Article

Toward real-time emotion-driven risk assessment for ship operators combining MobileViT neural network and fuzzy logic

Ocean Engineering

Li, Yibo; Wu, Bing; Guedes Soares, C. — 2025 — Elsevier

Key information

Authors:

Li, Yibo; Wu, Bing; Chen, Deshan; Guedes Soares, C. (Carlos Guedes Soares)

Published in

November 15, 2025

Abstract

This paper introduces a novel fuzzy logic model that provides a real-time risk assessment framework sensitive to the emotions of ship operators. A facial expression dataset is developed specifically for the ship bridge environment, utilising the MobileViT neural network, which has achieved a 99.5% accuracy rate. Fuzzy membership functions and rules are established with the guidance of expert knowledge. Finally, the emotional factors are integrated with pilotage and weather data within the fuzzy inference system, enabling the dynamic assessment of emerging risks. A notable feature of the model is the incorporation of a time factor, which transitions the risk analysis from an instantaneous view to a broader time-period perspective, which holds the potential to support a more nuanced and context-aware evaluation. A case study on a Panamanian ship reveals the potential that emotion-driven risk assessment holds the promise for the development of strategies that could mitigate the risk of accidents influenced by human emotions, thereby supporting safer maritime operations.

Publication details

Authors in the community:

Publication version

AO - Author's Original

Publisher

Elsevier

Link to the publisher's version

https://www.sciencedirect.com/science/article/pii/S0029801825018347

Title of the publication container

Ocean Engineering

First page or article number

122150

Volume

339

Issue

Part 2

ISSN

0029-8018

DOI

10.1016/j.oceaneng.2025.122150

Fields of Science and Technology (FOS)

other-engineering-and-technologies - Other engineering and technologies

Keywords

  • emotion recognition
  • risk assessment
  • fuzzy logic
  • maritime safety

Publication language (ISO code)

eng - English

Rights type:

Open access

Financing entity

Fundação para a Ciência e a Tecnologia

Title of the project, award or grant: CENTEC

Identifier for the funding entity: https://doi.org/10.13039/501100001871

Type of identifier of the funding entity: Crossref Funder

Number for the project, award or grant: UIDB-UIDP-00134-2020