John Krajewski has recently published a white paper on “Situational Awareness: The Next Leap in Industrial Human Machine Interface Design". This blog post reviews the industrial operators interface evolution and covers his findings on the trends that currently drive the needs of the industrial HMI application.
An Illustration of Industrial Operator Interface Evolution“Over the past several decades the way people have interacted with industrial systems has changed dramatically as depicted in Figure below. These changes were driven by the needs of the operations staff to improve the way that they use, manage, and maintain those systems coupled with advances in technology that facilitated those improvements. Changes like this will continue to naturally occur over the coming decades and will be driven by market needs coupled with advances in technology that address those needs and provide further opportunities for improvements. There are several trends in the current implementations of industrial automation human machine interface (HMI) systems that are driven by the current market needs. The trends that are currently driving the needs of industrial HMI application design are larger systems, greater volumes of data, increased levels of automation, staffing proficiency issues, and expanded use of remote operations. Each of these industry trends poses new challenges that can severely impact the ability for an operations team to achieve optimal business performance of their systems and safe operations.
Larger Systems and the Increased Span of Control
The number of pieces of equipment used in modern industrial systems is continuing to grow as the cost of connected devices continues to fall, the capabilities of control systems to handle more equipment rises, the reliability and bandwidth of networks grows, and the demands of industrial systems are driven by the business needs of larger global markets. While the technology has enabled more and more pieces of equipment to be connected into an integrated system, the user interfaces into these systems have not evolved at the same pace to effectively handle this increase. Modern operations teams are using fewer resources to staff these systems and the span of control of an operator is growing while the techniques he utilizes to manage his system were not designed for such volumes of equipment.
Another key factor in the growth of systems has been the integration of much larger geographic areas into single systems than was possible or practical in the past. These larger systems allow users to make key operational decisions in real time, such as determining which production facility can produce a service or product at the least cost. The costs and reliability of networking such systems together continues to improve and in result these types of systems are commonplace today. Whatever the business driver, the end result of these larger systems is an overload of the operator with much greater volumes of data than they can effectively manage.
Greater Volumes of Data and the Increased Operator Load
Even as the number of pieces of equipment grows, the equipment itself is generating more data. In the past, a single transmitter may have generated only a single value connected into the monitoring system, but modern transmitters have additional diagnostics, onboard control, and many tuning parameters all of which have increased the data density per piece of equipment by multiple orders of magnitude. In many cases the user interfaces that contain this data have not been designed to optimize the operator interpretation of this data and further compounds the operator overload described in the previous section.
Increased Levels of Automation and the Unintended Consequences
In an effort to reduce the variability that human operators can introduce, more and more of the functions performed in industrial automation systems are automated by control loops and process sequences. These control loops and process sequences do offer the operator some relief from the factors already discussed that increase their workload but also have unintended side effects. As the operations teams are rarely involved in the design and implementation of such systems they have little understanding of the actions being taken by the control system and they become disconnected from the process. This can lead to an over-dependency on the system to drive operator behaviors through mechanisms like alarms or process interlocks. It is very common to hear that operations teams are reduced to either resolving interlocks or reacting to process alarms. In this type of environment the operator is performing reactively and cannot prevent disruptions but instead can only react too them when they occur.
Staffing Issues and the Impact on Proficiency
As these systems evolve, and the user interface design techniques are kept mostly the same as has been done prior to these evolutions, it has driven up the amount of time that it takes to bring a new resource on board and make them proficient in utilizing these systems. It is common to hear that it will take about 2 years for an operator to become proficient on a system. This extended period of time is required because the operators need to become experts on the system to make up for deficiencies in the system design. However, other conditions in the market are shortening the length of employment terms. Operations staff have more freedom to seek employment elsewhere, advance through
their organizations and a variety of other causes that result in the average term of employment being near 2 years. This means that the operations teams are rarely at maximum proficiency. Another common concern in nearly every market is the impending retirement of the people who best understand the systems and the need to replace these experts and bring those replacements up to speed quickly. Something must be done to reduce the amount of time taken to achieve both operator proficiency and the variability in the quality of the process from one resource to another.
Remote Operations and the Challenge of Distance
With advances in networking technologies and reduction in costs for these technologies it is becoming more commonplace to remove the operator from the location where the process is actually taking place. This can often be driven by needs such as safety, optimizing staff utilization through increased span of control, or a need to locate the operations where subject matter experts are available. Whatever the reasoning, this separation is presenting further challenges to the operations teams as they can no longer employ the same number of senses as they could when they were located near the actual equipment. Many operators have described being able to understand the equipment and process status through sensing sounds, vibrations and smells alone. When the operations are remote and these additional senses are no longer able to be used, the operations team becomes even more dependent on
the effectiveness of the HMI in communicating the state of the system or process. But too often the user interface has been implemented by recreating and animating the Piping and Instrumentation Diagrams (P&ID). But these P&ID’s were never designed to overcome these challenges and as such this results in operators that poorly understand their systems and how to properly manage them.”
Krajewski, John (“Situational Awareness The Next Leap in Industrial Human Machine Interface Design”