An investigation on the performances of mode shift models in transit ridership forecasting

This paper aims at investigating the over-prediction of public transit ridership by traditional mode choice models estimated using revealed preference data. Five different types of models are estimated and analysed, namely a traditional Revealed Preference (RP) data-based mode choice model, a hybrid mode choice model with a latent variable, a Stated Preference (SP) data-based mode switching model, a joint RP/SP mode switching model, and a hybrid mode switching model with a latent variable. A comparison of the RP data-based mode choice model with the mode choice models including a latent variable showed that the inclusion of behavioural factors (especially habit formation) significantly improved the models. The SP data-based mode switching models elucidated the reasons why traditional models tend to over-predict transit ridership by revealing the role played by different transit level-of-service attributes and their relative importance to mode switching decisions. The results showed that traditional attributes (e.g. travel cost and time) are of lower importance to mode switching behaviour than behavioural factors (e.g. habit formation towards car driving) and other transit service design attributes (e.g. crowding level, number of transfers, and schedule delays). The findings of this study provide general guidelines for developing a variety of transit ridership forecasting models depending on the availability of data and the experience of the planner.     

Flexible substitution patterns in models of mode and time of day choice: new evidence from the UK and the Netherlands

Modelling the temporal response of travellers to transport policy interventions has rapidly emerged as a major issue in many practical transport planning studies and is recognised to hold particular challenges. The importance of congestion and its variation over the day, together with the emergence of time-dependent road user charging as a policy tool, emphasise the need to understand whether and how travellers will change the timing of their journeys. For practical planning studies, analysts face a major issue of relating temporal changes to other behavioural changes that are likely to result from policy or exogenous changes. In particular, the relative sensitivity of time and mode switching has been difficult to resolve. This paper describes a study undertaken to determine the relative sensitivity of mode and time of day choice to changes in travel times and costs and to investigate whether evidence exists of varying magnitudes of unobservable influences in time of day switching. The study draws on data from three related stated preference studies undertaken over the past decade in the United Kingdom and the Netherlands and uses error components logit models to investigate the patterns of substitution between mode and time of day alternatives. It is concluded that the magnitude of unobserved influences on time switching depends significantly on the magnitudes of the time switches considered. With time periods of the magnitude generally represented in practical modelling, i.e. peak periods of 2–3 hours, time switching is generally more sensitive in these data than mode switching. However, the context of the modelling and the extent to which relevant variables can be measured will strongly influence these results.     

Modelling users’ behaviour in inter-urban carsharing program: A stated preference approach

In this paper, the effects of a inter-urban carsharing program on users’ mode choice behaviour were investigated and modelled through specification, calibration and validation of different modelling approaches founded on the behavioural paradigm of the random utility theory. To this end, switching models conditional on the usually chosen transport mode, unconditional switching models and holding models were investigated and compared. The aim was threefold: (i) to analyse the feasibility of a inter-urban carsharing program; (ii) to investigate the main determinants of the choice behaviour; (iii) to compare different approaches (switching vs. holding; conditional vs. unconditional); (iv) to investigate different modelling solutions within the random utility framework (homoscedastic, heteroscedastic and cross-correlated closed-form solutions). The set of models was calibrated on a stated preferences survey carried out on users commuting within the metropolitan area of Salerno, in particular with regard to the home-to-work trips from/to Salerno (the capital city of the Salerno province) to/from the three main municipalities belonging to the metropolitan area of Salerno. All of the involved municipalities significantly interact each other, the average trip length is about 30 km a day and all are served by public transport. The proposed carsharing program was a one-way service, working alongside public transport, with the possibility of sharing the same car among different users, with free parking slots and free access to the existent restricted traffic areas. Results indicated that the inter-urban carsharing service may be a substitute of the car transport mode, but also it could be a complementary alternative to the transit system in those time periods in which the service is not guaranteed or efficient. Estimation results highlighted that the conditional switching approach is the most effective one, whereas travel monetary cost, access time to carsharing parking slots, gender, age, trip frequency, car availability and the type of trip (home-based) were the most significant attributes. Elasticity results showed that access time to the parking slots predominantly influences choice probability for bus and carpool users; change in carsharing travel costs mainly affects carpool users; change in travel costs of the usually chosen transport mode mainly affects car and carpool users.     

