Mixed traffic conditions of autonomous and human-driven ships: Assessing channel traffic capacity bounds and optimizing channel management

Abstract

As autonomous ships (ASs) advance, mixed traffic with both autonomous and human-driven ships (HSs) will increasingly dominate seaports. Different AS penetration levels affect channel traffic capacity, potentially increasing uncertainty and risks for ships queuing to enter or exit ports. This study explores the complexities of channel traffic capacity in mixed traffic scenarios and proposes optimal management policies. A novel assessment model was proposed to quantify the relationship between AS penetration rate and the upper and lower bounds of channel traffic capacity when human crews exhibit trustful or conservative attitudes towards AS. Moreover, the total sailing duration with mixed traffic flow is analytically modeled for various possible channel allocation policies: (a) mixed channels; (b) dedicated channels; (c) mixed-ASs channels; and (d) mixed-HSs channels. Microsimulation experiments indicate that channel traffic capacity is positively correlated with ASs penetration rates. Our findings also suggest that developing an optimal channel management strategy is closely related to the penetration of AS. The optimal channel allocation strategy is the dedicated channel policy when more than 60% of ships in the mixed-HSs channel or less than 55% in the mixed-ASs channel are ASs.