A Review on Risk Assessment Methodologies of Decision-making for Virtually Coupled Trains

Abstract

When virtually coupled trains are applied in the real world, there is a need to consider the associated risks stemming from unknown and unforeseen situations. This then requires decision-making systems of virtual coupling to be able to make appropriate decisions autonomously in the face of environmental and behavioral uncertainties and, more importantly, be able to perform appropriate risk assessments prior to decision-making. We provide an overview of risk assessment methodologies for virtually coupled trains decision-making in terms of both quantitative and qualitative analysis, respectively. Among them, the quantitative analysis methods can be further divided into three parts: risk identification, risk measurement and risk reasoning. By comparing the differences of each method, we find that the probabilistic approach can better handle the uncertainty of the input information of the decision-making system. Finally, we propose future research directions.