Machine learning applications in activity-travel behaviour research: a review

ABSTRACT

This paper reviews the activity-travel behaviour literature that employs Machine Learning (ML) techniques for empirical analysis and modelling. Machine Learning algorithms, which attempt to build intelligence utilizing the availability of large amounts of data, have emerged as powerful tools in the fields of pattern recognition and big data analysis. These techniques have been applied in activity-travel behaviour studies since the early ’90s when Artificial Neural Networks (ANN) were employed to model mode choice decisions. AMOS, an activity-based modelling system developed in the mid-’90s, has ANN at its core to model and predict individual responses to travel demand management measures. In the dawn of 2000, ALBATROSS, a comprehensive activity-based travel demand modelling system, was proposed by Arentze and Timmermans using Decision Trees. Since then researchers have been exploring ML techniques like Support Vector Machines (SVM), Decision Trees (DT), Neural Networks (NN), Bayes Classifiers, and more recently, Ensemble Learners to model and predict activity-travel behaviour. A large number of publications over the years and an upward trend in the number of published articles over time indicate that Machine Learning is a promising tool for activity-travel behaviour analysis and prediction. This article, first of its kind in the literature, reviews these studies and explores the trends in activity-travel behaviour research that apply ML techniques. The review finds that mode choice decisions have received wide attention in the literature on ML applications. It was observed that most of the studies identify the lack of interpretability as a serious shortcoming in ML techniques. However, very few studies have attempted to improve the interpretability of the models. Further, some studies report the importance of feature engineering in ML-based studies, but very few studies adopt feature engineering before model development. Spatiotemporal transferability of models is another issue that has received minimal attention in the literature. In the end, the paper discusses possible directions for future research in the area of activity-travel behaviour modelling using ML techniques.     

A latent class joint mode and departure time choice model for the Greater Toronto and Hamilton Area

This paper presents a closed-form Latent Class Model (LCM) of joint mode and departure time choices. The proposed LCM offers compound substitution patterns between the two choices. The class-specific choice models are of two opposing nesting structures, each of which provides expected maximum utility feedback to the corresponding class membership model. Such feedback allows switching class membership in response to the changes in choice contexts. The model is used for an empirical investigation of commuting mode and departure time choices in the Greater Toronto and Hamilton Area (GTHA) by using a large sample household travel survey dataset. The empirical model reveals that overall 38% of the commuters in the GTHA are more likely to switch modes than departure times and 62% of them are more likely to do the reverse. The empirical model also reveals that the average Subjective Value of Travel Time Savings (SVTTS) of the commuters in the GTHA can be as low as 3 dollars if a single choice pattern of departure time choices nested within mode choices is considered. It can also be as high as 67 dollars if the opposite nesting structure is assumed. However, the LCM estimates the average SVTTS to be around 27 dollars in the GTHA. An empirical scenario analysis by using the estimated model indicates that a 50% increase in morning peak period car travel time does not sway more than 4% of commuters from the morning peak period.

     

Understanding variability,habit and the effect of long period activity plan in modal choices: a day to day,week to week analysis on panel data

Understanding variability in individual behaviour is crucial for the comprehension of travel patterns and for the development and evaluation of planning policies. But, with only one notable exception, there are no studies on the intrinsic variability in the individual preferences for mode choices in absence of external changes in the transport infrastructures. This requires using continuous panel data. Few papers have studied mode choice with continuous panel data but mainly focused on the panel correlation. In this work we use a six-week travel diary survey to study the intrinsic variability in the individual preferences for mode choices, the effect of long period plans and habitual behaviour in the daily mode choices. Mixed logit models are estimated that account for the above effects as well as for systematic and random heterogeneity over individual preferences and responses. We also account for correlation over several time periods. Our results suggest that individual tastes for time and cost are fairly stable but there is a significant systematic and random heterogeneity around these mean values and in the preferences for the different alternatives. We found that there is a strong inertia effect in mode choice that increases with (or is reinforced by) the number of time the same tour is repeated. The sequence of mode choice made is influenced by the duration of the activity and the weekly structure of the activities     

