Search

In recent years, the influence of global tropical cyclones (TCs) has a tendency to expand towards the polar direction, and the proportion of high-intensity TCs has also increased greatly. The Northwest Pacific is a high incidence area of TCs, the northward shift of landfalling TCs in China and the increase of high-intensity TCs will lead to an increased risk of storm surge in the East China Sea (ECS). However, due to the lack of long-term observations in the ECS, previous studies have focused on the analysis of individual TC, and the response mechanism of storm surge to long-term trends of TCs is not clear. Based on the long-term observations of TCs from 1949 to 2022 and the numerical experiments of storm surge, the trend of landfalling TCs in China and the response characteristics of storm surge in the ECS were studied in this paper. The observations showed that the probability of TCs entering ECS became greater in the latter 37 years compared with the former 37 years. The average intensity of landfalling TCs (maximum wind speed) increased by 36% to 30 m/s, and the average speed of landfalling TCs increased by 100% to 8 m/s. The model results showed that the intensity of the storm surge is closely related to the intensity and translation speed of TCs. Under the condition of the same translation speed of TCs, the storm surge enhanced with the intensity of TCs. For the TCs with high intensity (strong typhoon and super typhoon), the intensity of the storm surge continues to increase as the translation speed of the TCs increases. However, for the TCs with low or medium intensity (severe tropical storms and typhoons), the intensity of the storm surge reaches a peak with the increase of the translation speed of TCs. The intensity of the storm surge is determined by both the wind speed and the duration of strong wind. The TCs with faster translation speed can produce a stronger wind field but shorter duration of strong wind. Therefore, the intensity of the storm surge decreases when the translation speed of TCs exceeds the threshold value.

Objective. The accurate automatic segmentation of tumors from computed tomography (CT) volumes facilitates early diagnosis and treatment of patients. A significant challenge in tumor segmentation is the integration of the spatial correlations among multiple parts of a CT volume and the context relationship across multiple channels. Approach. We proposed a mutually enhanced multi-view information model (MEMI) to propagate and fuse the spatial correlations and the context relationship and then apply it to lung tumor CT segmentation. First, a feature map was obtained from segmentation backbone encoder, which contained many image region nodes. An attention mechanism from the region node perspective was presented to determine the impact of all the other nodes on a specific node and enhance the node attribute embedding. A gated convolution-based strategy was also designed to integrate the enhanced attributes and the original node features. Second, transformer across multiple channels was constructed to integrate the channel context relationship. Finally, since the encoded node attributes from the gated convolution view and those from the channel transformer view were complementary, an interaction attention mechanism was proposed to propagate the mutual information among the multiple views. Main results. The segmentation performance was evaluated on both public lung tumor dataset and private dataset collected from a hospital. The experimental results demonstrated that MEMI was superior to other compared segmentation methods. Ablation studies showed the contributions of node correlation learning, channel context relationship learning, and mutual information interaction across multiple views to the improved segmentation performance. Utilizing MEMI on multiple segmentation backbones also demonstrated MEMI's generalization ability. Significance. Our model improved the lung tumor segmentation performance by learning the correlations among multiple region nodes, integrating the channel context relationship, and mutual information enhancement from multiple views. [ABSTRACT FROM AUTHOR]

This work investigates a new semi-online scheduling problem with lookahead. We focus on job scheduling on two identical parallel machines, where deterministic online algorithms only know the information of k initial jobs (i.e., the initial-lookahead information), while the following jobs still arrive one-by-one in an over-list fashion. We consider makespan minimization as the objective. The study aims at revealing the value of knowing k initial jobs, which are used to improve the competitive performance of those online algorithms without such initial-lookahead information. We provide the following findings: (1) For the scenario where the k initial jobs are all the largest jobs with length Δ , we prove that the classical LIST algorithm is optimal with competitive ratio (k + 3) / (k + 2) ; (2) For the scenario where the total length of these k   (> 1) jobs is at least Δ , we show that any online algorithm has a competitive ratio at least 3/2, implying that the initial-lookahead knowledge is powerless since there exists a 3/2-competitive online algorithm without such information; (3) For the scenario where the total length of these k   (> α) jobs is at least α Δ (α ≥ 2), we propose an online algorithm, named as LPT-LIST, with competitive ratio of (α + 2) / (α + 1) , implying that the initial-lookahead information indeed helps to improve the competitiveness of those online algorithms lacking such information. [ABSTRACT FROM AUTHOR]

