Expanding the pioneer shipping route for vaccine distribution: a reinforcement learning-based approach

Purpose In archipelagic countries, sea transport plays a vital role because islands are not always reachable by air because of the absence of airports. This study aimed to explore the use of pioneer shipping routes for delivering vaccines to these remote locations. Pioneer shipping is a service implemented by the government using “pioneer ships” to improve connectivity. This study explored the utilisation of pioneer shipping for vaccine distribution. Despite its potential, vaccine distribution using the pioneer shipping route is challenged by risks related to sea transport, such as delays, weather, sea conditions and handling errors, which have not been addressed in existing studies. Design/methodology/approach We introduce a new reinforcement learning-based model that optimises delivery while considering vaccine perishability and wave height. The model was tested in actual shipping routes by utilising historical wave-height data. Findings The numerical results indicate that without any optimisation effort, the route can cause 50% damage to the vaccine. Increasing the number of routes can help prevent damage during the voyage owing to the reduced shipping time. By drawing insights from numerical tests, managerial and policy implications can be derived, including the installation of vaccine refrigeration equipment in the vessel. If the equipment is unavailable, it is preferable to use direct shipping from port to port, as our analysis showed that wave variations significantly increase the average travel time and hinder the shipping of the vaccine. Originality/value This paper contributes to the literature by addressing the underexplored challenge of vaccine distribution in archipelagic and remote regions. This paper specifically optimises the utilisation of pioneering shipping routes by proposing a novel reinforcement learning-based vessel routing framework that optimises delivery considering goods’ perishability and wave-induced variations. The modelling framework leverages clustering and attention mechanisms to prioritise the delivery of vaccines across multiple shipping routes, enhancing both efficiency and vaccine preservation.