Including passengers’ response to crowding in the Dutch national train passenger assignment model

Transit passengers’ response to crowded conditions has been studied empirically, yet is limitedly included in transport models currently used in the design of policy and infrastructure investments. This has consequences for the practical applicability of these models in studies on, for instance, timetabling, train capacity management strategies, project appraisal, and passenger satisfaction. Here we propose four methods to include the effect of crowding, based on existing studies on passengers’ perception and response as well as often-used crowding indicators. These four alternative methods are implemented in the train passenger assignment procedure of the Dutch national transport model, and evaluated with respect to their impacts on the model results for the Dutch railway network. The four methods relate to four different ways in which an additive trip penalty and/or time-multiplier can be incorporated in the train utility function for different travel purposes, to capture the disutility of crowding as measured by the load factor. The analyses of the test case favor the hybrid method using both a boarding penalty (capturing seat availability upon boarding) and a time-multiplier (capturing physical comfort and safety throughout the trip). This method produces consistent results, while the additional computational effort that it imposes is acceptable. Further empirical underpinning is needed to conclusively show which of these methods best captures passengers’ response behavior quantitatively (for different travel purposes and conditions).     

A simulation evaluation of the maximum approximate composite marginal likelihood (MACML) estimator for mixed multinomial probit models

This paper evaluates the ability of the maximum approximate composite marginal likelihood (MACML) estimation approach to recover parameters from finite samples in mixed cross-sectional and panel multinomial probit models. Comparisons with the maximum simulated likelihood (MSL) estimation approach are also undertaken. The results indicate that the MACML approach recovers parameters much more accurately than the MSL approach in all model structures and covariance specifications. The MACML inference approach also estimates the parameters efficiently, with the asymptotic standard errors being, in general, only a small proportion of the true values. As importantly, the MACML inference approach takes only a very small fraction of the time needed for MSL estimation. In particular, the results suggest that, for the case of five random coefficients, the MACML approach is about 50 times faster than the MSL for the cross-sectional random coefficients case, about 15 times faster than the MSL for the panel inter-individual random coefficients case, and about 350 times or more faster than the MSL for the panel intra- and inter-individual random coefficients case. As the number of alternatives in the unordered-response model increases, one can expect even higher computational efficiency factors for the MACML over the MSL approach. Further, as should be evident in the panel intra- and inter-individual random coefficients case, the MSL is all but practically infeasible when the mixing structure leads to an explosion in the dimensionality of integration in the likelihood function, but these situations are handled with ease in the MACML approach. It is hoped that the MACML procedure will spawn empirical research into rich model specifications within the context of unordered multinomial choice modeling, including autoregressive random coefficients, dynamics in coefficients, space-time effects, and spatial/social interactions.     

Estimation of origin-destination matrices from link counts and sporadic routing data

Estimation of origin-destination (OD) matrices from link count data is a challenging problem because of the highly indeterminate relationship between the observations and the latent route flows. Conversely, estimation is straightforward if we observe the path taken by each vehicle. We consider an intermediate problem of increasing practical importance, in which link count data is supplemented by routing information for a fraction of vehicles on the network. We develop a statistical model for these combined data sources and derive some tractable normal approximations thereof. We examine likelihood-based inference for these normal models under the assumption that the probability of vehicle tracking is known. We show that the likelihood theory can be non-standard because of boundary effects, and provide conditions under which such irregular behaviour will be observed in practice. For regular cases we outline connections with existing generalised least squares methods. We then consider estimation of OD matrices under estimated and/or misspecified models for the probability of vehicle tracking. Theoretical developments are complemented by simulation experiments and an illustrative example using a section of road network from the English city of Leicester.     

Simultaneous optimization of transit line configuration and passenger line assignment

Passenger transportation in most large cities relies on an efficient mass transit system, whose line configuration has direct impacts on the system operating cost, passenger travel time and line transfers. Unfortunately, the interplay between transit line configuration and passenger line assignment has been largely ignored in the literature. This paper presents a model for simultaneous optimization of transit line configuration and passenger line assignment in a general network. The model is formulated as a linear binary integer program and can be solved by the standard branch and bound method. The model is illustrated with a couple of minimum spanning tree networks and a simplified version of the general Hong Kong mass transit railway network.     

