Towards system-theoretic risk assessment for future ships: A framework for selecting Risk Control Options

Research output: Contribution to journalArticleScientificpeer-review

35 Citations (Scopus)
140 Downloads (Pure)

Abstract

While the concept of smart shipping is expected to shape the future of the maritime industry, its safety is still a major concern. New risks might emerge when shifting from human controllers onboard, to autonomous software controllers and remote human controllers. The uncertainties associated with the emerging risks require an efficient decision-making methodology to ensure ship safety. This paper proposes a framework for selecting Risk Control Options (RCOs) of ships with higher degrees of autonomy in the context of marine risk assessment and Formal Safety Assessment (FSA). The framework uses the System Theoretic Process Analysis (STPA) for the hazard analysis and the identification of RCOs, while Bayesian Network is employed in the framework for estimating the system risk. Integrating STPA and BN offers the possibility to cover most of the steps of both risk assessment and FSA and permits the prioritization of the identified RCOs. The proposed method is applied to a concept of an autonomous seawater cooling system (SWC) as an illustrative case study. The results indicate that the RCOs including sensors health monitoring and software testing should be prioritized to reduce the risk. This is unveiled by the STPA analysis which shows the risk contribution of the associated causal scenarios.
Original languageEnglish
Article number111797
Number of pages15
JournalOcean Engineering
Volume259
Issue number111797
DOIs
Publication statusPublished - 1 Sept 2022
MoE publication typeA1 Journal article-refereed

Keywords

  • Risk control options
  • Bayesian network
  • STPA
  • Autonomous ships
  • Marine Formal Safety Assessment
  • Risk-based design

Fingerprint

Dive into the research topics of 'Towards system-theoretic risk assessment for future ships: A framework for selecting Risk Control Options'. Together they form a unique fingerprint.

Cite this