Babji Srinivasan

Operators' mental models play a central role in safety-critical domains like the chemical process industries. Accurate mental models, i.e., a correct understanding of the process and its causal linkages, are prerequisites for safe operation. Mental models are often defined and explained in abstract terms that make their interpretation subjective and prone to bias. In this work, we propose a Hidden Markov Model (HMM) based formalism to characterize control room operators' mental models while handling abnormal situations. We show that a suitable HMM representing the operator's mental model – including the states, state transition probabilities, and emission probability distributions – can be identified experimentally using data of the operator's control actions, eye gaze, and process variable values. This HMM can be used for the quantitative assessment of operators' mental models as illustrated using various case studies. We discuss the potential applications of the model in identifying various cognitive errors and human reliability assessment. In Part 2 of this paper, we use the proposed approach to assess operators' learning during training.

Process plant operators rely on their knowledge of process cause-and-effect relationships during abnormal situation management. Novice operators develop such process knowledge during training. Hence, holistic assessment of operators’ training is essential to ensure process safety. Currently, during training, operators’ process understanding is evaluated using criteria such as successful completion, task based measures, and operator actions that ignore their cognitive behavior. In this work, we propose an eye-tracking-based approach that uses the operator's attention allocation during different pre-specified training scenarios along with process data, alarm information, and operator actions. Our approach is based on the precept that an operator would focus their attention on those variables on the human-machine interface that they believe have a direct causal relationship to the situation at hand. Also, expert operators seek time-based information for proactive monitoring. Accordingly, to quantify the progress of a novice operator's learning, we develop two metrics — association metric and salience metric — using correspondence analysis of operators' eye gaze. To evaluate the applicability of the metrics, we conducted experiments with ten participants who performed 486 tasks. Statistical studies reveal that the proposed metrics can quantify operators’ learning and thus can be used to objectively evaluate training effectiveness.

Process industries are highly hazardous, and these hazards often lead to accidents. Over 70% of these accidents are attributed to human errors. With the advancements in technology and changing role of operators to the one involving an emphasis on cognitive aspects, most of these errors occur due to limitations in cognitive performance. One of the major constructs to understand cognitive performance is the cognitive workload. An increase in cognitive workload often leads to degradation in performance. Eye tracking has been used in several domains to assess cognitive workload. In this work, we propose a methodology to assess cognitive workload of control room operators during tasks that involve tackling process abnormalities. The methodology employs the fusion of metrics obtained from pupil and gaze data. Our results reveal that fusion of metrics provides better accuracies of classifying cognitive workload at three levels—low, medium and high workload.

Human performance is critical to ensure safety and health in hazardous chemical settings, wherein the majority of the accidents have been attributed to human error. As a result, both industry and academia have been concerned about the efficacy of knowledge and skill transfer. Virtual reality (VR) technology is gaining attention as a way to improve human performance. In this paper, we review the literature on the applications of VR to chemical safety in laboratories and industries. Our review reveals that VR offers much potential for training lab users and plant operators as well as for bridging the theoretical knowledge-practical skills gap. However, there is a need to develop systematic approaches to measure the effectiveness in achieving the desired training outcomes. Therefore, in this paper, we discuss the best practices for VR-based training. We also stress the need to incorporate physiological sensors into the VR environment.

Human errors are a root cause of majority of accidents occurring in the process industry. These errors are often a result of excessive workload on operators, especially during abnormal situations. Understanding and measurement of cognitive workload (overload), experienced by human operators while performing key safety critical tasks, is thus important to the understanding of human errors. Subjective measurements of workload are often not reliable and there is a need for physiological based parameters of workload. In this work, we propose a methodology to measure cognitive load of a control room operator in terms of a biomarker, specifically theta/alpha ratio, obtained from a single electrode EEG signal. Real-time detection of the biomarker can enable minimize errors and improve safety.

