The agile earth observation satellite scheduling problem (AEOSSP) with time-dependent transition times is a complex combinational optimization problem that has emerged from the development of large-scale satellite management techniques. To address this problem, we propose a deep reinforcement learning-based construction model (DRL-CM) that consists of five parts: 1) a Markov decision process (MDP); 2) a feature engineering; 3) a constructive heuristic neural network (CHNN); 4) an RL training method; and 5) an evaluation system. Specifically, the CHNN comprises six modules containing three special components that we propose: a dynamic encoder, a dynamic global layer, and a two-stage attention layer. First, we build the MDP of the AEOSSP and the feature engineering with effective features required for decision-making. Second, we design the CHNN to function as the MDP policy and train it with an RL model. Finally, we propose a comprehensive evaluation system for the validation of our model. The experimental results indicate that the proposed DRL-CM outperforms the state-of-the-art algorithm in terms of both optimization speed and quality. In addition, the feature engineering and network architecture built in our model are verified to be effective in comprehensive experiments.

The capacitated electric vehicle routing problem (CEVRP) extends the traditional vehicle routing problem by simultaneously considering the service order of the customers and the recharging schedules of the vehicles. Due to its NP-hard nature, we decompose the original problem into two sub-problems: a capacitated vehicle routing problem (CVRP) and a fixed route vehicle charging problem (FRVCP). A highly effective threshold acceptance based multi-layer search (TAMLS) algorithm is proposed to quickly obtain high-quality solutions. TAMLS consists of three layers. An iterated thresholding search procedure and a thresholding selection procedure are employed to produce diversified CVRP solutions in the first layer and to screen out high quality ones in the second layer, respectively. In the third layer, a removal heuristic coupling with an enumeration method is adopted to solve FRVRP, which produces optimized charging schedules. Extensive computational results show that TAMLS outperforms the state-of-the-art algorithms in terms of both solution quality and computation time. In particular, it is able to obtain new best results for 11 out of 17 benchmark instances, and reach the best known results on the remaining 6 instances. Additional experimental analyses are performed to better understand the contributions of key algorithmic components.

In the face of a sudden public health emergency caused by a new infectious disease, it is necessary to establish a multi-level emergency medical facility (including primary and superior facilities) to address the surge in medical needs. In this context, traditional hospitals are responsible for patient screening, primary emergency medical facilities are responsible for treating mild cases, and superior emergency medical facilities are responsible for treating severe cases. Against the backdrop of uncertainties such as patient self-referral and the autonomous progression of the disease, we address an important problem of integrated emergency medical facility location and patient dispatching under uncertainty and propose a multi-stage stochastic programming model to formulate the problem. For a deterministic model under a given set of scenarios, a Decomposition-based Dual-level Heuristic (DDH) algorithm is proposed to efficiently solve the problem, where the upper level employs tabu search to optimize the location scheme, and the lower level utilizes a patient allocation heuristic to provide an optimized patient dispatching solution. Numerical experiments are conducted using Wuhan, China, the epicenter of the COVID-19 outbreak, as an example. The results show that the DDH algorithm achieves high quality solutions close to those obtained by state-of-the-art solver CPLEX but with significantly reduced computational overload. The DDH algorithm is also compared with the progressive hedging algorithm and genetic algorithm, showing its superior performance in terms of solution quality and computational efficiency. Through extensive data analysis, valuable conclusions and managerial insights are obtained, providing useful references for emergency response in similar public health emergencies in the future.

Wireless capsule endoscopy is a non-invasive and painless approach for diagnosing gastrointestinal diseases. An automated medical decision support system can significantly enhance clinician efficiency and reduce the incidence of misdiagnosis when analyzing lesions within the 50,000 to 100,000 image frames generated for each individual. However, only a small number of previous studies have focused on fine-grained recognition characteristics, and very few have simultaneously addressed both fine-grained recognition and class imbalance issues in wireless capsule endoscopy. To address these issues, we propose a novel medical decision support network that includes a prototype attention enhancement module and a dynamic curriculum learning approach. The prototype attention enhancement module improves lesion-sensitive compact representation learning by leveraging cosine similarity between the representation and a prototype memory with multi-class centers. The dynamic curriculum learning adopts triplet loss and weighted cross-entropy, facilitated by a progressive factor that controls between fine-grained representation learning and class-sensitive learning. Through extensive comparative experiments on three public datasets, the proposed network demonstrated competitive performance, outperforming previous methods with an F1-score of 96.6% on the 10-class Kvasir-Capsule dataset, an accuracy of 98.3% on the CAD-CAP dataset, and an accuracy of 93.7% on the mixed KID dataset. Two additional gastrointestinal histopathology datasets confirm the generalization of the proposed network. The code will be available at https://github.com/Xingcun-Li/MDSN-WCE Note to Practitioners—The primary aim of this study is to address diagnostic challenges in automated wireless capsule endoscopy (WCE). WCE provides a non-invasive diagnostic method for gastrointestinal diseases, but manual analysis of the resulting large volume of image data is time-consuming and error-prone. To this end, we propose a novel medical decision support network, which is also applicable to similar medical image analysis issues characterized by class imbalance and fine-grained recognition. Our approach integrates a prototype attention enhancement (PAE) module and dynamic curriculum learning (DCL), enabling the model to accurately identify lesions within WCE images and exhibit favorable performance in gastrointestinal histopathology image recognition beyond the WCE domain. Additionally, our approach can be applied to various common deep learning models without incurring significant additional computational costs, thereby aiding endoscopists in significantly reducing the time required for diagnosing WCE video frames. The code will be uploaded to GitHub for direct use in the automated diagnosis of WCE video frames. While our method shows promise, its main focus is on rapidly identifying a large number of frames within WCE videos. However, further work is needed to extract key frames and improve the interpretability of outputs from WCE videos, which is also a future research direction.