A recent survey by the City of Calgary, Canada, found that more than 75% of cyclists commuting to downtown Calgary are male. The intent of this research is to determine whether this is also true for cyclists commuting to a university campus located in the second most popular employment area of the city, what obstacles are preventing women from bicycling, and what measures could increase the number of female commuter cyclists. An online survey was conducted to collect information that allowed the grouping of respondents as potential, occasional, or regular cyclists. Analysis showed that women are more likely than men to be possible or occasional cyclists, while men are more likely than women to be regular cyclists. These findings suggest that if women's cycling needs were addressed, the modal share of bicycle commuting could be increased. Investigation of cycling barriers indicated that women are more concerned than men about safety issues associated with cycling, with being able to carry daily items while cycling, and with the need to fix their hair on arrival. In analysis of desired improvements, women were found to place a higher value on bicycle maps and literature but share similar facility preferences with men. High proportions of both genders indicated a desire for bicycle lanes, more pathways, and more direct bicycle routes. Analysis of falls and collisions suggested that men and women experience a similar number of falls per unit of exposure, while men experience more collisions per unit of exposure than women do.

As the number of bicyclists in urban areas continues to increase, the need to realistically model the movement and interactions of bicyclists is rapidly gaining importance in the accurate modeling of mixed urban traffic. In response to this need, several approaches to modeling bicyclists’ movements and interactions have been developed. This study summarized selected modeling approaches that depict the state of the art in bicycle modeling. The overall modeling of bicycles was divided into modeling of uninfluenced operational and tactical behavior and influenced operational and tactical behavior. The ability to model bicyclist behavior on each of these levels was evaluated on the basis of the results of an extensive literature review and input from an expert workshop that included industry professionals and academics with extensive experience in traffic modeling. The results of the assessment indicate that although the approaches used to model uninfluenced and influenced behavior on the operational level vary in their level of detail and ability to reproduce reality correctly, it is possible to model most bicyclist behavior at this level. There is a need to validate and calibrate these models with empirical data collected from a variety of locations and traffic situations. The state of the art in modeling the tactical behavior of bicyclists is, however, less developed. It is important to model the uninfluenced and influenced tactical behavior of bicyclists accurately because bicycle behavior is less constrained by road markings and traffic regulations.

Bicyclists are extremely flexible road users who employ various tactical behaviours to optimise comfort, directness and time efficiency while crossing a signalised intersection. Tactical choices faced by bicyclists at signalised intersections include whether to use the bicycle lane, roadway or sidewalk, to stop at or violate a red traffic signal, to ride with or against the mandatory direction of travel and the method of executing a left turn. The outcome of these choices has a direct impact on traffic safety and efficiency at intersections. In this paper, revealed choice data from 4710 bicyclists at four intersections in Munich, Germany are used to estimate binomial and multinomial logistic regression models to predict tactical choice outcomes. Optimal predictor sets are selected from the main and two-way interaction effects of 43 independent variables describing the situation, strategic behaviour and prior tactical choices of bicyclists using recursive feature elimination. A simplified model is estimated using the statistically significant variables of the optimal predictor set. The prediction power of the resulting regression model is assessed using k-fold cross validation. The models to predict response to a red signal and the type of left-hand turn exhibit high predictive power while the prediction of infrastructure selection and the direction of travel proves to be difficult.

Models were developed, calibrated, and evaluated to describe the acceleration and deceleration processes of bicyclists in three states: while they accelerate from a stop, decelerate to a stop, and fluctuate around the desired traveling speed. Such models are necessary to simulate the speed profiles of bicyclists reliably in microscopic traffic simulations. To accomplish this aim, a sample of 1,030 processed trajectories from bicyclists at four intersections in Munich, Germany, was used to analyze the dynamic characteristics of bicyclists. The average crossing speed, the fluctuation in crossing speed, and the minimum and the maximum speeds of uninfluenced bicyclists who crossed at a green light were analyzed, and correlations between these variables were investigated. The acceleration and deceleration profiles of bicyclists who stopped at a red light but were uninfluenced by other bicyclists, were used to evaluate four acceleration–deceleration models: the constant model, the linear decreasing model, the two-term sinusoidal models, and a polynomial model. Two adaptations of the models were developed and evaluated: one to derive acceleration and deceleration as a function of speed rather than time, and the other to account for the observed fluctuation in bicyclist traveling speed. The polynomial model was found to be the most flexible and to produce the overall best estimates of the acceleration profiles. The constant model was found to estimate best the deceleration, acceleration, and deceleration while fluctuation occurred around the desired speed.

No abstract