50 Shades of (Grey Scaled Graphics)

ucds 50 shadesWhy Switch Your Rainbow Graphics to a Boring Shade of Grey? First of all, you're not just changing the background color on your existing graphics, your reducing the number of graphics by incorporating human factors into the design, using analog objects, trends, and regrouping objects so that the graphics provide visual interpretation of what's important to the operator. Color is still used, but it is only visible when the operator needs it to be. Just look at the above example, what stands out...color! We realize that many operators are perfectly happy with their existing red and green HMIs and have a problem with change when they can't even imagine why the change is so important. Besides, what benefits do grey backgrounds give me. They have to see a level one display in action. When they see that they used to have to click through 22 screens to get the same information on a 3 second glance at a single screen, they are blown away. Operators need to see it before they jump onboard. This is an important step to create a culture change in the control room and a great way to introduce human factors and best practices. Today humans spend almost 90% of their time staring at screens between PC monitors, phones, i-pads, and TV's. The average American watches up to 3 hours of TV per day and spends close to five hours on their smart phones each day. While on the phone, watching TV, or even watching a funny video we miss things going on in the background, sure it's not a big deal during a movie or funny video but for a control room operator, failure to notice a runaway reaction could have severe consequences. You've seen the same movie more than once and always notice something you missed the first time. Operators can't afford to miss important changes in their process and have to be prepared for the worst situations at all times. That is why it is so important that we reevaluate how data is presented and understand human limitations so we have no excuse or reason for missing an important event. Make it impossible to error. We have talked to hundreds of managers that are considering the change to high performance graphics and the same questions always come up: Aren't we just changing the backgrounds to greyscale? Not if you're doing it correctly. You have to perform a task analysis to identify what operators do, what they need to know, understand their limitations, and turn data into presentation. Get away from using P&ID's and focus on the "objectives" and "the need to know information" that allows the operators to achieve those objectives. Why do you think operators request new alarms, it's because they need to know when things happen. Instead of giving them alarm, we use alarms to figure out what they are missing on the graphics. Take a list of all the alarms that do not require operator action, ask them why they have the alarm, if it's important information, put it on the graphic in a way that allows the operator to capture that information. We often take the most critical alarms and most important information that an operator needs, then we design that information into an overview screen. A properly designed overview screen can substantially reduce the total number of graphics that operators have to navigate through. Besides improving situation awareness you also decrease engineering overhead. Alarm management and HMI design should be done together. You can't take away important alarms during an alarm rationalization without providing that information on the graphic, if it's important to the operator it should be visible. We have seen operators that have embraced our designs and are now five times more likely to identify abnormal process situations before the alarm is activated. Just think of that. You can have operators that maintain a steady state, a world where alarms are very rare and investigated on a case by case basis. We have some clients that don't get alarms very often and when they do they look at them as opportunities to improve the graphics. Should the operator have seen that coming, if so, why didn't he or she prevent the alarm? What if the operators hate the new graphics and want to go back? Sure, most people don't like being out of their comfort zone but when we design the screens we do it with the operators, they are driving the change, we guide them based on experience but they feel like they designed them. It's hard to change back to something that you yourself designed. Most operators love the exercise and really get into it, they take pride in the new interface, they embrace the new objects and have fun designing. Once they grasp the concepts and learn the rules they want to run with it. We keep them focused on the guidelines and often refer back to the HMI philosophy and style guide to ensure that situation awareness is never compromised. It helps to remind everyone that this has been tested and proven. The Abnormal Situation Management research reports on the performance of old graphics versus the new HP Graphics:
  • Operators are five times more likely to detect abnormal situations before alarms occurred
  • 22% increase in success rate for handling abnormal situations
  • 41% decrease in time from alarm to resolving the abnormal situation
In the last 20 years, HP graphics have come a long-long way. Most of the new things we see are from our past clients that have taken the graphics to higher levels. One Operator said "I can't believe I actually used to operate with those graphics, I was so out of the loop with what was actually going on, I relied on alarms, and spent all my time going through them. Trying to keep the process steady was challenging, now I actually know the status of all the equipment and processes and can see issues forming. I can see how close I am to an alarm and where my operating limits are, much like flying a plane on auto pilot". This looks like a big investment, is this going to take up a lot of resources? So how do you eat an elephant? One bite at a time. Some of the biggest results come from developing level one overview displays, the design taking just a week per operator position and the development a few weeks later. Some of our customers only developed Level one and Level 2 graphics, which reduced the total number of graphics they have by as much as 40% and reduced navigation clicks by as much as 60%. The improvements in situation awareness are incredible and they made these changes by focusing on one console position as a pilot, made adjustments and learned everything they needed before going to the next console months later. Today we have technology that allows you to use a software graphics package that has all the high performance objects ready to snap onto any DCS or SCADA system. We recommend you get the operators together to see a before and after, let them see an overview design of their processes, they will be the driving force and will embrace the change. Educate, Design, Develop, Test, Repeat