Ozone pollution has emerged as a pressing issue in urban environments, with its adverse effects on human health and ecosystems increasingly being scrutinized. This study proposes a novel model to achieve accurate prediction of near-surface ozone concentrations. The model integrates the strengths of Graph Convolutional Networks (GCN) and Gated Recurrent Units (GRU) to capture spatial dependencies among monitoring stations and temporal dynamics of ozone concentrations, respectively. Additionally, an attention-based feature fusion module is developed to dynamically balance the contributions of spatial and temporal features, further enhancing prediction performance. Furthermore, a multi-stage training strategy is introduced to adapt the model better to autoregressive time series prediction tasks. Using data from 30 air quality monitoring stations in Beijing as test subjects, the model demonstrates excellent predictive performance, even for longer-term predictions (+6 hours). Compared to commonly used methods, the proposed model significantly reduces the average prediction error, exhibiting greater stability and robustness, especially in multi-step prediction tasks where it effectively mitigates error accumulation. This study provides a reliable and robust spatiotemporal air quality modeling framework for predicting near-surface ozone concentrations, offering a significant reference for improving air quality management and understanding the dynamics of ozone pollution.

The on-demand food delivery (OFD) service has gained rapid development in the past decades but meanwhile encounters challenges for further improving operation quality. The order dispatching problem is one of the most concerning issues for the OFD platforms, which refer to dynamically dispatching a large number of orders to riders reasonably in very limited decision time. To solve such a challenging combinatorial optimization problem, an effective matching algorithm is proposed by fusing the reinforcement learning technique and the optimization method. First, to deal with the large-scale complexity, a decoupling method is designed by reducing the matching space between new orders and riders. Second, to overcome the high dynamism and satisfy the stringent requirements on decision time, a reinforcement learning based dispatching heuristic is presented. To be specific, a sequence-to-sequence neural network is constructed based on the problem characteristic to generate an order priority sequence. Besides, a training approach is specially designed to improve learning performance. Furthermore, a greedy heuristic is employed to effectively dispatch new orders according to the order priority sequence. On real-world datasets, numerical experiments are conducted to validate the effectiveness of the proposed algorithm. Statistical results show that the proposed algorithm can effectively solve the problem by improving delivery efficiency and maintaining customer satisfaction.