Modelling crew performance variability in emergency situations from simulator data for human reliability analysis
Metadata only
Date
2020-11Type
- Conference Paper
ETH Bibliography
yes
Altmetrics
Abstract
Several international initiatives have been recently undertaken in Human Reliability Analysis (HRA) with the goal to collect data in main control room simulators and provide HRA models with new empirical evidence to inform the error probability estimates. Recent studies highlighted large performance variability across operating crews in emergency scenarios, as a result of factors such as team dynamics, work processes, communication strategies, sense of urgency. The present paper, building on previous works from the authors (Greco et al., 2019, 2020a), investigates how to formally model crew variability aspects by using common behavioural tendencies activated during task performance (e.g. in communication or decision-making). This approach allows identifying the crew features that determine performance in a specific crew task and explicitly model their influence on the error probability. This paper introduces the multi-step methodology to derive set of crew behavioural tendencies from simulator data collection. A numerical application with artificial data shows the implications of modelling performance variability with behavioural tendencies. Show more
Publication status
publishedBook title
Proceedings of the 30th European Safety and Reliability Conference andthe 15th Probabilistic Safety Assessment and Management ConferencePages / Article No.
Publisher
Research Publishing ServicesEvent
Subject
human reliability analysis; nuclear power plant simulator; SACADA; HAMMLAB; HuREX; performance variability; Bayesian inferenceOrganisational unit
03725 - Prasser, Horst-Michael (emeritus) / Prasser, Horst-Michael (emeritus)
More
Show all metadata
ETH Bibliography
yes
Altmetrics