The electric scooter (e-scooter) sharing service has attracted significant attention because of its extensive usage and eco-friendliness. Since e-scooters are mostly accessed by foot, the presence of e-scooters within walking distance has a crucial effect on the service quality. Therefore, to maintain appropriate service quality, relocation strategies are often used to properly distribute e-scooters within service areas. There are extensive literatures on demand forecasting for an efficient relocation. However, the study of the relocation of small-scale spatial units within walking distance level is still inadequate because of the sparsity of demand data. This research aims to establish an effective methodology for predicting the demand for e-scooters in high spatial resolution. A new grid-based spatial setting was created with the usage data. The model in the methodology predicts not only the identified demand but also the unmet demand to increase practicality. A convolutional autoencoder is used to obtain the latent feature that can reduce the problem of representing sparse data. An encoder–recurrent neural network–decoder (ERD) framework with a convolutional autoencoder resulted in a huge improvement in predicting spatiotemporal events. This new ERD framework shows enhanced prediction performance, reducing the mean squared error loss to 0.00036 from 0.00679 compared with the baseline long short-term memory model. This methodological strategy has its significance in that it can solve any prediction issue with spatiotemporal data, even those with sparse data problems.
With the growth of the bike-sharing system, the problem of demand forecasting has become important to the bike-sharing system. This study aims to develop a novel prediction model that enhances the accuracy of the peak hourly demand. A spatiotemporal graph convolutional network (STGCN) is constructed to consider both the spatial and temporal features. One of the model’s essential steps is determining the main component of the adjacency matrix and the node feature matrix. To achieve this, 131 days of data from the bike-sharing system in Seoul are used and experiments conducted on the models with various adjacency matrices and node feature matrices, including public transit usage. The results indicate that the STGCN models reflecting the previous demand pattern to the adjacency matrix show outstanding performance in predicting demand compared with the other models. The results also show that the model that includes bus boarding and alighting records is more accurate than the model that contains subway records, inferring that buses have a greater connection to bike-sharing than the subway. The proposed STGCN with public transit data contributes to the alleviation of unmet demand by enhancing the accuracy in predicting peak demand.
Traffic density, which is a critical measure in traffic operations, should be collected precisely at various locations and times to reflect site-specific spatiotemporal characteristics. For detailed analysis, heavy vehicles have to be separated from ordinary vehicles, since heavy vehicles have a significant effect on traffic flow as well as traffic safety. With unmanned aerial vehicles (UAVs), it is easy to acquire video for vehicle detection by collecting images from above the traffic without any disturbances. Despite previous studies on vehicle detection, there is still a lack of research on real-world applications in estimating traffic density. This study investigates the effects of several influential factors: the size of objects, the number of samples, and a combination of datasets, on detecting multi-class vehicles using deep learning models in various UAV images. Three detection models are compared: faster region-based convolutional neural networks (faster R-CNN), region-based fully convolutional network (R-FCN), and single-shot detector (SSD), to suggest guidelines for model selection. The results provided several findings: (i) vehicle detection from UAV images showed sufficient performance with a small number of samples and small objects; (ii) deep learning-based multi-class vehicle detectors can have advantages compared with single-class detectors; (iii) among all the models, SSD showed the best performance because of its algorithmic structure; (iv) simply combining datasets in different environments cannot guarantee performance improvement. Based on these findings, practical guidelines are offered for estimating multi-class traffic density using UAV.
The dockless e-scooter sharing service is rapidly spreading, replacing existing transportation, and improving last-mile accessibility. User segmentation with travel regularity and segment-level behavior analysis, which are already conducted in public transit, also benefits e-scooter sharing service to enhance service quality and increase usage. In this work, we group e-scooter users according to their travel regularity and identify each group’s usage characteristics. Through the dockless e-scooter usage data, as operated in six cities in South Korea, travel regularity measured by users’ repetitive departure time and destination is discovered and spatiotemporal usage patterns are identified. We divide e-scooter users into three groups by type of travel regularity: irregular user, spatially regular user, and regular user. Regular users more frequently use e-scooters, travel shorter distances, and walk longer distances to find an e-scooter than other groups. It is also revealed that the use in morning peak hours only occurs in the regular user group. By decomposing the temporal patterns of spatially regular and regular users, we discover that spatially regular users are composed of daytime, evening peak, and nighttime users. In contrast, regular users are composed of morning peak, evening peak, and lockdown (restriction in response to COVID-19 pandemic) peak users. This research suggests user segmentation based on travel regularity in e-scooter sharing services, enabling multiple strategies to be drawn to retain users with high regularity and convert users with low regularity to regular users.
The estimation of traffic variables and provision of traffic information are the most important components of intelligent transportation systems. Advances in technology have led to the collection of various traffic sensor data, and nonlinear dependencies between traffic variables have enabled the development of models based on deep learning approaches. However, there is a missing data segment where data collection is not possible because of the non-installation of the sensor, malfunction of the sensor, or error in communication. In this study, a deep multimodal model is proposed for traffic speed estimation of the missing data segment. We implement the proposed model using two heterogeneous traffic sensors, that is, a vehicle detection system and dedicated short-range communication. The structure of the proposed model consists of three multilayer perceptron models, two of which receive each modality as input data and one fusion model that receives the concatenated outputs from each modality model as input data. To evaluate the estimation performance of the deep multimodal model, we use three performance measures to compare the multimodal model with the arithmetic average model and a single-modality model. The results show that the single-modality model and the proposed deep multimodal model outperform the arithmetic average model. In particular, the deep multimodal model shows the highest accuracies of 90.5% and 92.1% on weekends and peak hours, respectively, without reflecting the true value. The proposed deep multimodal model has three contributions, that is, high accuracy using two different sensors, robustness in various periods, and real-time application with fast computational time.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.