An adaptive dynamic vehicle dispatching policy for intelligent-manual mixed fleet in semi-automated container terminals

This study addresses the vehicle dispatching problem in a semi-automated container terminal with an intelligent-manual mixed fleet consisting of electric intelligent guided vehicles and diesel container trucks. We propose a mixed-integer linear programming model with an adaptive dynamic vehicle dispatching policy to optimize multiple objectives, including energy consumption, transportation task completion time, and workload balance. The proposed model requires only minor configuration adjustments to deploy across three common real-world operational modes, leveraging a target-based optimal solution that dynamically adjusts objective weights for equitable trade-offs. Extensive numerical experiments are conducted to validate the proposed policy against three benchmark policies, demonstrating the superior adaptability and efficiency of our proposed policy in real-time dispatching across diverse scenarios. Furthermore, an in-depth analysis of energy replenishment processes underscores the complementary roles and operational distinctions of the two types of vehicles. Sensitivity analyses further reveal that a 60 % electric intelligent guided vehicles fleet proportion minimizes energy-time trade-offs, and energy consumption increases linearly with a higher export container proportion. Additional robustness tests under stochastic disruptions such as vehicle breakdowns and quay crane outages confirm that our proposed policy maintains stable performance and workload balance even under operational uncertainties. These findings provide actionable management insights for semi-automated container terminal operators to configure an intelligent-manual mixed fleet and adapt dispatching policies across dynamic operational modes, ensuring efficiency, sustainability, and scalability in real-world implementations.