PTQ Q3 2025 Issue

Operator performance advances enable safer operations

Aligning best practices with industry standards and recommendations while developing a sustainable lifecycle for key aspects that impact long-term operator performance

David Lee User Centered Design Service, Inc. Tim Olsen Emerson Automation Solutions

F or many operating manufacturers, decisions about control room design and operator workload were made during the original plant start-up and commissioning. Those design decisions, including alarm management and human-machine interface (HMI) graphics, may have been developed several years or decades ago, and industry stand- ards and good engineering practices have improved signifi - cantly since then. In addition, operators may have different behaviour expectations based on new digital transformation enhancements, which can be challenging with older designs. A good practice when considering alarm management and HMI graphics is that ‘less is more’. The focus for alarm man- agement should be quality alarms without duplication rather than quantity of alarms. For example, if an abnormal event generated 20 alarms to the console operator, but they only took two actions to avoid a consequence, then the operator received 18 more alarms than needed. Similar to HMI graphics, when considering level 1 process area overviews and level 2 process unit control information to be displayed, consider only information the console opera- tor uses to evaluate the plant performance quickly and effec- tively (not all information available needs to be shown on the main graphics). It is still good to identify a problem within level 1 and 2 graphics and where to go to get more details (level 3 process unit details and level 4 process unit support and diagnostics). The overriding principle for a level 1 graphic is to provide the operator with optimal situational aware- ness. However, as demands on the operator grow, the

adoption of new technology provides them with ever- increasing amounts of data, some of which require additional interpretative skills. It is important to consider how to man- age cognitive workload. One way to achieve this is to build intelligence into the graphics. So, as designs have evolved from simple data to contex- tual data in increasingly complex forms, it is important to consider how to best provide the information in a form that supports intuitive/easy decisions. A good example is a face that is smiling or frowning, from which it is easy to determine good or bad, and, if placed in context of the process, where to go next to diagnose and take action (see Figure 1 ). Consider a car’s dashboard display. Not all information available about the car performance is presented, just the information required to safely operate the vehicle at a quick glance. For example, individual tyre pressures are not pre- sented on the main display, but a low tyre pressure alarm informs the driver of an abnormal condition. Individual tyre pressures can be viewed by going deeper into the displays (similar to level 3 and 4 detailed HMI graphics within the dis- tributed control system, DCS). In the last two decades, there has been a lot of focus on operator performance initiatives with the development of new industry standards and an improved understanding of how to better apply human factors in an operational envi- ronment. There have also been significant advances in ena - bling technologies, most recently under the broad category of digital transformation. A good starting point is first to understand industry standards and good practices and then

Mass balance

04PX1234

08PX0250

0.0 psig

Data

Information

Relationships

Decision support

Cognitive workload

Figure 1 Cognitive workload

PTQ Q3 2025