Distributionally robust battery investment and replacement for AGV battery swapping stations with demand uncertainty in automated container terminals

As the maritime industry accelerates its transition toward decarbonization, electric automated guided vehicles utilizing battery swapping stations (BSSs) have emerged as a critical solution for green automated container terminals. However, the adoption of this low-carbon technology faces dual challenges: high capital investment and significant demand uncertainty. Traditional replacement strategies, often relying on fixed cycles or empirical judgment, fail to adequately account for battery performance degradation and demand fluctuations, leading to resource mismatches and hindering the economic sustainability of electrification. To address these issues, this paper proposes a multi-period decision-making model for optimizing battery investment and replacement strategies under uncertainty. The model manages batteries in age-based groups and optimizes procurement timing and usage allocation to minimize the total operational cost in net present value, ensuring a cost-effective transition to green logistics. To handle demand uncertainty without relying on precise distributional information, we establish distributionally robust chance constraints based on the Wasserstein distance. Furthermore, we propose an approximation method using Conditional Value-at-Risk (CVaR) and derive its closed-form expression through duality theory. Numerical experiments validate the model’s effectiveness. Comparative analysis demonstrates that the CVaR method exhibits superior robustness in extreme demand scenarios compared to expectation-based approaches, providing a theoretical foundation for reliable and resilient energy management in decarbonized terminals.