The shift towards Maritime Autonomous Surface Ships is a significant development in the maritime logistics industry, with the potential to enhance efficiency, safety, and environmental sustainability. However, the integration of autonomous systems also presents new challenges and risks, particularly in the absence of empirical data for traditional risk assessment methodologies. This research tackles this problem by utilizing the Net-worked Hazard Analysis and Risk Management System (Net-HARMS) method, a systems thinking method that hasn’t been previously employed in examining MASS. The method analyses the risks associated with the EC-funded, H2020, MOSES Project, which included a concept for automating the manoeuvring and docking processes with autonomous tugboats. The Net-HARMS method offers a comprehensive and holistic approach to risk assessment, overcoming the limitations of conventional probabilistic models. By constructing a Hierarchical Task Analysis and a task network, the research maps the system’s operational framework and explores task interdependencies. The use of a risk mode taxonomy allows for the identification of task-specific and emergent risks, which are then assessed by utilising the risk matrix of the Risk-Based Assessment Tool developed by DNV, to assess the final risk as a function of the effectiveness of each risk mitigation layer and the severity of the identified task consequences. The findings provide valuable insights into critical tasks requiring enhanced risk control measures and contribute to the development of safety constraints necessary for the successful implementation of autonomous shipping technologies. By applying Net-HARMS method to the realm of autonomous ships, this research not only fills a significant gap in maritime risk analysis but also sets a precedent for future studies in this rapidly evolving field.