Operator Training by Doing was a session at the 2019 ARC Industry Forum. Here was the session’s abstract:
Operator training continues to be very important as the millennial generation assumes operational responsibility for industrial manufacturing. Industry has learned that training by doing is highly effective and different simulation technologies can allow operators to practice doing their job. Training simulators continue to evolve with new 3D virtual technologies. This session will focus on new trends for process, control system, and 3D spatial simulations used for training.
One of the presenters from Georgia Tech and Fortiphy Logic gamified the Tennessee Eastman simulation developed back in the 1990s for cybersecurity learning. The simulation contains an exothermic chemical reaction that was originally designed for process control engineers to better understand this process.
The simulation was built with 3D modeling, open source PLC and Human Machine Interface (HMI) software on a Linux open source operating system.
The simulation contains two input feeds, a product output and a purge valve. Key measurements include reactor pressure and level and computed costs by measurement how much product is wasted through the purge valve.
Scenarios were created on how to attack the simulated process to better understand how to defend against these attacks. Some of the simulated attacks included man-in-the-middle, command injection, false data injection, reprogramming logic (ex. Stuxnet), loading malicious binary payload, password cracking and buffer overflow.
From a how-to-defend standpoint, the simulation included network segmentation based on the ISA 95 reference model and Intrusion Detection Systems (IDS) and Intrusion Prevention Systems (IPS) to detect and/or stop buffer overflow.
You can download this open source simulation in the GRFICS [graphical realism framework for industrial control simulations] GitHub project.
The next speaker from a major global chemical company presented their use of operator training systems. At the heart of these systems is a dynamic simulation of the plant. They quickly discovered that beyond operator training, these dynamic simulations became indispensable in looking for optimization opportunities. New strategies could be tried that never would be tried on the live process due to safety and availability concerns.
The costs of developing the operator training was more than offset by its use in optimization efforts. Having a sandbox area to test new control strategies and other actions help to prove out the improvements before adding to the live control system.
Another presenter focused their efforts on adding alarm descriptions on ways to address alarm conditions by documenting possible causes, consequences and suggested corrective actions. The effort was driven by the large number of retirements occurring in the operations staff. Before retirement, their experience was captured through extensive interviews, documented and added into the control system alarm configuration.