Nowadays cross-docking operations play a significant role in the cold-chain logistics. This paper addresses the cold-chain cross-docking truck scheduling problem where two types of products, i.e. refrigerated and frozen ones, demand separate trucks and storage areas with distinct temperature settings during their storage and transportation. We present a mixed-integer linear programming model with the objective of minimising the total operational costs that consist of inbound truck arrival penalties for violating contracted time windows, product delivery tardiness penalties, inventory costs and outbound truck transportation costs. Due to the strong NP-hardness of the considered problem, we solve it in two phases where the inbound truck arrival schedule and the schedule of outbound truck departure together with product processing are produced, respectively. Four heuristic algorithms are proposed to generate complete solutions of the considered two-stage problem, which are the combinations of two solution frameworks for the first stage and two methods for the second stage. Computational experiments are carried out to verify the effectiveness and efficiency of the proposed heuristic algorithms in terms of the solution quality and running time, respectively. [ABSTRACT FROM AUTHOR]

Increased blood demand and limited blood supply make blood shortages a significant problem in the blood supply chain, which may cause immeasurable losses. The objective of this paper is to develop the scheduling scheme of blood products during blood shortages, considering limited blood supply. For this purpose, a bi-objective mixed-integer programming model is proposed, in which one objective minimises the maximum blood shortage of hospitals as well as the other one minimises the latest arrival time. To solve this model, a $ \varepsilon $-constraint-based hybrid algorithm called $ \varepsilon $-GA-VNS is presented, which benefits from exploration of the genetic algorithm (GA) and exploitation of the variable neighbourhood search (VNS) approaches. Then, a series of numerical experiments based on Solomon's benchmark were performed to evaluate the proposed model and algorithm. A performance comparison of $ \varepsilon $-GA-VNS and NSGA-II indicated that $ \varepsilon $-GA-VNS was superior to NSGA-II in both efficiency and effectiveness. Finally, sensitivity analyses of the uncertain and stochastic blood supply and demand impart several managerial insights. [ABSTRACT FROM AUTHOR]

Upwelling is a widespread phenomenon in the ocean and plays key roles in the marine environment, marine fishery and air–sea carbon exchange. In coastal regions, the upwelling is usually modulated by tides and complex topography, but the dynamical mechanism is still unclear and yet to be quantified. In this study, a three-dimensional (3D) regional ocean model is used to investigate tide-induced upwelling and its mechanisms quantitatively in the mouth of a semi-closed bay, the Bohai Strait, which is a tide-dominated coastal region. The results show that the upwelling mainly occurs near the tidal front in the north of the Laotieshan Channel and the southern region of the front, with the most active upwelling existing off promontories and small islands. Numerical sensitivity experiments indicate that the upwelling in the study area is mainly caused by tides, accounting for approximately 86% of the total. The 3D balance of the vertical component of the vorticity based on the model results quantifies the dynamic processes of the upwelling and reveals that tides induce the upwelling through tidal mixing and nonlinear effects. In the tidal front zone, the upwelling is mainly caused by baroclinic processes related to tidal mixing. Off promontories and small islands, we first reveal that the upwelling is driven by both the tidal mixing and nonlinear effect related to centrifugal force rather than just one of the two mechanisms, and the latter plays a dominant role in producing the upwelling. The strong nonlinear effect is attributed to the periodic movement of barotropic tidal currents rather than the mean flow. [ABSTRACT FROM AUTHOR]

Under support of industry 4.0, researchers have shown an increased interest in MapReduce scheduling problems to process big data. However, very few studies investigate MapReduce scheduling problems under parallel batch machine environment, which is also common in practice. Motivated by this, we study a parallel batch machine scheduling problem in which all the jobs are belonging to MapReduce type. The objective of the considered problem is of minimizing the total weighted tardiness. For solving this problem, we first establish a mixed integer linear programming model, and then a rule-based genetic algorithm is developed to solve it. Numerical experiments are also conducted to demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]