Search

This paper discusses the implementation of an asymmetric network equilibrium model with detailed representation of unsignalized and signalized urban intersections. A software has been developed to solve the deterministic user equilibrium (DUE) problem which takes into account real urban intersections in their detailed configurations. During the first phase this software was tested on a “toy” network and then on the real network of Villafranca (a town near Verona Italy). The comparison between the equilibrium flow patterns resulting from the model and some traffic counts on the Villafranca network confirms that the model is good.

This research aims at investigating the direct and indirect influence of network structures on urban transportation performance with a macroscopic perspective. Transport systems are complex – the functional properties of a transportation network can affect mobility patterns which in turn changes the network performance. Understanding the topology of transportation networks is important in order to upgrade transport network design and to improve transportation performance. This paper attempts to determine important network indicators such connectivity, centrality and clustering measures for different network types (road, rail and bike) from 86 urban areas and 32 countries, based on compa- rable, directly observable open-source data such as OpenStreetMap (OSM) and the TomTom congestion database. Relations between indicators are identified through correlation measures. In addition, regression models are calibrated which quantify the relations between infrastructure accessibility (IA) and network indicators and average traffic delay times. The indicator average road connectivity over average road circuity (RCRC), which is proposed in this study, has not been cited before in literature. The main results suggest that the determination of distance-based connectivity of networks is an important proxy to understand road transportation performance. Consequently, two main results were obtained: (1) an increase in average short-distance connectivity of road networks (average closeness centrality and RCRC) eases road congestion, presumably because the network distributes road traffic more homogenously while decreasing low-permeability choke points, (2) an increase of the average short-distance connectivity of networks of alternative modes such as rail or bike (average weighted rail clustering coefficient and average cycle closeness centrality) does alleviate road congestion. In particular, for cities with over 0.4 km per km2 cycleway density, an increase in cycleway closeness centrality decreases road congestion and it does so almost as efficiently as an increase in road infrastructure accessibility. Presumably, well-connected, alternative networks with short and direct routes convince car users to shift to the alternative mode, which decreases road traffic volumes.

Cyclist physical attributes are important for performance aspects such as speed, effort, and energy expenditure, and could also be systematically related to preferences and behaviour. Casual assumptions (stereotypes) about cyclists based on their appearance are common among road users but largely untested. This study examines whether readily observable physical attributes are significantly associated with cycling efficiency, preferences, and habits for a sample of 531 intercepted cyclists in Vancouver, Canada. Due to strong correlations among physical attributes, a typology is developed using cluster analysis based on physical aspects of the bicycle (bicycle type, tire type, tire width, tire pressure, and cargo) and rider (apparel and riding position). Results show that Mountain, Hybrid and Road type cyclists are, in that order, systematically more efficient, more comfortable on major roads, cycle more consistently year-round, cycle faster, and engage in more vigorous physical activity. Still, the hypothesis of significant relationships between appearance and behaviour is only weakly supported: behaviour differences among cyclist types are modest and wide ranges of preferences and behaviours within the physical clusters could be viewed as a refutation of common cyclist stereotypes. For application of the physical typology, readily observable attributes such as tire type can be used as indicators to infer resistance parameters and more generally characterize a sample of cyclists. This study establishes associations, not causality, and future work should examine a potential positive feedback effect between equipment efficiency and cycling frequency.

The increase of cyclist presence in urban areas and of the number of cyclist accidents on roads lead researchers to explore the in-traffic visual behavior and hazard perception of cyclists. In this study the actual cyclist gaze behavior while cycling on bicycle tracks—exclusive or shared with pedestrians is analyzed. The intent is to allow a better comprehension of those elements representing interferences, which can influence user’s trip. Field tests were performed in the urban center of Bologna, Italy. 16 participants were asked to wear mobile eye tracking glasses and cycle along a defined route. From gaze data recorded by the mobile eye detector, we analyzed which visual information are detected. By applying fixations detection algorithm and then a frame-by-frame analysis we calculated the proportion of fixations—number and duration-across different areas of interest. Proportion of fixations and fixation time are assumed as a proxy of visual workload. Thus, the relative frequency of fixation has been used to rank those elements that draw cyclist attention. Three are the main outcomes: first, an equilibrium of attention location between the central (trajectory) and lateral parts of the visual scene can be assumed as the optimal cycling visual condition. This condition results compromised when the presence of pedestrians is high. Second, discontinuities of the path (like intersections and crosswalks) and the presence of pedestrians are the elements requiring more attention. Third, the absence of physical and visual separation between cyclists and pedestrians seems to lead to a lack of attention to these risk elements. These outcomes about cyclists’ visual behavior allowed to recommend design measures to increase comfort and safety on shared-with pedestrian-cycling paths. Thus, suggestions are addressed in the conclusions.

