Human errors during operations, probably more clearly referred to as human or action failures, play an important role in causing industrial accidents. The assessment of human performance, through the identification and measurement of human failures, is a complicated, but essential, task to accomplish in real process plants. Virtual Reality (VR) provides a suitable mean to identify human failures, measure human performance and train field operators to risky situations. Nevertheless, not all the aspects relevant to Human Factors (HF) can be easily identified, assessed and reproduced in Virtual Environments (VE). Indeed, VR seems to be better suited to measure cognitive capabilities, such as Command, Control, and Communication capabilities (commonly referred to as C3 capabilities), rather than anthropometric ones like physical coordination, precision in manipulating and ability to reach. Actually, this is certainly not due to the intrinsic nature of VR but, rather, to its current state of development. Industrial environments, if properly recreated in VR, can allow anticipating people behavior, thus enabling to identify whether critical actions have been identified and to measure human performance. Further, by changing in real-time those experiment parameters, such as weather conditions (e.g., wind speed, direction, intensity) and process variables (e.g., pressure, flow rate, heat duties), the strength of environmental stressors, singularly or in a combined fashion, on cognitive capabilities such as recognition, anticipation, prioritization, and planning, can be suitably measured and assessed. The consequences of actions performed by operators can even be experienced instantly, thus allowing for an incisive and persistent training effect. The manuscript presents an integrated approach to step towards the use of VR to (a) verify whether the identified human failure types are all of those that might occur in reality, (b) identify additional human failure types that might affect plant safety, (c) measure the influence of environmental stressors on human performance. Further, the approach presents a way to collect automatically HF data to be used and manipulated for giving rise to Human Performance Indexes (HPI). Eventually, HPI can then be of real help in supporting decision-making processes for industrial safety.
Towards the Automatic Measurement of Human Performance in Virtual Environments for Industrial SafetyASME 2011 World Conference on Innovative Virtual Reality
COLOMBO, SIMONE;MANCA, DAVIDE;BRAMBILLA, SARA;TOTARO, ROBERTO;
2011-01-01
Abstract
Human errors during operations, probably more clearly referred to as human or action failures, play an important role in causing industrial accidents. The assessment of human performance, through the identification and measurement of human failures, is a complicated, but essential, task to accomplish in real process plants. Virtual Reality (VR) provides a suitable mean to identify human failures, measure human performance and train field operators to risky situations. Nevertheless, not all the aspects relevant to Human Factors (HF) can be easily identified, assessed and reproduced in Virtual Environments (VE). Indeed, VR seems to be better suited to measure cognitive capabilities, such as Command, Control, and Communication capabilities (commonly referred to as C3 capabilities), rather than anthropometric ones like physical coordination, precision in manipulating and ability to reach. Actually, this is certainly not due to the intrinsic nature of VR but, rather, to its current state of development. Industrial environments, if properly recreated in VR, can allow anticipating people behavior, thus enabling to identify whether critical actions have been identified and to measure human performance. Further, by changing in real-time those experiment parameters, such as weather conditions (e.g., wind speed, direction, intensity) and process variables (e.g., pressure, flow rate, heat duties), the strength of environmental stressors, singularly or in a combined fashion, on cognitive capabilities such as recognition, anticipation, prioritization, and planning, can be suitably measured and assessed. The consequences of actions performed by operators can even be experienced instantly, thus allowing for an incisive and persistent training effect. The manuscript presents an integrated approach to step towards the use of VR to (a) verify whether the identified human failure types are all of those that might occur in reality, (b) identify additional human failure types that might affect plant safety, (c) measure the influence of environmental stressors on human performance. Further, the approach presents a way to collect automatically HF data to be used and manipulated for giving rise to Human Performance Indexes (HPI). Eventually, HPI can then be of real help in supporting decision-making processes for industrial safety.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
