Decision tree method for modeling travel mode switching in a dynamic behavioral process

As road congestion is exacerbated in most metropolitan areas, many transportation policies and planning strategies try to nudge travelers to switch to other more sustainable modes of transportation. In order to better analyze these strategies, there is a need to accurately model travelers’ mode-switching behavior. In this paper, a popular artificial intelligence approach, the decision tree (DT), is used to explore the underlying rules of travelers’ switching decisions between two modes under a proposed framework of dynamic mode searching and switching. An effective and practical method for a mode-switching DT induction is proposed. A loss matrix is introduced to handle class imbalance issues. Important factors and their relative importance are analyzed through information gains and feature selections. Household Travel Survey data are used to implement and validate the proposed DT induction method. Through comparison with logit models, the improved prediction ability of the DT models is demonstrated.     

A tour-based model of travel mode choice

This paper presents a new tour-based mode choice model. The model is agent-based: both households and individuals are modelled within an object-oriented, microsimulation framework. The model is household-based in that inter-personal household constraints on vehicle usage are modelled, and the auto passenger mode is modelled as a joint decision between the driver and the passenger(s) to ride-share. Decisions are modelled using a random utility framework. Utility signals are used to communicate preferences among the agents and to make trade-offs among competing demands. Each person is assumed to choose the best combination of modes available to execute each tour, subject to auto availability constraints that are determined at the household level. The households allocations of resources (i.e., cars to drivers and drivers to ride-sharing passengers) are based on maximizing overall household utility, subject to current household resource levels. The model is activity-based: it is designed for integration within a household-based activity scheduling microsimulator. The model is both chain-based and trip-based. It is trip-based in that the ultimate output of the model is a chosen, feasible travel mode for each trip in the simulation. These trip modes are, however, determined through a chain-based analysis. A key organizing principle in the model is that if a car is to be used on a tour, it must be used for the entire chain, since the car must be returned home at the end of the tour. No such constraint, however, exists with respect to other modes such as walk and transit. The paper presents the full conceptual model and estimation results for an initial empirical prototype. Because of the complex nature of the model decision structure, choice probabilities are simulated from direct generation of random utilities rather than through an analytical probability expression.     

Dynamic travel mode searching and switching analysis considering hidden model preference and behavioral decision processes

This paper proposes a conceptual framework to model the travel mode searching and switching dynamics. The proposed approach is structurally different from existing mode choice models in the way that a non-homogeneous hidden Markov model (HMM) has been constructed and estimated to model the dynamic mode srching process. In the proposed model, each hidden state represents the latent modal preference of each traveler. The empirical application suggests that the states can be interpreted as car loving and carpool/transit loving, respectively. At each time period, transitions between the states are functions of time-varying covariates such as travel time and travel cost of the habitual modes. The level-of-service (LOS) changes are believed to have an enduring impact by shifting travelers to a different state. While longitudinal data is not readily available, the paper develops an easy-to-implement memory-recall survey to collect required process data for the empirical estimation. Bayesian estimation and Markov chain Monte Carlo method have been applied to implement full Bayesian inference. As demonstrated in the paper, the estimated HMM is reasonably sensitive to mode-specific LOS changes and can capture individual and system dynamics. Once applied with travel demand and/or traffic simulation models, the proposed model can describe time-dependent multimodal behavior responses to various planning/policy stimuli.     

A utility-theory travel demand model incorporating travel budgets

A model of traveler behavior is proposed which is consistent with the possibility that travelers expend average daily amounts of time and money on travel with stable regularities both among urban areas and over time in the same area. The model is founded on economic utility theory. It is designed to forecast: (1) the amount of total travel generated by types of households, (2) the division of travel among available modes, and (3) the relationship between the amounts of time and money allocated to travel expenditures. The qualitative properties of the model are shown to be consistent with economic principles. Specific theoretical results reveal that, in the simultaneous presence of constraints on both time and money, travel budgets are not strictly constant proportions of income and time available as they are in the cases of single constraints relevant to classes of travelers to whom time is scarce compared to money, or conversely. Constant expenditure proportions are shown to be linear approximations which are subject to empirical validation. The relevant economic principle is that expenditures can be considered fixed in the short run but become flexible in the long run when utility maximization is applied to the expenditures themselves and not just to their allocation. Empirical tests of the model using data from three urban areas are positive, but additional tests are called for. The most important output of the research is deemed to be the establishment of theoretical hypotheses which can be used in continuing tests of travel budgets.     