Cellpath: Fusion of cellular and traffic sensor data for route flow estimation via convex optimization

A new convex optimization framework is developed for the route flow estimation problem from the fusion of vehicle count and cellular network data. The issue of highly underdetermined link flow based methods in transportation networks is investigated, then solved using the proposed concept of cellpaths for cellular network data. With this data-driven approach, our proposed approach is versatile: it is compatible with other data sources, and it is model agnostic and thus compatible with user equilibrium, system-optimum, Stackelberg concepts, and other models. Using a dimensionality reduction scheme, we design a projected gradient algorithm suitable for the proposed route flow estimation problem. The algorithm solves a block isotonic regression problem in the projection step in linear time. The accuracy, computational efficiency, and versatility of the proposed approach are validated on the I-210 corridor near Los Angeles, where we achieve 90% route flow accuracy with 1033 traffic sensors and 1000 cellular towers covering a large network of highways and arterials with more than 20,000 links. In contrast to long-term land use planning applications, we demonstrate the first system to our knowledge that can produce route-level flow estimates suitable for short time horizon prediction and control applications in traffic management. Our system is open source and available for validation and extension.     

Mining missing train logs from Smart Card data

This paper shows how to recover the arrival times of trains from the gate times of metro passengers from Smart Card data. Such technique is essential when a log, the set of records indicating the actual arrival and departure time of each bus or train at each station and also a critical component in reliability analysis of a transportation system, is missing partially or entirely. The procedure reconstructs each train as a sequence of the earliest exit times, called S-epochs, among its alighting passengers at each stations. The procedure first constructs a set of passengers, also known as reference passengers, whose routing choices are easily identifiable. The procedure then computes, from the exit times of the reference passengers, a set of tentative S-epochs based on a detection measure whose validity relies on an extreme-value characteristic of the platform-to-gate movement of alighting passengers. The tentative S-epochs are then finalized to be a true one, or rejected, based on their consistencies with bounds and/or interpolation from prescribed S-epochs of adjacent trains and stations. Tested on 12 daily sets of trains, with varying degrees of missing logs, from three entire metro lines, the method restored the arrival times of 95% of trains within the error of 24 s even when 100% of logs was missing. The mining procedure can also be applied to trains operating under special strategies such as short-turning and skip-stop. The recovered log seems precise enough for the current reliability analysis performed by the city of Seoul.     

A new estimation approach for the multiple discrete–continuous probit (MDCP) choice model

This paper develops a blueprint (complete with matrix notation) to apply Bhat’s (2011) Maximum Approximate Composite Marginal Likelihood (MACML) inference approach for the estimation of cross-sectional as well as panel multiple discrete–continuous probit (MDCP) models. A simulation exercise is undertaken to evaluate the ability of the proposed approach to recover parameters from a cross-sectional MDCP model. The results show that the MACML approach does very well in recovering parameters, as well as appears to accurately capture the curvature of the Hessian of the log-likelihood function. The paper also demonstrates the application of the proposed approach through a study of individuals’ recreational (i.e., long distance leisure) choice among alternative destination locations and the number of trips to each recreational destination location, using data drawn from the 2004 to 2005 Michigan statewide household travel survey.     

Managing evacuation networks

This paper proposes a non-anticipative, adaptive, decentralized strategy for managing evacuation networks. The strategy is non-anticipative because it does not rely on demand forecasts, adaptive because it uses real-time traffic information, and decentralized because all the information is available locally. It can be used with a failed communication network.The strategy pertains to networks in which no links backtrack in the direction of increased risk. For these types of networks, no other strategy exists that can evacuate more people in any given time, or finish the evacuation in less time. The strategy is also shown to be socially fair, in the sense that the time needed to evacuate all the people exceeding any risk level is, both, the least possible, and the same as if less-at-risk individuals did not participate in the evacuation. The strategy can be proven optimal even when backflows happen due to driver gaming.