The sustainable development of the maritime supply chain is an undeniable trend. Low-carbon port operations are a vital component of creating an eco-friendly maritime supply chain, requiring substantial investments in technologies that reduce carbon emissions. However, the key factors influencing investment decisions by ports and shipping companies in these green technologies, particularly government subsidies, remain poorly understood. Hence, this paper proposes a game-based framework to explore the impact of government subsidies. Through numerical analysis, this study first demonstrates that the pricing decisions, investment level, and profits of ports and shipping companies are sensitive to government subsidies and low-carbon preferences of the market; however, the influence of government subsidies and low-carbon preferences varies with different adopted investment strategies. Furthermore, investment decisions are mainly influenced by investment costs, low-carbon preferences, government subsidies, and cost-sharing ratios. Ports are more sensitive to government subsidies and low-carbon preferences while shipping companies are more sensitive to government subsidies and cost-sharing ratios. In addition, government subsidies and low-carbon preferences are substitutes for each other and can balance cost-sharing ratios between ports and shipping companies. Finally, recommendations are provided to the government, ports, and shipping companies for promoting low-carbon port operations based on the findings of this study.

Accurate vessel traffic flow prediction provides an important data basis for intelligent shipping management, including maritime traffic monitoring, navigational risk awareness, waterway organization, and emission-related assessment. Although recent studies have advanced spatiotemporal, graph-based, and hybrid forecasting methods, improving the predictive ability of a conventional ConvLSTM backbone without introducing substantially more complex model structures remains underexplored in grid-based waterway scenarios. This study proposes a lightweight CA-ConvLSTM framework for grid-level vessel inflow and outflow prediction. AIS-derived flow data and MERRA-2 meteorological variables are rasterized onto a common spatial grid and fused at an early stage. A residual dilated convolution module with dilation rates of 1, 2, and 4 is used to extract multi-scale spatial dependencies, and a channel attention mechanism is applied before ConvLSTM-based temporal prediction to adaptively reweight the fused flow-meteorological feature channels. Experiments using AIS and MERRA-2 data from the northern Bohai Strait waterway show that the proposed framework improves baseline ConvLSTM performance. Compared with ConvLSTM, CA-ConvLSTM reduces MSE and MAE by 24.93% and 12.55% for outflow prediction, and by 24.80% and 12.82% for inflow prediction. These results suggest that spatially aligned meteorological fusion, multi-scale spatial feature extraction, and channel-wise feature weighting can effectively enhance ConvLSTM-based grid-level vessel traffic flow prediction without relying on complex model fusion or heavy graph-based architectures.