A practical approach to identifying the market potential for high speed rail: A case study in the Sydney-Canberra corridor

This paper presents the methodology and selective empirical results from a study of the demand for a high speed rail system serving the Sydney-Canberra corridor currently dominated by air travel for business trips and car travel for non-business trips. We outline the steps involved in the study from problem specification, data needs, development of base year trip tables, model specification and estimation to establish switching behaviour in the presence of a new mode and calculation of induced demand for current travellers. A stated choice heteroskedastic extreme value switching model is used to evaluate the choice of fare type for business and non-business travel given the current mode used in the corridor for each sampled traveller conventional train, charter coach, scheduled coach, plane or car. Starting with the current travel profile, patronage can be predicted under alternative fare regimes, taking into account diverted traffic, induced traffic and growth. Treating fare class as endogenous enhances the real choice context facing potential patrons.     

Development and testing of a fully Adaptive Cruise Control system

Adaptive Cruise Control systems have been developed and introduced into the consumer market for over a decade. Among these systems, fully-adaptive ones are required to adapt their behaviour not only to traffic conditions but also to drivers’ preferences and attitudes, as well as to the way such preferences change for the same driver in different driving sessions. This would ideally lead towards a system where an on-board electronic control unit can be asked by the driver to calibrate its own parameters while he/she manually drives for a few minutes (learning mode). After calibration, the control unit switches to the running mode where the learned driving style is applied. The learning mode can be activated by any driver of the car, for any driving session and each time he/she wishes to change the current driving behaviour of the cruise control system.The modelling framework which we propose to implement comprises four layers (sampler, profiler, tutor, performer). The sampler is responsible for human likeness and can be calibrated while in learning mode. Then, while in running mode, it works together with the other modelling layers to implement the logic. This paper presents the overall framework, with particular emphasis on the sampler and the profiler that are explained in full mathematical detail. Specification and calibration of the proposed framework are supported by the observed data, collected by means of an instrumented vehicle. The data are also used to assess the proposed framework, confirming human-like and fully-adaptive characteristics.     

Nonlinear multivariate time–space threshold vector error correction model for short term traffic state prediction

We propose Time–Space Threshold Vector Error Correction (TS-TVEC) model for short term (hourly) traffic state prediction. The theory and method of cointegration with error correction mechanism is employed in the general design of the new statistical model TS-TVEC. An inherent connection between mathematical form of error correction model and traffic flow theory is revealed through the transformation of the well-known Fundamental Traffic Diagrams. A threshold regime switching framework is implemented to overcome any unknown structural changes in traffic time series. Spatial cross correlated information is incorporated with a piecewise linear vector error correction model. A Neural Network model is also constructed in parallel to comparatively test the effectiveness and robustness of the new statistical model. Our empirical study shows that the TS-TVEC model is an effective tool that is capable of modeling the complexity of stochastic traffic flow processes and potentially applicable to real time traffic state prediction.     

A discrete choice framework for modeling and forecasting the adoption and diffusion of new transportation services

