-
Before-after safety analysis of a shared space implementation
Authors:
Federico Orsini,
Mariana Batista,
Bernhard Friedrich,
Massimiliano Gastaldi,
Riccardo Rossi
Abstract:
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,…
▽ More
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.
△ Less
Submitted 3 July, 2023;
originally announced July 2023.
-
Neural networks trained with WiFi traces to predict airport passenger behavior
Authors:
Federico Orsini,
Massimiliano Gastaldi,
Luca Mantecchini,
Riccardo Rossi
Abstract:
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…
▽ More
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).
△ Less
Submitted 30 October, 2019;
originally announced October 2019.
-
On the Runtime-Efficacy Trade-off of Anomaly Detection Techniques for Real-Time Streaming Data
Authors:
Dhruv Choudhary,
Arun Kejariwal,
Francois Orsini
Abstract:
Ever growing volume and velocity of data coupled with decreasing attention span of end users underscore the critical need for real-time analytics. In this regard, anomaly detection plays a key role as an application as well as a means to verify data fidelity. Although the subject of anomaly detection has been researched for over 100 years in a multitude of disciplines such as, but not limited to,…
▽ More
Ever growing volume and velocity of data coupled with decreasing attention span of end users underscore the critical need for real-time analytics. In this regard, anomaly detection plays a key role as an application as well as a means to verify data fidelity. Although the subject of anomaly detection has been researched for over 100 years in a multitude of disciplines such as, but not limited to, astronomy, statistics, manufacturing, econometrics, marketing, most of the existing techniques cannot be used as is on real-time data streams. Further, the lack of characterization of performance -- both with respect to real-timeliness and accuracy -- on production data sets makes model selection very challenging. To this end, we present an in-depth analysis, geared towards real-time streaming data, of anomaly detection techniques. Given the requirements with respect to real-timeliness and accuracy, the analysis presented in this paper should serve as a guide for selection of the "best" anomaly detection technique. To the best of our knowledge, this is the first characterization of anomaly detection techniques proposed in very diverse set of fields, using production data sets corresponding to a wide set of application domains.
△ Less
Submitted 12 October, 2017;
originally announced October 2017.
-
Shift Aggregate Extract Networks
Authors:
Francesco Orsini,
Daniele Baracchi,
Paolo Frasconi
Abstract:
We introduce an architecture based on deep hierarchical decompositions to learn effective representations of large graphs. Our framework extends classic R-decompositions used in kernel methods, enabling nested part-of-part relations. Unlike recursive neural networks, which unroll a template on input graphs directly, we unroll a neural network template over the decomposition hierarchy, allowing us…
▽ More
We introduce an architecture based on deep hierarchical decompositions to learn effective representations of large graphs. Our framework extends classic R-decompositions used in kernel methods, enabling nested part-of-part relations. Unlike recursive neural networks, which unroll a template on input graphs directly, we unroll a neural network template over the decomposition hierarchy, allowing us to deal with the high degree variability that typically characterize social network graphs. Deep hierarchical decompositions are also amenable to domain compression, a technique that reduces both space and time complexity by exploiting symmetries. We show empirically that our approach is able to outperform current state-of-the-art graph classification methods on large social network datasets, while at the same time being competitive on small chemobiological benchmark datasets.
△ Less
Submitted 18 March, 2024; v1 submitted 16 March, 2017;
originally announced March 2017.