Safety is paramount aspect of any chemical plant. In this paper various approaches to enhance process safety are evaluated. The specific enhancements include process design methodologies for improving inherent safety and cognitive engineering to reduce human errors. Their aim is to reduce the number and the consequences of possible deviation events, which depends predominantly on quality of the equipment and human error potential. The consequences are linked to the substances and their inventories. An inherent safety index is used to assess the properties of substances and process units while the potential for human error is characterized using various physiological measures. Our research indicates that application of process synthesis methodologies for simultaneous inherent safety assessment and advanced cognitive engineering approaches for human error reduction will lead to enhanced process safety.

Operator training is critical to ensure safe operation in safety-critical domains such as chemical process industries. Training enhances the operator's understanding of the process, which is then encapsulated as mental models. Typically, the operator's learning in traditional training programs is assessed using expert judgment or in terms of process- and operator action-based metrics. These assessment schemes, however, ignore the cognitive aspects of learning, such as mental model development and cognitive workload. The HMM-based model proposed in Part 1 offers a systematic way to quantify operators' cognition during abnormalities. In this Part 2, we show that the cognitive behaviors displayed by expert operators can be represented as target values on the HMM's state transitions and emission probability distributions. Further, we propose two axioms of learning that can capture the evolution of the operator's mental models as they learn the causal relationships in the process and gain expertise in handling abnormal situations. We validate the proposed axioms by conducting training experiments involving 10 participants performing 486 tasks. Our results reveal that the axioms can accurately assess the progress of operators' learning.

In modern plants with high levels of automation, acquiring an adequate mental model of the process has become a challenge for operators. Studies indicate that sub-optimal decisions occur when there is a mismatch between the demands of the process and the human's capability. This mismatch leads to high cognitive workload in human operators, often a precursor for poor performance. Recently, researchers in various safety critical domains (aviation, driving, marine, NPP, etc.) have started to explore the use of physiological measurements from humans to understand their cognitive workload and its effect. In this work, we evaluate the potential of EEG to measure cognitive workload of human operators in chemical process control room. We propose a single dry electrode EEG based methodology for identifying the similarities and mismatch between the operators’ mental model of the process and the actual process behaviour during abnormal situations. Our results reveal that SƟ(ω), the power spectral density of theta (ɵ) waves (frequency range 4–7 Hz) in the EEG signal has the potential to identify such mismatches. Results indicate that SƟ(ω) is positively correlated with workload and hence can be used for assessing the cognitive workload of operators in process industries.

While modern chemical plants have numerous layers of protection to ensure safety, the human operator is often the final arbiter, especially during abnormal situations. It is therefore not surprising that when operators lose control over the plant, undesirable consequences including property damage, injury, and sometimes loss of lives follow. It is therefore important to continuously monitor the plant operators' situation awareness based on their cognitive state. In this study, we make the first known attempt to infer the cognitive state of control room operators and its evolution over the course of carrying out tasks in a control room. First, we study the operator's actions to distinguish consistent actions from inconsistent ones that allows us to identify major events in the evolution of their cognitive state. Next, we conduct experimental studies with human participants and explore the evolution of their cognitive state through patterns in their eye tracking data. Our studies reveal that two eye tracking measures, fixation duration and saccade duration, are sensitive to the cognitive state and can be used to monitor control room operators and thus prevent human error.

Process industries rely on effective decision-making by human operators to ensure safety. Control room operators acquire various inputs from the DCS, interpret them, make a prognosis, and respond through appropriate control actions. In order to perform these effectively, the operator needs to have appropriate mental models of the process. Poor mental models would increase the operator's cognitive workload and make them prone to errors. Traditionally, operator training systems are used to help operators learn appropriate mental models. However, performance assessment metrics used during training do not explicitly account for their cognitive workload while performing a task. In this work, we demonstrate that this leads to an incorrect assessment of operators’ abilities. We propose an Electroencephalography (EEG) power spectral density-based metric that can quantify the cognitive workload and provide detailed insight into the evolution of the operator's mental models during training. To demonstrate its utility, we have conducted training experiments with ten participants performing 438 tasks. Statistical studies reveal that the proposed metric can quantify the cognitive workload and therefore be used to assess operator training accurately.