Major technological and infrastructural changes over the next decades, such as the introduction of autonomous vehicles, implementation of mileage-based fees, carsharing and ridesharing are expected to have a profound impact on lifestyles and travel behavior. Current travel demand models are unable to predict long-range trends in travel behavior as they do not entail a mechanism that projects membership and market share of new modes of transport (Uber, Lyft, etc.). We propose integrating discrete choice and technology adoption models to address the aforementioned issue. In order to do so, we build on the formulation of discrete mixture models and specifically Latent Class Choice Models (LCCMs), which were integrated with a network effect model. The network effect model quantifies the impact of the spatial/network effect of the new technology on the utility of adoption. We adopted a confirmatory approach to estimating our dynamic LCCM based on findings from the technology diffusion literature that focus on defining two distinct types of adopters: innovator/early adopters and imitators. LCCMs allow for heterogeneity in the utility of adoption for the various market segments i.e. innovators/early adopters, imitators and non-adopters. We make use of revealed preference (RP) time series data from a one-way carsharing system in a major city in the United States to estimate model parameters. The data entails a complete set of member enrollment for the carsharing service for a time period of 2.5 years after being launched. Consistent with the technology diffusion literature, our model identifies three latent classes whose utility of adoption have a well-defined set of preferences that are significant and behaviorally consistent. The technology adoption model predicts the probability that a certain individual will adopt the service at a certain time period, and is explained by social influences, network effect, socio-demographics and level-of-service attributes. Finally, the model was calibrated and then used to forecast adoption of the carsharing system for potential investment strategy scenarios. A couple of takeaways from the adoption forecasts were: (1) placing a new station/pod for the carsharing system outside a major technology firm induces the highest expected increase in the monthly number of adopters; and (2) no significant difference in the expected number of monthly adopters for the downtown region will exist between having a station or on-street parking.     

Investigation and analysis of evidence of asymmetric churn in travel demand models

There is a large amount of research work that has been devoted to the understanding of travel behaviour and for the prediction of travel demand and its management. Different types of data including stated preference and revealed preference, as well as different modelling approaches have been used to predict this. Essential to most travel demand forecasting models are the concepts of utility maximisation and equilibrium, although there have been alternative approaches for modelling travel behaviour. In this paper, the concept of asymmetric churn is discussed. That is travel behaviour should be considered as a two way process which changes over time. For example over time some travellers change their mode of travel from car to bus, but more travellers change their mode from bus to car. These changes are not equal and result in a net change in aggregate travel behaviour. Transport planners often aim at producing this effect in the opposite direction. It is important therefore to recognise the existence of churns in travel behaviour and to attempt to develop appropriate policies to target different groups of travellers with the relevant transport policies in order to improve the transport system. A data set collected from a recent large survey, which was carried out in Edinburgh is investigated to analyse the variations in departure time choice behaviour. The paper reports on the results of the investigation.     

Optimal deployment of autonomous vehicle lanes with endogenous market penetration

This paper develops a mathematical approach to optimize a time-dependent deployment plan of autonomous vehicle (AV) lanes on a transportation network with heterogeneous traffic stream consisting of both conventional vehicles (CVs) and AVs, so as to minimize the social cost and promote the adoption of AVs. Specifically, AV lanes are exclusive lanes that can only be utilized by AVs, and the deployment plan specifies when, where, and how many AV lanes to be deployed. We first present a multi-class network equilibrium model to describe the flow distributions of both CVs and AVs, given the presence of AV lanes in the network. Considering that the net benefit (e.g., reduced travel cost) derived from the deployment of AV lanes will further promote the AV adoption, we proceed to apply a diffusion model to forecast the evolution of AV market penetration. With the equilibrium model and diffusion model, a time-dependent deployment model is then formulated, which can be solved by an efficient solution algorithm. Lastly, numerical examples based on the south Florida network are presented to demonstrate the proposed models.     

Volatility of car ownership, commuting mode and time in the UK

This paper has two objectives: to examine the volatility of travel behaviour over time and consider the factors explaining this volatility; and to estimate the factors determining car ownership and commuting by car. The analysis is based on observations of individuals and households over a period of up to 11 years obtained from the British Household Panel Survey (BHPS). Changes in car ownership, commuting mode and commuting time over a period of years for the same individuals/households are examined to determine the extent to which these change from year-to-year. This volatility of individual behaviour is a measure of the ease of change or adaptation. If behaviour changes easily, policy measures are likely to have a stronger and more rapid effect than if there is more resistance to change. The changes are “explained” in terms of factors such as moving house, changing job and employment status. The factors determining car ownership and commuting by car are analysed using a dynamic panel-data models.     

