An Evidential Reasoning-Based CREAM to Human Reliability Analysis in Maritime Accident Process.

Area of Science:

• Maritime Safety
• Human Reliability Analysis
• Risk Assessment

Background:

• Existing Cognitive Reliability and Error Analysis Method (CREAM) models struggle with uncertainty and linguistic variables in human error probability estimation.
• Maritime accident analysis often lacks a sequential perspective, hindering a comprehensive understanding of accident development.

Purpose of the Study:

• To propose a modified CREAM approach integrating evidential reasoning for enhanced human error probability estimation in maritime accidents.
• To develop a scenario- and barrier-based framework for describing maritime accident development sequentially.
• To improve the quantification of linguistic variables for common performance conditions in human reliability analysis.

Main Methods:

• Modification of the Cognitive Reliability and Error Analysis Method (CREAM) using an evidential reasoning approach.
• Development of a sequential, scenario- and barrier-based framework for accident process modeling.
• Application of the enhanced CREAM and framework to a ship capsizing accident case study.

Main Results:

• The modified CREAM effectively quantifies linguistic variables and handles uncertainty from incomplete information.
• The sequential accident development framework provides a structured approach to analyzing maritime incidents.
• The integrated approach demonstrated successful human reliability analysis for a ship capsizing scenario.

Conclusions:

• The proposed evidential reasoning-based CREAM, coupled with the sequential accident framework, offers a robust method for human reliability analysis in maritime safety.
• This approach enhances the precision of human error probability estimation, particularly in systems with inherent uncertainties.
• The methodology is adaptable for human reliability analysis in various engineering systems facing similar challenges with uncertainty and linguistic data.