We are happy to present this special issue of Transportmetrica A on “Vehicular and pedestrian flow: from data to models”. It bundles eight papers, which describe the ever progressing state-of-the-a...
In this study we analyse the impact of congestion in dynamic origin-destination (OD) estimation. This problem is typically expressed using a bi-level formulation. When solving this problem the relationship between OD flows and link flows is linearised. In this article the effect of using two types of linear relationship on the estimation process is analysed. It is shown that one type of linearisation implicitly assumes separability of the link flows, which can lead to biased results when dealing with congested networks. Advantages and disadvantages of adopting non-separable relationships are discussed. Another important source of error attributable to congestion dynamics is the presence of local minima in the objective function. It is illustrated that these local minima are the result of an incorrect interpretation of the information from the detectors. The theoretical findings are cast into a new methodology, which is successfully tested in a proof of concept.
The LWR model is of interest since it is simple and can successfully reproduce some essential features of traffic flow, such as the formation and propagation of traffic disturbances. In this article, we investigate the LWR model from an uncertainty perspective. We attempt to analyse how reliable the LWR model prediction will be if the fundamental diagram (FD) in use is not accurately specified. To fulfil this end, we postulate a flux function (equivalently a FD) driven by a random free flow speed, which accommodates the uncertain feature observed in the speed–density data. We provide essential mathematical properties and solution schemes of the LWR model with the probabilistic FD. In case studies, the approach to evaluate the uncertainty of traffic disturbance propagation with this model is presented. We find that if FD in a LWR model cannot be perfectly specified, the uncertainty associated with the location of a traffic disturbance would increase over time. In contrast, the magnitude of the traffic disturbance can still be accurately predicted.
In this article, a hybrid predictive control (HPC) strategy is formulated for the real-time optimisation of a public transport system operation run using buses. For this problem, the hybrid predictive controller corresponds to the bus dispatcher, who dynamically provides the optimal control actions to the bus system to minimise users' total travel time (on-vehicle ride time plus waiting time at stops). The HPC framework includes a dynamic objective function and a predictive model of the bus system, written in discrete time, where events are triggered when a bus arrives at a bus stop. Upon these events, the HPC controller makes decisions based on two well-known real-time transit control actions, holding and expressing. Additionally, the uncertain passenger demand is included in the model as a disturbance and then predicted based on both offline and online information of passenger behaviour. The resulting optimisation problem of the HPC strategy at every event is Np-hard and needs an efficient algorithm to solve it in terms of computation time and accuracy. We chose an ad hoc implementation of a Genetic Algorithm that permits the proper management of the trade-off between these two aspects. For real-time implementation, the design of this HPC strategy considers newly available transport technology such as the availability of automatic passenger counters (APCs) and automatic vehicle location (AVL) devices. Illustrative simulations at 2, 5 and 10 steps ahead are conducted, and promising results showing the advantages of the real-time control schemes are reported and discussed.
Non-recurrent traffic congestion has been estimated using the capacity and the number of closed lanes in work zones and the upstream traffic demand of work zones. However, the number of closed lanes may be meaningless due to operational strategies, such as using the shoulder area and composing an additional lane by temporarily reducing the existing lane width to mitigate traffic congestion. The objective of this study is to develop a method to quantify non-recurrent traffic congestion caused by freeway work zones based on traffic flow data and spatio-temporal work zone information. In addition, to demonstrate the efficacy of the developed method, a case study is performed using 1-year historical traffic data and work zone data on major freeways in Korea.
This article contributes to the growing research and policy interest on the challenges of achieving socially sustainable transportation. It analyses the determinants of transport mode use for journey to work among population groups considered as vulnerable to mobility and accessibility limitations. Using the 2001 Census of Canada, multilevel multinomial logistic regression models were estimated to assess the personal, social and economic factors that affect travel mode use of low-income persons in their journey to work in urban areas in Ontario and Quebec. The findings show important differences in the factors associated with car driving and public transit, between genders, and according to income level, educational achievement, household structure and immigration status. Furthermore, it is found that significant factors affecting travel mode use among low-income people in various urban areas are differentiated by province. The results point towards a geographic-based and balanced promotion of public and private mobility programmes and policies to address transport needs of low-income workers.
