Business systems designed to report business results to management were commonly referred to as Management Information Systems (MIS) and were typically designed around traditional human schedules with many business functions executed monthly. Automation systems, on the other hand, were designed to control the manufacturing and production processes to maximise efficiency and had to operate in a time frame relative to the operation of the physical process – real-time. This functional separation and these distinctions served industrial operations quite well as long as the time frame of fluctuations in key industrial variables occurred on a greater than monthly basis.
Over the past decade a number of key business variables of industrial operations, which had previously been constants over extended time periods, started to experience more frequent fluctuation rate, often changing multiple times an hour. This began with the deregulation of the electric power grids. Electricity prices, which had been stable for months at a time started to change every 15 minutes or even more frequently. This caused a domino effect to other types of energy, such as natural gas, then to raw materials, and to the production value of many industrial operations. Monthly business measurement systems proved to be inadequate for decision support because the information they generated came too late for effective decisions in most instances.
Measuring business variables in the required time frame
Measuring variables in the time frame relative to the rate at which they may change – real-time – has been a staple for manufacturing and production operations for well over a century. Flows, levels, temperatures, pressure, compositions, speeds and other critical process variables change frequently and must be measured at corresponding frequency if a manufacturing operation has any hope of being controlled efficiently. This has been common practice and a core competency at plant engineering levels for decades. Unfortunately, this same real-time measurement competency is not common in MIS organisations. As business variables started to experience real-time fluctuation the competencies required to address the situation were not in the organisation charged with business management, but what the MIS teams lacked the engineering talent possessed.
The transition to real-time fluctuation of business variables has taken place steadily over a number of years. Because of this gradual transition many industrial business managers did not realise that the profitability of their plants had similarly become out of control. Therefore they often did not immediately realise that relying on the traditional business management approaches was an impediment to increasing the profitability of their operations. Once they realised the problem they were often perplexed as to how to solve it.
Not all variables have the same frequency
Business financials of a company are often presented in the form of a balance sheet or other financial statement and can be quite complex. It is daunting to consider all the variables on such reports fluctuating in very short time frames and how to manage such complex fluctuations. But it is important to understand that not all of these variables experience frequent change. In fact there are only a few critical business variables experiencing less than monthly fluctuation in most industrial operations. Those real-time business variables include energy costs, material costs and the production value of the operation. The primary business objective of any industrial operation is to maximise the production value while minimising the costs of the operation. The two variable costs of operation that experience real-time variability are energy and material costs. So the real-time profit model is actually much simpler than what may have been presumed. It involves maximising production value while reducing energy and material costs in real-time. However, these three vectors do not represent the overall profitability of the operation; rather they are those components that change in real-time. The other key variables typically fluctuate over periods of greater than monthly and can therefore be effectively managed using the traditional MIS systems. The real-time measurement of the variables that comprise these three vectors is referred to as real-time accounting.
As with any optimisation problem there are constraints on the vector lengths. Certainly the physical equipment presents some constraints since there are limits to the capacity and capability of the equipment. In many industrial operations the equipment constraints are not the first encountered. Since the safety of the people, equipment and environment are typically of prime importance in these operations, the safety constraints tend to be encountered prior to other constraints in the production processes. Referring to this model, the primary business measurements that must be made on a real-time basis in order to be able to effectively manage and control the profitability industrial operations are production value, energy cost, material cost and safety of people, equipment and the environment. Providing the real-time accounting and safety measures associated with these variables is a critical first step in enabling improved operational profitability.
Most industrial operations supplement the business measures with sets of operational measures often referred to as key performance indicators (KPI). These KPIs may measure the throughput of a process, maintenance requirements, quality, work-in-process and other important variables across the operation. Many of these KPIs also fluctuate in real-time and, therefore, must be measured on a real-time basis for operational decision support to be effective.
Dynamic performance measures
Teams charged with managing the performance of industrial operations require both real-time business and operational measures of performance to effectively control today’s dynamic industrial operations. Complete dependence on business measures may cause the teams to de-emphasise critical safety and operational conditions, while complete dependence on KPIs may cause them to operate at less that desirable profitability levels. A balanced view is required. The problem is that the sheer number of real-time accounting measures and KPIs can present too much dynamic information to operations personnel and can easily overwhelm them. The solution is to present the measures that are of prime importance to the operation at any time, based on the current operational strategy and to present them in priority order. Experience has shown that operations personnel required to make real-time decisions can typically handle about four measures at a time. Applying a strategic filter to the combined real-time accounting measures and KPIs that prioritised the measures according to the current operating state and strategy can meet this requirement. One example of such a filter can be produced through a Vollmann decomposition process as described in the Business Point of View on Real-time Empowerment. The resulting prioritised measures are referred to as the Dynamic Performance Measures (DPM) of the operation. This description has both operational and business connotations and provides a nice bridge phrase between the traditionally separate functions.
Manufacturing strategies are not as stable today as they had previously been. In interviews with one food company it was discovered that years ago they would establish a manufacturing strategy that would be expected to be effective for decades. Today’s industrial market is much more dynamic. Operational and manufacturing strategies often require adjustment multiple times daily in the most dynamic industry segments, such as power generation, transmission and distribution. Other industries may not be quite so dynamic, but the outside market drivers are certainly causing them to change or tune their operational strategies much more frequently. Each time there is a strategic change the strategic filter for the DPMs must be adjusted to reflect the change. The result will be a new prioritisation of the measures that reflects the new strategy. In this way the operational and business control system can be reprioritised to the current environment. The result is a set of prioritised measures that can be contextualised to empower every person in the operation who has impact on profitability with the exact actionable information they need to make better decisions that drive improved profitability.
Summary
A fundamental rule in control theory is that what is not measured cannot be controlled. This is certainly true. But there is more to it than merely measuring the critical variables to be controlled. They must be measured in a time frame appropriate to how frequently they may change. These rules have been employed for decades for plant level manufacturing process control. In fact, they have also been employed by default for business level control since traditionally most business variables did not change appreciably in shorted than one month time frames.
Today the business of industry is very different and much more dynamic than it has ever been. Continuing to try to manage the profitability of industrial operations on monthly data has proven to be futile. The skills exist in most industrial operations to meet the challenge, but they tend to be in different organisations. Pulling these skills together across the engineering, MIS and business teams is critical to increasing operational profitability. New real-time performance measurement systems based on dynamic performance measures are required to enable both real-time operational and business controls. The challenge may be daunting, but the solution is at hand.
For more information contact Jaco Markwat, Invensys Operations Management, +27 (0)11 607 8100, [email protected], www.iom.invensys.co.za
© Technews Publishing (Pty) Ltd | All Rights Reserved
printer friendly version