A bayesian dynamic linear model approach for real-time short-term freeway travel time prediction

This paper presents a Bayesian inference-based dynamic linear model (DLM) to predict online short-term travel time on a freeway stretch. The proposed method considers the predicted freeway travel time as the sum of the median of historical travel times, time-varying random variations in travel time, and a model evolution error, where the median is employed to recognize the primary travel time pattern while the variation captures unexpected supply (i.e. capacity) reduction and demand fluctuations. Bayesian forecasting is a learning process that revises sequentially the state of a priori knowledge of travel time based on newly available information. The prediction result is a posterior travel time distribution that can be employed to generate a single-value (typically but not necessarily the mean) travel time as well as a confidence interval representing the uncertainty of travel time prediction. To better track travel time fluctuations during non-recurrent congestion due to unforeseen events (e.g., incidents, accidents, or bad weather), the DLM is integrated into an adaptive control framework that can automatically learn and adjust the system evolution noise level. The experiment results based on the real loop detector data of an I-66 segment in Northern Virginia suggest that the proposed method is able to provide accurate and reliable travel time prediction under both recurrent and non-recurrent traffic conditions.     

A model for assessing the effects of dynamic travel time information via variable message signs

In view of the serious traffic congestion during peak hours in most metropolitan areas around the world and recent improvement of information technology, there is a growing aspiration to alleviate road congestion by applications of electronic information and communication technology. Providing drivers with dynamic travel time information such as estimated journey times on major routes should help drivers to select better routes and guide them to utilise existing expressway network. This can be regarded as one possible strategy for effective traffic management. This paper aims to investigate the effects and benefits of providing dynamic travel time information to drivers via variable message signs at the expressway network. In order to assess the effects of the dynamic driver information system with making use of the variable message signs, a time-dependent traffic assignment model is proposed. A numerical example is used to illustrate the effects of the dynamic travel time information via variable message signs. This revised version was published online in June 2006 with corrections to the Cover Date.     

Semiparametric specification of the utility function in a travel mode choice model

We propose a semiparametric approach that can capture the nonlinearity of deterministic components of the utility functions in discrete choice models and demonstrate it by analyzing travel mode choice behaviour for an interregional trip. The proposed smoothing spline-based specification method can be used to make ex ante evaluations regarding the parametric specifications of the deterministic utility functions in discrete choice models.     

Values,attitudes and travel behavior: a hierarchical latent variable mixed logit model of travel mode choice

Values lie at the heart of an individual’s belief system, serving as prototypes from which attitudes and behaviors are subsequently manufactured. Attitudes and behaviors may evolve over time, but values represent a set of more enduring beliefs. This study examines the influence of values on travel mode choice behavior. It is argued that personal values influence individual attitudes towards different alternative attributes, which in turn impact modal choices. Using data from a sample of 519 German commuters drawn from a consumer panel, the study estimates an integrated choice and latent variable model of travel mode choice that allows for hierarchical relationships between the latent variables and flexible substitution patterns across the modal alternatives. Results from the empirical application support the value-attitude-behavior hierarchical model of cognition, and provide insights to planners and policy-makers on how better to sell public transit as a means of travel.     

A demand model for recreational travel

A model is developed to describe and to predict the patterns of regional recreational travel. The model is designed in such a manner to allow its calibration and use without the need to conduct extensive travel surveys in a large region. To allow its use for prediction, the model is based on a causal structure and attempts to derive recreational travel demand from behavioural variables. The main hypothesis of the model is that the amount of recreational travel a recreation area attracts is affected by the accessibility of this area to points of demand potential and by its attractiveness relative to the recreation areas.

The calibration is founded on actual data on recreational travel to national forests in California, U.S.A. It is found in the calibration that accessibility to demand potential is the single most important determinant of recreational travel attraction. A simple relationship is derived to relate travel to each national forest to the relative accessibility of the forest. The model is calibrated and statistically validated.

It is suggested that when constructing travel demand models simplicity be sought, even at the risk of the loss of some explanatory power. In the calibration of such models statistical significant is more important than the ability to reproduce observed patterns.     

