Abstract
Numerous accident investigations highlight lack of proper overview displays and alarms as root causes or contributing factors to accidents. One example of this is the Texas City incident in 2005, in which “the control board display did not provide adequate information on the imbalance of flows in and out of the tower to alert the operators to the dangerously high level” (Source: U.S. Chemical Safety and Hazard Investigation Board, Final Investigation Report, March 23, 2007).
Industry has spent billions to design and install automated safety systems, in accordance with highly detailed standards such as ISA-84 (IEC 61511), Safety Instrumented Systems. Despite these efforts, the accidents continue and are often attributed to human error. In many companies, management is highly concerned at verifying, at all times, whether the processes are within a variety of acceptable boundaries.
While operating inside safe boundaries sounds simple, modern control systems (e.g., DCS, SCADA) are not designed to track boundaries other than process alarms. Indeed, alarms setpoints and activation rates are enough of a challenge to control; visualization, management and control of operational boundaries are even more complex.
Consolidating operational boundaries is difficult because the information resides in multiple databases, or worse – in hard copy files. Additionally, capacity “creep” and debottleneck activities or minor betterments will change the throughput of the process, pushing the process closer to or beyond the original design limits. From an operational perspective, the process needs to be compared to these documented limits in real-time for effective operator situation awareness. The actual operational information resides in the automation system, making this comparison a challenge. How can an operator recognize approaching operational, design or safe limit violations in a timely manner without adding excessive alarms? Moreover, how do violations of the limits get logged, tracked, and investigated to prevent recurrence? Managers, engineers, and operators are responsible for making sure that easily-changed automation systems remain both configured and operated within appropriate boundaries. In this paper, we discuss new technology and methods for aggregating, analyzing, depicting, and controlling process boundary information to increase awareness of the operator while enabling engineers and managers to ensure that the process is always within safe limits.