In this article, we propose an extended multiclass gas-kinetic theory applicable to mixed traffic flow of manual and adaptive cruise control (ACC) vehicles. In the model, the acceleration/deceleration of ACC vehicles is specified explicitly using microscopic models. The macroscopic multiclass traffic equations are obtained from the proposed gas-kinetic model using the well-known method of moments. The influencing conditions to traffic flow stability with respect to a small perturbation are derived by the linear stability method. The analytical results show that ACC vehicles contribute to the improved stabilisation of traffic flow. The numerical simulations of our developed multiclass macroscopic model on a circular freeway support our analytical findings and indicate that increasing the penetration of ACC vehicles results in more stable traffic flow. Simulations of merging flows at an on-ramp in an open freeway are carried out to describe the effects of the penetration of ACC vehicles on the bottleneck capacity. It is found that around 30% ACC vehicles in traffic flow leads to significantly increased capacity and reduced travel time. We argue that, together with other microscopic models, our model provides a wider picture of the effects of ACC vehicles on traffic flow characteristics.
This study examines the household interactions in the context of the car allocation choice decision in car-deficient households as part of an activity-scheduling process, focusing on non-work tours. A chi-square automatic interaction detector-based algorithm is applied to derive a decision tree using a large activity diary dataset recently collected in the Netherlands. The results show a satisfactory improvement in the goodness-of-fit of the decision tree model compared to the null model. Gender still plays a role. A descriptive analysis indicates that men, more often than women, get the car for non-work tours for which a car allocation decision needs to be made. Tour-level attributes also influence the household car allocation decision for non-work tours. Overall, men exert more influence on the car allocation decision for non-work tours, as indicated by the number of influential variables that relate to males. The developed models will be incorporated in a refinement of the ALBATROSS model - an existing computational process model of activity-travel choice.
We examine the impact of individual-specific information processing strategies (IPSs) on the inclusion/exclusion of attributes on the parameter estimates and behavioural outputs of models of discrete choice. Current practice assumes that individuals employ a homogenous IPS with regards to how they process attributes of stated choice (SC) experiments. We show how information collected exogenous of the SC experiment on whether respondents either ignored or considered each attribute may be used in the estimation process, and how such information provides outputs that are IPS segment specific. We contend that accounting the inclusion/exclusion of attributes will result in behaviourally richer population parameter estimates.
This special issue provides additional insights into activity-travel behaviour analysis for universal mobility design. Four papers are included, focusing on behavioural issues of elderly who need nursing care, children and parents, low-income persons and car-deficient households. Time use, travel mode usage, car allocation within households and activity scheduling behaviours are analysed in these papers, using large-scale samples. Promising methods about the representation of inter-personal interactions and diverse behavioural variations as well as empirical findings are provided. Methodological challenges and mobility policies in the future are also discussed.
This study addresses complex daily activity-travel routines of households with young children and their proper representation in a computational process model of travel demand using family skeletons expressed as family sequence patterns. Building on qualitative interview research findings, an a priori classification of family types is defined according to the distribution of care and work responsibilities in the household on a typical weekday. Enriched census data are examined to calculate the share of each family type in the region of Flanders in Belgium. Next, individual activity-travel sequence patterns are drawn for children and adults. Finally, these individual sequences are combined to family sequence patterns, yielding a concise representation of skeletal information in activity-travel patterns of household members and their interrelationships. This process is tested and the method offers a promising approach to both household activity-travel analysis and travel demand modelling.
In general, almost everyone talks about transaction costs in regard to vertical separation of rail infrastructure from operation but there is still only little or almost no systematic work on their measurement in railways. This paper applies a top down approach to reveal the size of the transaction sector within train operation and rail infrastructure firms. It also aims to estimate the resulting transaction costs and their share in total operating costs for the first time. The results reveal differences in the relative size of the transaction sector on firm level, supporting in general the argument that vertical separation results in relatively higher transaction costs. However, their level is less decisive than suggested by earlier studies and might be offset by parent company support and the size of the rail system wide transaction sector.
In the study of traffic safety, expected crash frequencies across sites are generally estimated via the negative binomial model, assuming time invariant safety. Since the time invariant safety assumption may be invalid, Hauer (1997) proposed a modified empirical Bayes (EB) method. Despite the modification, no attempts have been made to examine the generalisable form of the marginal distribution resulting from the modified EB framework. Because the hyper-parameters needed to apply the modified EB method are not readily available, an assessment is lacking on how accurately the modified EB method estimates safety in the presence of the time variant safety and regression-to-the-mean (RTM) effects. This study derives the closed form marginal distribution, and reveals that the marginal distribution in the modified EB method is equivalent to the negative multinomial (NM) distribution, which is essentially the same as the likelihood function used in the random effects Poisson model. As a result, this study shows that the gamma posterior distribution from the multivariate Poisson–gamma mixture can be estimated using the NM model or the random effects Poisson model. This study also shows that the estimation errors from the modified EB method are systematically smaller than those from the comparison group method by simultaneously accounting for the RTM and time variant safety effects. Hence, the modified EB method via the NM model is a generalisable method for estimating safety in the presence of the time variant safety and the RTM effects.