The relationship between transportation perceptions and behavior over time

The aim of this paper is to discuss cross-lagged panel analysis in terms of the causal inferences it generates about the relationship of beliefs about modes and mode choice behavior. Frequencies of use of the single-occupant automobile (SOA), bus and carpool, as well as beliefs about each of the modes, were collected from a sample of central business district commuters at two points in time. The belief variables for each mode were summed to form composite measures and were corrected for unreliability due to measurement error.Perceptions of each mode and the frequency of its use were analyzed for influences operating over time. A time interval was assumed to exist during which the variables causally operated on each other. It was assumed that the time necessary for an individual to change modes based on his perception was equivalent to the interval required for a person to alter perceptions based on his experience. The causal structure relating the two variables was also assumed to be stable over time. An additional assumption was required to distinguish between third variable effects, or spuriousness, and dual causation: if a third variable were to be causing the relationship, it would be operating at a relatively constant rate over time.A strong causal relationship was found to be operating between beliefs about SOA and bus and use of those modes over time. The relationship is mutually causative; beliefs determine behavior and behavior reinforces and changes perceptions. Analysis of the carpool data indicated that the causal structure had changed over time and could not be analyzed with this technique. In general, support is evidenced for an adaptation or learning process interpretation of the relationship between beliefs and mode choice behavior.     

Does regional travel time unreliability influence mode choice?

Researchers and practitioners highlight the unreliability of travel as a potential weak link in the transportation system which may inhibit individuals’ accessibility and urban economic activity. With the trend towards increasing traffic congestion, the outlook suggests that travel conditions will become structurally less reliable over time, but that not all places will be equally affected. But is travel time unreliability a problem? This study uses global positioning systems travel survey data for Chicago to build a regional model of travel time unreliability. The results suggest that unreliability varies spatially during different time periods, but that the average overall network unreliability varies little across times in the day. Using the Chicago Metropolitan Agency for Planning (CMAP)’s 2007 Travel Tracker Survey, a household travel diary survey including both GPS and non-GPS components, we estimate a mode choice model for work trips to explore the influence of unreliability on travel behavior. The results suggest that unreliable auto travel conditions induce mode switching to transit and that the influence is strongest when service by train is already faster than by car. This further suggests that auto travel unreliability may have the strongest influence in metropolitan regions with highly-competitive transit systems. Nevertheless, the influence of travel unreliability is limited and is not the underlying driver of travel decision-making.     

Motorcycle-based adolescents’ travel behaviour during the school morning commute and the effect of intra-household interaction on departure time and mode choice

We hypothesise that intra-household interaction influences home departure time and mode choice for the morning commute. In Indonesia, over 71% of vehicles on the road are motorcycles. This fact increases the significance of household interaction in influencing transport mode choice since the simplicity of the motorcycle allows a great degree of versatility in regard to multiple family member transport. To emphasise this point, our study focuses on the unique travel behaviour of adolescents during the school morning commute which, due to the use of the motorcycle, is a combination of the travel behaviour of accompanied children and escorting adults. Our study discovers that adolescents are likely to shift their school arrival time very early or close to the designated starting time in relation to motorcycle-based parental escort to school. In regard to mode choice, adolescent students prefer to be escorted by motorcycle rather than take public transport.     

A comprehensive study of life course,cohort, and period effects on changes in travel mode use

This paper studies changes in people’s travel mode use from one year to the next. It is informed by three distinct discourses: travel behaviour change, the mobility biographies approach, and cohort analysis. The data used is the German Mobility Panel (GMP) 1994–2008 in which households and their members are asked three times in three subsequent years to report the trips they made over a week. The changes reported are regressed to key events over the life course, cohort effects and period effects, while various sociodemographic and spatial attributes are controlled. Due to the non-independent nature of panel observations, a cluster-robust regression approach is used. The findings suggest that behind the aggregate stability in travel mode use over time there is much change ‘under the surface’, induced by life course changes, individual and household sociodemographic, and spatial context. The changes found induced by life course related key events favour the notion of mobility biographies. However, taken over all key events seem to be relatively loosely associated with mode use changes. Nonetheless, various significant effects of baseline variables suggest that mode use may change even in the absence of a key event.     