Habitual travel behaviour: Evidence from a six-week travel diary

This paper introduces different methods to measure similarity of travel behaviour addressing the question of how repetitious travel behaviour actually is. It compares empirical results of the different methods based on the data from a six-week travel diary. In general, the results show that the day-to-day behaviour is more variable if measured with trip-based methods instead of methods based on time budgets. Furthermore, it is confirmed that the similarity declines if the method captures more of the complexity of the travel pattern. It is also shown that travel behaviour is neither totally repetitious nor totally variable. Even for the whole observation period, it is demonstrated that two days always have some common elements. Additionally, it is found that the different methods yield the same pattern of variability for different types of day. Travel behaviour is clearly more stable on work days. Similar results for all methods are also obtained concerning the question of how long the minimum period of observation should be. All measures show that the period should not be less than two weeks if one aims at measuring variability.     

Activity settings and travel behaviour: A social contact perspective

Using time-use data from Canada, Norway, and Sweden, this study briefly outlines the essence of the activity setting approach and illustrates one aspect of its usefulness by exploring the impact of social contact on travel behaviour. The activity system approach views behaviour in context. Activity settings are generic components of the activity system and studying them using time-use diaries can provide major insights into travel behaviour. Focusing on social contact, this paper characterizes the social environment in terms of social circle (interaction partners) and social space (location). The analysis shows that there are clear differences in the levels of social interaction across various groups, including those who work at home. The 1992 Canadian data showed people working at the workplace spend relatively more time with others, about 50% of total time awake. Working at home reduced the time with others to a low of 15.7%. when people worked at home the family benefited, almost doubling the time spent with them compared to those working at the workplace. Persons working at home only spend the most time alone. There is a tendency for persons with low social interaction to travel more. It is argued that individual need, or want, social contact and if they cannot find it at the workplace they will seek it elsewhere thus generating travel. Whether this is the result of need or opportunity is of minor relevance, what it does suggest is that working in isolation at home will not necessarily diminish travel but rather may simply change its purpose. This revised version was published online in June 2006 with corrections to the Cover Date.     

Some issues of dynamics in forecasting travel behaviour — A discussion paper

In the next few years there is likely to be a large growth of interest in the dynamic modelling of travel behaviour. In order to try to avoid the eventual collapse of enthusiasm which has sometimes occurred with other new developments when they turn out not to provide transport planning with the hoped-for panacea, this paper aims to demonstrate the diversity of approaches which will be required to tackle the subject of dynamics. In particular, it identifies three overlapping but distinct levels at which dynamics interact with travel behaviour — microdynamics, which is concerned with the detailed scheduling of activities and travel within a day — macrodynamic modifiers, whereby changes in medium- and long-term behaviour which are often considered to be instantaneous are in fact subject to important phasing considerations — and macrodynamic processes, which deal with the overriding demographic processes of birth, ageing and death. The paper suggests approaches to the incorporation of these three topics into the forecasting of travel behaviour.     

Effects of increased awareness on choice of travel mode

This paper reports a field experiment with the purpose of studying the effects of increased awareness on travel mode choice. One hundred fifteen subjects were randomly assigned to an experimental and a control group. In the experimental group, a more deliberate choice of travel mode was induced and expected to result in a stronger relationship between attitude and behavior, a weaker relationship between habit and behavior, and a behavioral change among individuals with a strong habit. Attitude, habit, and behavior were measured in travel diaries and questionnaires. The results indicated no significant change in the relationship between attitude and behavior and no significant change in the relationship between habit and behavior. However, a temporally extended decrease in car use was observed in the experimental group. The effect was noted for individuals with a strong habit who reduced their car use but not for subjects with a weak habit.     

The significance and measurement of variability in travel behaviour

It is argued that an understanding of variability is central to the modelling of travel behaviour and the assessment of policy impacts, and is not the peripheral issue that it has often been considered. Drawing on recent studies in the UK and Australia, in conjunction with a review of the literature, the paper first examines the policy and analytical rationale for using multi-day data, then illustrates different ways of measuring variability, and finally discusses issues relating to the collection of suitable data for such analyses. In a policy context, there is a growing need for multi-day data to examine issues that affect general rather than one-day behaviour (e.g. to assess the distribution of user charges for road pricing, or patterns of public transport usage); while analytically, multi-day data is needed to improve our ability to identify the mechanisms behind travel behaviour and to derive better empirical relationships. Three measures of variability are presented: a graphical form showing daily differences in behaviour at the individual level; an aggregate, similarity index; and a hybrid graphical/numerical measure, which provides new insights into variability in daily patterns of behaviour. The paper raises a number of issues for debate, probably the most crucial of which is: variability in what? The way in which behaviour is measured crucially affects our conception of stability and variability.     

