,” a configurable, trustworthy combination of technologies that handle the dynamics and nonlinearities of this information management challenge. The result is double-digit percentage energy performance improvement, which has been maintained by average experience and less specialized technical professionals.
Despite the frequent claims from suppliers and from user champions of pilot projects, most potential users of information systems—from process historians and mobile worker systems to portals and various analytics software—do not have time to use the software. And many reports and displays do not have sufficient information. For example, how can users delay startup of some equipment, or adjust setpoints, based on a monthly site report? Another challenge is energy measurement. Besides instrumentation challenges for energy sources (fuel, steam, power, chilled water, etc.), there is often a critical lack of measuring energy performance context.
Without actionable information, organizational behaviors will not be changed. Without behavior change, an energy culture will not be established. The leading users of successful information management approaches have gained experience over years of using the “four rights” to change the interaction between information and users and achieve the double-digit energy performance improvements beyond equipment projects and customized reports.
Effective information management for energy management is a significantly different experience for organizations. These successful organizations are proactive; teamwork occurs much more often; and the dominant perspectives are time and money, not pounds of steam and megawatts of power. And with these information management methods, manufacturing organizations achieve and sustain significant improvement in energy performance, which come from behavior changes, and produce new energy cultures.
ABOUT THE AUTHOR
Stan DeVries (firstname.lastname@example.org) brings 34 years experience in designing and applying automation and collaboration systems. DeVries has worked in applications engineering, business development, and product management roles in automation and information technologies including instrumentation, distributed control systems, programmable logic controllers, SCADA, maintenance management systems, and workflow systems. A true believer in the power of human-systems integration, he delivers solutions for improved operations, HSE, and business performance. In his current role as director of Energy Management Solutions at Invensys Operations Management, he is responsible for identifying, defining, developing, and helping to implement innovative, yet reproducible energy management solutions for clients in the hybrid and heavy process industries.