On the reliability of expert’s assessments for autonomous underwater vehicle risk of loss prediction: are optimists better than pessimists?
On the reliability of expert’s assessments for autonomous underwater vehicle risk of loss prediction: are optimists better than pessimists?
Expert judgment elicitation is a key element of formal risk assessment. Some research in this subject has focused on identifying the best way to aggregate expert judgments. In this paper we explore this problem. Given the divergence in expert judgments, when using mathematical aggregation it is possible to group expert judgments according to their mood, this is optimists and pessimists. Using hard data, gathered after the expert judgment elicitation process we test which group performs better. In this paper, we group the expert judgments elicited for building the risk model for the Nereid-UI hybrid autonomous underwater vehicle into optimists and pessimists. After the risk assessment the vehicle conducted 16 missions. We compared the two risk models from the risk assessment against the observed risk from actual missions. Our results showed that differences between the pessimist risk model estimates and the observed risk are not statistically significant. On the other hand differences between the predicted risk using the optimistic risk model and the observed risk are statistically significant. We conclude that for early missions in extreme environment it is imperative to use the pessimistic risk model estimates to inform decision making.
expert judgment aggregation, autonomous underwater vehicles, RISK, risk of loss, Reliability
Brito, Mario
82e798e7-e032-4841-992e-81c6f13a9e6c
16 September 2018
Brito, Mario
82e798e7-e032-4841-992e-81c6f13a9e6c
Brito, Mario
(2018)
On the reliability of expert’s assessments for autonomous underwater vehicle risk of loss prediction: are optimists better than pessimists?
12 pp
.
Record type:
Conference or Workshop Item
(Paper)
Abstract
Expert judgment elicitation is a key element of formal risk assessment. Some research in this subject has focused on identifying the best way to aggregate expert judgments. In this paper we explore this problem. Given the divergence in expert judgments, when using mathematical aggregation it is possible to group expert judgments according to their mood, this is optimists and pessimists. Using hard data, gathered after the expert judgment elicitation process we test which group performs better. In this paper, we group the expert judgments elicited for building the risk model for the Nereid-UI hybrid autonomous underwater vehicle into optimists and pessimists. After the risk assessment the vehicle conducted 16 missions. We compared the two risk models from the risk assessment against the observed risk from actual missions. Our results showed that differences between the pessimist risk model estimates and the observed risk are not statistically significant. On the other hand differences between the predicted risk using the optimistic risk model and the observed risk are statistically significant. We conclude that for early missions in extreme environment it is imperative to use the pessimistic risk model estimates to inform decision making.
This record has no associated files available for download.
More information
Published date: 16 September 2018
Additional Information:
Proceedings International Conference on Probabilistic Safety Assessment and Management (PSAM14), Los Angeles, California
Keywords:
expert judgment aggregation, autonomous underwater vehicles, RISK, risk of loss, Reliability
Identifiers
Local EPrints ID: 425027
URI: http://eprints.soton.ac.uk/id/eprint/425027
PURE UUID: e9bfcd5d-2949-4529-ae89-b02bee865aa0
Catalogue record
Date deposited: 09 Oct 2018 16:30
Last modified: 28 Apr 2022 02:00
Export record
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