Introduction The accurate analysis and comparison of transport indicators from a large variety of urban areas can help to evaluate the performance of different adopted transport policies. This paper attempts to determine important transport and socio-economic indicators from 151 urban areas and 51 countries, based on comparable, directly observable open-source data such as OpenstreetMap (OSM) and the TomTom database. Analysis This is the first, systematic indicator-analysis using recent, open source data from different urban areas around the world. The indicator road kilometers per person, sometimes cited as infrastructure accessibility is calculated by processing OSM data. Information on congestion levels have been taken from the TomTom database and socio-economic data from various, publicly accessible databases. Relations between indicators are identified through correlations and regression models are calibrated, quantifying the relation between transport infrastructure and performance indicators. Three sub-categories of cities with different population sizes (small cities, large cities and metropolises) are defined and studied individually. In addition, a qualitative analysis is performed, putting five different indicators into relation. Results & Conclusions The main results reconfirm previous findings but with a larger sample size and more comparable data. Good correlation values between infrastructure accessibility, socio-economic indicators, and congestion levels are demonstrated. It is shown that cities with higher GDP have generally built more infrastructure which in turn reduces their congestion levels. In particular, for cities with low population density (above approximately 1500 inh. Per sq.km), more roads per inhabitant lead to lower congestion levels; cities with high population density have in general lower congestion levels if the rail infrastructure per person ratio is high. Furthermore, these cities increasing railways per person is more effective in reducing congestions than increasing road length per person.

Knowing daily traffic for the current year is recognized as being essential in many fields of transport analysis and practice, and short-term forecasting models offer a set of tools to meet these needs. This paper examines and compares the accuracy of three representative parametric and non-parametric prediction models, selected by the analysis of the numerous methods proposed in the literature for their good combination of forecast accuracy and ease of calibration, using real-life data on Italian motorway stretches. Non-parametric K-NN regression model, Gaussian maximum likelihood model and double seasonality Holt–Winters exponential smoothing model confirm their goodness to predict the weekly and monthly fluctuations of average daily traffic with varying degrees of performance, while maintaining an easy use in professional practice, i.e. requiring ordinary professional skills and conventional analysis tools. Since combining several prediction models can give, on average, more accuracy than that of the individual models, the paper compares two weighting methods of easy implementation and susceptible to a direct use, namely the widely used information entropy method and the less widespread Shapley value method. Despite being less common than the information entropy method, the Shapley value method proves to be more capable in better combining single forecasts and produces improvements in the predictions for test data. With these remarks, the paper might be of interest to traffic technicians or analysts, in various and not uncommon tasks they might find in their work. Keywords: Short-term traffic forecasting, Non-parametric regression, Gaussian maximum likelihood, Double seasonal Holt–Winters exponential smoothing, Entropy weighting method, Shapley value weighting method

Traffic flow, its characteristics, and road quality service are usually estimated by using aggregated and averaged data from multiple lanes; however, in real experiences, there is significant variability in traffic-flow parameters across the different freeway lanes. Among the numerous factors that affect traffic distribution are vehicular composition, driver behavior, regulations about passing and overtaking, lane discipline, and designated lanes. Because these factors are very specific, the models and values for traffic distribution in a multilane freeway segment are highly sensitive to their national context. This study starts from an analysis of the capacity of a three-lane test segment within the Italian freeway system, and then examines the manner in which the flow and density vary across the freeway lanes. The analysis is restricted to up-to-capacity conditions and nonlinear regression models of the distribution relationships are calibrated for each lane, for both flow and density ratios, as a function of the total flow. The nonhomogeneous use of freeway lanes in Italy can be qualified and quantified through the trends found in these ratios, and then compared with similar internationally focused studies. This provides a useful interpretation of driver behavior, and is also a reference for calibrating traffic models, supporting design and operational processes and the implementation of intelligent transportation systems (ITSs) that are consistent with the national context.

The increasing sensitivity of policy-makers towards more sustainable and healthy transport is leading to increased interest in cycling, especially in urban areas. However, at the same time, recent studies in Europe, US and other countries have stressed the fact that cyclist fatalities are still alarmingly frequent, and lead researchers to want improved knowledge about bicycle traffic flow theory and modeling. The challenge is to make available robust analysis methods and models for building effective and safe infrastructures, for increased cycling mobility combined with positive effects on transport and social systems. This work presents the application of a procedure for fitting bicycle time headways and bicycle speed distributions from traffic data collected along bike tracks. The general frame of the procedure, together with functional components and their mutual interactions, are reported here. The effects of flow rate in both directions (analyzed and opposite) on time headway and vehicle speed distributions were examined. The possibility of associating the probability density functions of bicycle time headways and speeds in various cycling traffic conditions is a significant and interesting advance with respect to previous works. The procedure was applied to cross-sections belonging to the cycling network of the city of Bologna (Italy). The analysis compared a set of headway and speed distribution models, highlighting their goodness-of-fit with reference to empirical distributions.