A data-driven travel mode share estimation framework based on mobile device location data

Transportation - Mobile device location data (MDLD) contains abundant travel behavior information to support travel demand analysis. Compared to traditional travel surveys, MDLD has larger...     

A time-use activity-pattern recognition model for activity-based travel demand modeling

This study develops a new comprehensive pattern recognition modeling framework that leverages activity data to derive clusters of homogeneous daily activity patterns, for use in activity-based travel demand modeling. The pattern recognition model is applied to time use data from the large Halifax STAR household travel diary survey. Several machine learning techniques not previously employed in travel behavior analysis are used within the pattern recognition modeling framework. Pattern complexity of activity sequences in the dataset was recognized using the FCM algorithm, and resulted in identification of twelve unique clusters of homogeneous daily activity patterns. We then analysed inter-dependencies in each identified cluster and characterized the cluster memberships through their socio-demographic attributes using the CART classifier. Based on the socio-demographic characteristics of individuals we were able to correctly identify which cluster individuals belonged to, and also predict various information related to their activities, such as start time, duration, travel distance, and travel mode, for use in activity-based travel demand modeling. To execute the pattern recognition model, the 24-h activity patterns are split into 288 three dimensional 5 min intervals. Each interval includes information on activity types, duration, start time, location, and travel mode if applicable. Results from aggregated statistical evaluation and Kolmogorov–Smirnov tests indicate that there is heterogeneous diversity among identified clusters in terms of temporal distribution, and substantial differences in a variety of socio-demographic variables. The homogeneous clusters identified in this study may be used to more accurately predict the scheduling behavior of specific population groups in activity-based modeling, and hence to improve prediction of the times and locations of their travel demands. Finally, the results of this study are expected to be implemented within the activity-based travel demand model, Scheduler for Activities, Locations, and Travel (SALT).

     

An activity-based approach of investigating travel behaviour of older people

The paper presents the results of an investigation on daily activity-travel scheduling behaviour of older people by using an advanced econometric model and a household travel survey, collected in the National Capital Region (NCR) of Canada in 2011. The activity-travel scheduling model considers a dynamic time–space constrained scheduling process. The key contribution of the paper is to reveal daily activity-travel scheduling behaviour through a comprehensive econometric framework. The resulting empirical model reveals many behavioural details. These include the role that income plays in moderating out-of-home time expenditure choices of older people. Older people in the highest and lowest income categories tend to have lower variations in time expenditure choices than those in middle-income categories. Overall, the time expenditure choices become more stable with increasing age, indicating that longer activity durations and lower activity frequency become more prevalent with increasing age. Daily activity type and location choices reveal a clear random utility-maximizing rational behaviour of older people. It is clear that increasing spatial accessibility to various activity locations is a crucial factor in defining daily out-of-home activity participation of older people. It is also clear that the diversity of out-of-home activity type choices reduces with increasing age and older people are more sensitive to auto travel time than to transit or non-motorized travel time.     

A real time forecasting tool for dynamic travel time from clustered time series

This paper addresses the problem of dynamic travel time (DTT) forecasting within highway traffic networks using speed measurements. Definitions, computational details and properties in the construction of DTT are provided. DTT is dynamically clustered using a K-means algorithm and then information on the level and the trend of the centroid of the clusters is used to devise a predictor computationally simple to be implemented. To take into account the lack of information in the cluster assignment for the new predicted values, a weighted average fusion based on a similarity measurement is proposed to combine the predictions of each model. The algorithm is deployed in a real time application and the performance is evaluated using real traffic data from the South Ring of the Grenoble city in France.     

Experiments with an activity-based travel model

Transportation - Experiments are described with an activity-based travel model, estimated on a 7-day activity-diary survey. The first part of the paper describes the model system in its final form,...