An Object-Oriented Bayesian Network (OOBN) Prototype for modeling the safety risk of an unmanned rotorcraft

James Luxhoj, Matthew B. Harrell

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

This paper documents a modeling exercise intended to supplement a NAVAIR System Safety Risk Assessment performed at the Lakehurst Naval Air Station. The System Safety Risk Assessment focused on an event that involved a mode mismatch for the Guidance, Navigation and Control (GNC) software for an MQ-8B Fire Scout unmanned rotorcraft which had potential to result in catastrophic consequences. A modeling exercise was performed in parallel to the traditional NAVAIR System Safety Risk Asessment as part of an Office of Naval Research (ONR) Summer Faculty Fellowship. The purpose was to explore the use of an Object-Oriented Bayesian Network (OOBN) as a possible supplementary tool for NAVAIR System Safety analysts. The Hugin Bayesian Belief Network (BBN) software tool was used to construct the OOBN. A probabilistic model supports making inferences about changes to the states of the causal factors or the presence or absence of mitigations. The Hugin BBN software tool facilitates the computation of sensitivity values based on perturbations to the contributing factors identified in the conditional probability tables. This leads to identification of the most sensitive causal factors with respect to a Mishap probability. Lessons learned from the modeling exercise regarding the analytics for constructing the OOBN are provided.

Original languageEnglish (US)
Title of host publicationIIE Annual Conference and Expo 2015
PublisherInstitute of Industrial Engineers
Pages183-192
Number of pages10
ISBN (Electronic)9780983762447
StatePublished - Jan 1 2015
EventIIE Annual Conference and Expo 2015 - Nashville, United States
Duration: May 30 2015Jun 2 2015

Publication series

NameIIE Annual Conference and Expo 2015

Other

OtherIIE Annual Conference and Expo 2015
CountryUnited States
CityNashville
Period5/30/156/2/15

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Industrial and Manufacturing Engineering

Keywords

  • Bayesian network
  • Safety risk
  • Unmanned aircraft system

Fingerprint Dive into the research topics of 'An Object-Oriented Bayesian Network (OOBN) Prototype for modeling the safety risk of an unmanned rotorcraft'. Together they form a unique fingerprint.

Cite this