Search

While automated vehicle (AV) research has grown steadily in recent years, the impact of automated lane changing behavior on transportation systems remains a largely understudied topic. The present work aims to explore the effects of automated lane changing behavior on urban network efficiency as the penetration rate of AVs increases. To the best of the authors knowledge, this represents the first attempt to do so by isolating the effects of the lane changing behavior; this was obtained by considering AVs with automated lateral control, yet retaining the same longitudinal control characteristics of conventional vehicles (CV). An urban road network located in Hannover, Germany, was modeled with the microsimulation software SUMO, and several scenarios were analyzed, starting from a baseline with only CVs and then progressively increasing the AV penetration rate with 10% increments. Results highlight a modest, but statistically significant, decrease in system performance, with travel times increasing, and average speed and network capacity decreasing, as penetration rates increase. This was likely caused by a more prudent behavior of AVs, which accepted larger gaps than CVs when performing lane changing maneuvers., Comment: Accepted article version of paper presented at the 2023 8th International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS)

Shared spaces aim to reduce the dominance of motor vehicles by promoting pedestrian and cyclist activity and minimizing segregation between road users. Despite the intended scope to improve the safety of vulnerable road users, only few works in the literature focused on before after safety evaluations, mainly analyzing changes in users trajectories and speeds, traffic volumes, and conflict counts, which, while useful, cannot univocally quantify road safety. Here, we propose a more advanced methodology, based on surrogate measures of safety and Extreme Value Theory, to assess road safety before and after the implementation of a shared space. The aim is to produce a crash risk estimation in different scenarios, obtaining a quantitative and comprehensive indicator, useful to practitioners for evaluating the safety of urban design solutions. A real world case study illustrates the proposed procedure. Video data were collected on two separate days, before and after a shared space implementation, and were semiautomatically processed to extract road users trajectories. Analysis of traffic volumes, trajectories, speeds and yield ratios allowed to understand the spatial behavior of road users in the two scenarios. Traffic conflicts, identified with an innovative surrogate measure of safety called time to avoided collision point, TTAC, were then used to estimate a Lomax distribution, and therefore to model the probabilistic relationship between conflicts and crashes, eventually retrieving a crash risk estimate. Results show that the analyzed shared space was able to significantly reduce the risk of crashes, and these findings are consistent with the observed changes in users speed and spatial behavior. The analyzed case study and its limitations were useful in highlighting the methodology main features and suggesting practical prescriptions for practitioners., Comment: This is the Accepted Manuscript version. The version of record of this manuscript has been published and is available on Case Studies on Transport Policy, Volume 13, 101021

The use of neural networks to predict airport passenger activity choices inside the terminal is presented in this paper. Three network architectures are proposed: Feedforward Neural Networks (FNN), Long Short-Term Memory (LSTM) networks, and a combination of the two. Inputs to these models are both static (passenger and trip characteristics) and dynamic (real-time passenger tracking). A real-world case study exemplifies the application of these models, using anonymous WiFi traces collected at Bologna Airport to train the networks. The performance of the models were evaluated according to the misclassification rate of passenger activity choices. In the LSTM approach, two different multi-step forecasting strategies are tested. According to our findings, the direct LSTM approach provides better results than the FNN, especially when the prediction horizon is relatively short (20 minutes or less)., Comment: Post-print of paper presented at the 2019 6th International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS)