Assessment of virtual reality based safety training simulator for electric overhead crane operations

• Published in: Safety science 2021-07, Vol.139, p.105241, Article 105241
• Main Authors: Dhalmahapatra, Krantiraditya, Maiti, J., Krishna, O.B.
• Format: Article
• Language: English
• Subjects: Analytic hierarchy process; Computer applications; Concordance measure; Cranes; Hazard assessment; Hazard identification improvement index; Hazards; Integrated iron and steel plants; Occupational safety; Operators; S/N ratio; Safety; Safety management; Safety training; Simulation; Simulator effectiveness; Simulators; Training simulators; Virtual reality; VR simulator
• ISSN: 0925-7535; 1879-1042
• DOI: 10.1016/j.ssci.2021.105241

•Assessment of virtual reality (VR) based immersive safety training simulator•Analysis on presence, system usability, and simulation sickness questionnaires•Hazard identification improvement index for evaluation of training effectiveness•A novel ‘signal to noise (S/N) ratio’ approach for VR simulator effectiveness assessment In this paper, we have assessed the efficiency and effectiveness of a developed immersive VR-based safety training simulator that allowed procedural training to electric overhead crane (EOT) operators. The primary objective of the training simulator is to help the novices to understand the sequence of operations as well as manage the potential hazards while working. Two types of assessment are carried out: simulator effectiveness and safety training effectiveness. We have developed three experimental setups, namely practice environment, familiarization environment, and virtual workplace environment, for improving acquaintance, learning of navigation and control, and operational and safety training, respectively. 19 operators from an integrated steel plant have participated in the experiments; among them, 16 are young operators (average age: 32.68 years, average experience: 4 years) and 3 are experienced operators (average age: 49.66 years, average experience: 12.5 years). The participants have provided their responses on the effectiveness of the simulator using the presence, system usability, and simulation sickness questionnaires. Cronbach’s alpha (>0.7), concordance (>0.5), and convergent validity (>0.7) show that the responses are consistent and reliable. Further, the t-test shows that the simulator is equally effective for both young and experienced operators. For validation purposes, the proposed VR simulator’s performance is compared with the desktop simulator using t-test and signal to noise (S/N) ratio and the results show that the VR simulator is better than the desktop simulator. Finally, the safety training effectiveness is assessed based on the improvement of identifying accident path elements for five tasks. 18 hazardous elements (HEs) and 14 initiating mechanisms (IMs) were identified by the operators before VR-based safety training and additional 10 hazardous elements (HEs) and 13 initiating mechanisms (IMs) were identified after the training. Further, hazard-identification improvement index is proposed as a quantitative measure for safety training effectiveness. Then, task module prioritization analysis using analytic h