Citation

Musharraf M, Smith J, Khan F, Veitch B. Reliab. Eng. Syst. Safety 2020; 194: e106179.

Copyright

(Copyright © 2020, Elsevier Publishing)

DOI

10.1016/j.ress.2018.06.007

PMID

unavailable

Abstract

Offshore emergency conditions are dynamic in nature and personnel on board are challenged with high risk, time pressure, uncertainty, and the complexity of the situation. This paper investigates how different attributes of emergency scenarios influence people's choice of egress route subsequent to training. An empirical study was carried out in a virtual environment (VE) with 17 naïve participants. The participants were trained to muster during emergencies using a lecture based training (LBT) approach. Training sessions in LBT consisted of computer based training tutorials and simulated training scenarios. Participants' performance was then tested in simulated testing scenarios. It was observed that given the same training, people used different sets of attributes to make decisions on the egress route. This can help to diagnose causes of poor performance and to design adaptive training lessons. Such identification can also help in the assessment of the efficacy of the training curriculum, or the pedagogical approach. To evaluate the prediction accuracy of the decision trees, the outcomes were compared to the actual observed outcomes of the participants in scenarios in the testing data set.

RESULTS show an average of 95% prediction accuracy of the decision trees.

Language: en

Keywords

Decision making; Emergency situation; Human factor risk analysis; Human factors