Using repeated cross-sectional travel surveys to enhance forecasting robustness: Accounting for changing mode preferences

Transportation system capacity and performance, urban form and socio-demographics define the influences and constraints conditioning the preferences of urban residents for different transport modes. Changes in characteristics of urban areas are likely to lead to changes in preferences for alternative modes of transport over time; as a consequence, statistical models to forecast mode choice need to be sensitive to both purposeful changes to urban systems as well as exogenous shocks. We make use of the 1996, 2001 and 2006 household surveys conducted in the Greater Toronto and Hamilton Area to study mode preference evolution and model forecasting performance. These repeated cross-sectional household surveys provide an opportunity to investigate aggregate structural changes in commuting mode preferences over time, in a manner sensitive to changes in the urban area. We focus on commuting mode choices because these trips are prime determinants of peak period congestion and peak spreading. We then address how to combine the three cross-sections econometrically in a robust way that allows for use of a single mode choice model across the entire period. Using independent data from 2012, we are able to compare the individual year and combined models in terms of forecasting performance to demonstrate the combined model’s more robust forecasting performance into the future.     

Train routing and timetabling problem for heterogeneous train traffic with switchable scheduling rules

This paper proposes a mathematical model for the train routing and timetabling problem that allows a train to occasionally switch to the opposite track when it is not occupied, which we define it as switchable scheduling rule. The layouts of stations are taken into account in the proposed mathematical model to avoid head-on and rear-end collisions in stations. In this paper, train timetable could be scheduled by three different scheduling rules, i.e., no switchable scheduling rule (No-SSR) which allows trains switching track neither at stations and segments, incomplete switchable scheduling rule (In-SSR) which allows trains switching track at stations but not at segments, and complete switchable scheduling rule (Co-SSR) which allows trains switching track both at stations and segments. Numerical experiments are carried out on a small-scale railway corridor and a large-scale railway corridor based on Beijing–Shanghai high-speed railway (HSR) corridor respectively. The results of case studies indicate that Co-SSR outperforms the other two scheduling rules. It is also found that the proposed model can improve train operational efficiency.     

Integrated models of land use and transportation for the autonomous vehicle revolution

ABSTRACT

The advent of the autonomous vehicle (AV) will affect not only the transportation system, but also future patterns of land development. Integrated land use and transportation models will be critical tools in assessing the path forward with this technology. Key questions with respect to land use impacts of AVs arise from potential changes in sensitivity to travel and reduced demand for parking. It is an open question whether AVs will induce urban sprawl, or whether spatial economies of agglomeration will mitigate any reductions in travel time sensitivity. The deployment of shared fleets of AVs would likely reduce parking demand, producing yet to be explored impacts on property development within existing urban footprints. We perform a critical assessment of currently operational models and their ability to represent the adoption of AVs. We identify the representation of time in such models as a vital component requiring additional development to model this new technology. Existing model applications have focused on the discrete addition of new infrastructure or policy at a fixed point in time, whereas AV adoption will occur incrementally through time. Stated adaptation surveys are recommended as tools to quantify preferences and develop relevant model inputs. It is argued that existing models that assume comparatively static equilibrium have been convenient in the past, but are insufficient to model technology adoption. In contrast, dynamic model frameworks lack sufficient structure to maintain reasonability under large perturbations from base conditions. The ongoing advancement of computing has allowed models to move away from being mechanistic aggregate tools, towards behaviourally rich depictions of individual persons and firms. However, much work remains to move from projections of existing conditions into the future, to the evolution of the spatial economy as it evolves through time in response to new technologies and exogenous stresses. Principles from complex and evolutionary systems theory will be important in the development of models with the capacity to consider such dynamics.