Shuguang Li

With the rapid rise of e-commerce, the logistics and distribution industry is experiencing unprecedented growth. In particular, intra-city distribution is the crucial “last mile" of logistics and plays a decisive role in determining overall customer satisfaction. This study improves an inclusive vehicle routing optimization framework for intra-city distribution under dynamic demand. The initiative of a novel memetic algorithm that efficiently solves the NP-hard dynamic vehicle routing problem while guaranteeing high service quality and cost reduction. However, modern intercity distribution systems often struggle with low information, unpredictable demand patterns, and high operational costs due to scattered customer locations and dynamic order information. Addressing these challenges, this study suggests a comprehensive and intelligent vehicle routing optimization framework tailored for intracity distribution under dynamic demand conditions. The proposed system begins with a grey prediction model for short-term demand forecasting across many distribution regions, permitting differentiated vehicle loading methods to optimize transportation costs and improve operational effectiveness. Building upon this, a dynamic vehicle routing optimization model is formulated to reduce costs while assuring high levels of customer satisfaction within strict delivery time windows. To competently manage fluctuating demand, a dynamic information processing approach is introduced; prioritizing customer needs based on their urgency and importance, thereby guaranteeing the timely delivery of critical orders with minimal computational overhead. Moreover, a novel memetic algorithm is considered to solve the complex NP-hard dynamic vehicle routing problem. This algorithm integrates an adaptive elite genetic algorithm for global search with improved crossover and mutation operators, improved by local search methods such as 2-opt and swap methods to refine solutions. Numerical experiments validate the feasibility and performance of the proposed method, indicating significant improvements over conventional fully loaded vehicle schemes and regular route update methods. The results highlight the practical value of the system in attractive intra-city logistics efficiency, reducing costs, and inspiring customer service standards.

Medical images, particularly chest radiographs and computed tomography (CT), are essential noninvasive diagnostic tools for detecting chest diseases, such as long-term obstructive pulmonary disease, and other respiratory diseases. However, the manual process of detecting diseases and writing their radiology reports is labor-intensive and time-consuming for radiologists. As a result, there has been a growing interest in the development of automated systems for generating radiological chest X-ray (CXR) reports. Although previous research has concentrated on improving the quantitative performance of these automated reports, the quality of the reports has often been neglected. Radiology reports are crucial for communicating with physicians and patients. Therefore, developing automated systems to generate these reports can significantly reduce radiologists’ workload and enhance efficiency in clinical practice by automating the diagnosis of CXR through artificial intelligence (AI) and generating CXR reports. This paper aims to ease the burden especially on radiologists. In this paper, we propose a novel approach using the long short-term memory (LSTM) based model and gated recurrent unit (GRU) model as a decoder in combination with five different convolutional neural network (CNN) models, in which the DenseNet169 model shows the most promising results. Our LSTM-based decoder with the DenseNet169 model achieves the highest results in terms of the BLEU score B1 at 0.5856, B2 at 0.4982, B3 at 0.3470, B4 at 0.1269, ROUGE at 0.4534, and CIDEr at 0.463. Although our GRU-based model achieves the results in terms of the BLEU score (B1: 0.5619, B2: 0.4682, B3: 0.3370, B4: 0.1169), ROUGE (0.4401), and CIDEr (0.4320), the empirical evaluations demonstrate that our proposed approach, especially when combined with DenseNet169, achieves more accurate disease identification and generates reports of fluent and accurate radiological findings. The approach also compared with existing baseline methods.

The distribution of pedestrians in urban space reflects the status of urban spatial planning to some extent. The reasonable prediction of pedestrian concentration is of great significance to the evaluation of urban vitality, urban comfort, and urban spatial layout planning. In this paper, a method for predicting pedestrian concentration is proposed, which can estimate pedestrian concentration in a whole city without being limited to a specific intersection or city node. According to the characteristics of three kinds of transportation accessibility based on space syntax and commercial vitality index, a dynamic distribution estimation model of pedestrian concentration is proposed. Taking Xi’an city of China as a case study, through multiple linear regression (MLR), a support vector regression (SVR) algorithm, and random forest (RF) algorithm, the pedestrian concentration in five periods of a day was predicted and analyzed, and the spatial and temporal characteristics of crowd distribution are comprehensively described. The results show that the dynamic distribution model of pedestrian concentration constructed by RF is superior to the MLR and SVR, and its average prediction accuracy can reach 93.86%.

We propose a dynamic combined-mode traffic network model considering transfer behaviors. We assume that travelers can be classified into two classes: one class is pure-mode travelers who complete a trip by single transportation mode, and another is combined-mode travelers who cover a journey by car, bus, and so forth. The multimode point queue model is used to model the interaction of cars and buses on the network. We present an integrated variational inequality formulation to capture the complex traveler choice behaviors such as departure time choices, transfer point, and route choices. Finally, a numerical example is given to illustrate the effectiveness of the proposed heuristic algorithm and model.

Traffic data of multiple vehicle types are important for pavement design, traffic operations and traffic control. A new video‐based traffic data collection system for multiple vehicle types is developed. By tracking and classifying every passing vehicle under mixed traffic conditions, the type and speed of every passing vehicle are recognised. Finally, the flows and mean speeds of multiple vehicle types are output. A colour image‐based adaptive background subtraction is proposed to obtain more accurate vehicle objects, and a series of processes like shadow removal and setting road detection region are used to improve the system robustness. In order to improve the accuracy of vehicle counting, the cross‐lane vehicles are detected and repeated counting for one vehicle is avoided. In order to reduce the classification errors, the space ratio of the blob and data fusion are used to reduce the classification errors caused by vehicle occlusions. This system was tested under four different weather conditions. The accuracy of vehicle counting was 97.4% and the error of vehicle classification was 8.3%. The correlation coefficient of speeds detected by this system and radar gun was 0.898 and the mean absolute error of speed detection by this system was only 2.3 km/h.

A multiple-vehicle type reactive dynamic user equilibrium model with physical queues is presented. The single-vehicle type cumulative flow given by Newell is extended into multiple-vehicle types. The flow conditions for multiple-vehicle types on link can be divided into two classes: Class 1 is a congested condition – various types of vehicles cannot overtake; Class 2 is an uncongested condition – each type of vehicles can run according to each type of free-flow velocity. Further, multiple-vehicle type reactive dynamic user equilibrium condition is described. Finally, one iterative algorithm is proposed, and the results show that the model can demonstrate multiple-vehicle type physical queue conditions among the links and multiple-vehicle type reactive dynamic user equilibrium condition.