This article studies route choices and traffic equilibria when travel times are risky, and travellers are risk averse and regret averse. It is shown how regret theory, being one of the most popular contenders of expected utility theory throughout the social sciences, can be applied to model risky route choices by means of an expected modified utility function. Subsequently this function is used to study numerically how risk aversion and regret aversion jointly determine equilibrium outcomes in a simple binary route choice situation. It is found that increasing levels of regret aversion lead to equilibrium shifts towards routes whose mean travel time is low, routes that are less risky and especially routes whose worst-case travel time is low compared to that of the competing route. Furthermore, risk aversion and regret aversion are found to reinforce each other's impact on equilibrium towards a situation where safer routes are preferred over riskier (but faster) ones.
The capacity within an extended theory of planned behaviour (TPB) to change young drivers' intentions and reduce their commission of driving violations was tested using regression-based statistical simulations. Participants (N = 198) completed questionnaire measures of TPB variables, plus moral norm and anticipated regret, each with respect to 11 different driving violations. One month later, subsequent behavioural performance was measured, again using self-completion questionnaires. Statistical simulations indicated substantial capacity within the extended TPB to reduce driving violations, with maximum changes to all of the cognitive predictors generating large degrees of intention and behaviour change (i.e. d>0.80). However, the degree of intention change that was generated was greater than the degree of behaviour change, and sensitivity analyses demonstrated that behavioural interventions need to successfully change multiple cognitive variables in order to achieve meaningful reductions in driving violations. Implications of the findings for developing behaviour change interventions are discussed.
Junction modelling is described within the framework of macroscopic traffic flow models. Given previous models, the focus is on the most important factor: first-in first-out (FIFO) effects. According to the number of lanes, current models assume either a perfectly FIFO traffic or a perfectly non-FIFO traffic. Although we assume a single-flow highway traffic model, the improvement takes into account lane changes close to merges and diverges. Thus the traffic is neither perfectly FIFO nor perfectly non-FIFO and so outflows at junctions depend considerably on junction topology, traffic distribution in lanes and driver behaviour. As a consequence, this article suggests a simple improvement of the description of merges and diverges which takes into account some of these factors. With respect to the previous models, the best improvement concerns the outflow forecasting on an uncongested exit of a diverge when the other exit is congested.
Three new models for the flow-density relationship are proposed in this work. The resulting flow-density curves are concave in the whole range of feasible values for the parameters. These models have four parameters, three of them being the jam density, the free-flow speed and the kinematic wave speed. The fourth parameter is a shape parameter. The models allow for a great flexibility for fitting of real traffic flow and density data. A remarkable property of these models is the fact that they yield a bilinear fundamental diagram when the shape parameter tends to infinity. The models have been tested with freeway data and urban data. The results demonstrate that the models achieve an excellent goodness of fit and yield realistic estimates of the parameters. The models proposed in this work are a valuable tool not only for fitting flow-density data but also for its use in traffic flow dynamic models.
This article considers the stochastic user equilibrium (SUE) problem with the route choice model based on the C-logit function. The C-logit model has a simple closed-form analytical probability expression and requires relatively lower calibration efforts and represents a more realistic route choice behaviour compared with the multinomial logit model. This article proposes two versions of the C-logit SUE model that captures the route similarity using different attributes in the commonality factors. The two versions differ with respect to the independence assumption between cost and flow. The corresponding stochastic traffic equilibrium models are called the length-based and congestion-based C-logit SUE models, respectively. To formulate the length-based C-logit SUE model, an equivalent mathematical programming formulation is proposed. For the congestion-based C-logit SUE model, we provide two equivalent variational inequality formulations. To solve the proposed formulations, a new self-adaptive gradient projection algorithm is developed. The proposed formulations and new solution algorithm are tested in two well-known networks. Numerical results demonstrate the validity of the formulations and solution algorithm.
