Big Trajectory Data Mining: A Survey of Methods, Applications, and Services.

The increasingly wide usage of smart infrastructure and location-aware terminals has helped increase the availability of trajectory data with rich spatiotemporal information. The development of data mining and analysis methods has allowed researchers to use these trajectory datasets to identify urban reality (e.g., citizens' collective behavior) in order to solve urban problems in transportation, environment, public security, etc. However, existing studies in this field have been relatively isolated, and an integrated and comprehensive review is lacking the problems that have been tackled, methods that have been tested, and services that have been generated from existing research. In this paper, we first discuss the relationships among the prevailing trajectory mining methods and then, classify the applications of trajectory data into three major groups: social dynamics, traffic dynamics, and operational dynamics. Finally, we briefly discuss the services that can be developed from studies in this field. Practical implications are also delivered for participants in trajectory data mining. With a focus on relevance and association, our review is aimed at inspiring researchers to identify gaps among tested methods and guiding data analysts and planners to select the most suitable methods for specific problems.

Subjective evaluations of self and others' driving behaviors: A comparative study involving data from drivers in Japan, China, and Vietnam.

INTRODUCTION: This study explored the influence of personal attributes on subjectively-reported aggressive driving behaviors, with an emphasis on the inter-influences between subjectively-reported aggressive driving behaviors between self and other individuals. To determine this, a survey was conducted comprising participants' socio-demographic data, information on their history with automotive accidents, and subjective scales to report on the driving behaviors between self and others. More specifically, a four-factor shortened version of the Manchester Driver Behavior Questionnaire was used to collect data on the aberrant driving behaviors of "self" and "others." METHOD: Participants were recruited from three countries, namely, Japan (1,250 responses), China (1,250), and Vietnam (1,000). This study only considered the "aggressive violations' factor," which was referred to as self-aggressive driving behaviors (SADB) and others' aggressive driving behaviors (OADB). After collecting the data, univariate and bivariate multiple regression models were employed to better understand the response patterns from both scales. RESULTS: This study found that accident experience had the strongest influence on the reporting of aggressive driving behaviors (followed by education level). However, variation in countries was also found between both the rate of engagement in aggressive driving behavior and its recognition. In this study, highly educated Japanese drivers tended to evaluate others as safe, whereas highly educated Chinese drivers tended to evaluate others as aggressive. This discrepancy can likely be attributed to cultural norms and values. Meanwhile, evaluations from Vietnamese drivers seemed to differ depending on whether they drove cars or bikes, with additional influences as a result of the driving frequency. Furthermore, this study found that it was most difficult to explain the driving behaviors on the "other" scale reported by Japanese drivers. PRACTICAL APPLICATIONS: These findings can aid policymakers and planners to develop road safety measures that reflect the behaviors of drivers in their respective countries.

An alternative closed-form crash severity model with the non-identical, heavy-tailed, and asymmetric properties.

Crash severity model is a classical topic in road safety research. The multinomial logit (MNL) model, as a basic discrete outcome method, is widely applied to measure the association between crash severity and possible risk factors. However, the MNL model has several assumptions and properties that are possibly not consistent with the actual crash mechanism, and therefore with the association measure for crash severity. One significant attribute is the variation in drivers' safety perception. Risk-taking drivers tends to drive at a higher speed, which increases the likelihood of severe crashes. However, the variations in speed and other driving performance lead to the error in the utility function more profound. This violates the assumption of identical error distributions between different crash severity outcomes. In this paper, we propose a multinomial multiplicative (MNM) model, as an alternative for crash severity model. There are two possible formulations for the proposed MNM model: (1) Weibull and (2) Fréchet, according to the distributions of random propensities and subject to the signs of the systematic parts of the regression equation. The two heavy-tailed distributions can capture the effect of unobserved contributory factors on crash injury severity. Additionally, the MNM model can incorporate the effects of the non-identical, heavy-tailed, and asymmetric properties of the distribution, whereas the conventional MNL model cannot. Several operational considerations are also attempted in this study, including the specifications of the systematic parts and the interpretations of the parameters. The MNM model is further extended to the mixed MNM (MMNM) model by considering unobserved heterogeneities using random coefficients, while the mixed MNL (MMNL) model is used as the benchmark model. The proposed MMNM model is calibrated using the crash dataset obtained from the Guangdong Province, China. Results indicated that the proposed MMNM model outperformed the MMNL model in this case. Also, the results of parameter estimates are indicative to impact factors on crash severity as well as the design and implementation of policies. This justified the use of MMNM model as an alternative for crash severity model in practice. This is the first application of MMNM model in the traffic safety literature, it is worth exploring the application of other advanced multiplicative models for safety analysis in the future.

Determining passenger car equivalent (PCEs) for pretimed signalized intersections with severe motorcycle composition using Bayesian linear regression.

Pretimed signalized intersection is known as a common source of congestion, especially in urban heterogeneous traffic. Furthermore, the accuracy of saturation flow rate is found to cause efficient and vital capacity estimation, in order to ensure optimal design and operation of the signal timings. Presently, the traffic also consists of diverse vehicle presence, each with its own static and dynamic characteristics. The passenger car equivalent (PCE) in an essential unit is also used to measure heterogenous traffic into the PCU (Passenger Car Unit). Based on the collection of observed data at three targets in Banda Aceh City, this study aims to redetermine the PCEs by using Bayesian linear regression, through the Random-walk Metropolis-Hastings and Gibbs sampling. The result showed that the obtained PCE values were 0.24, 1.0, and 0.80 for motorcycle (MC), passenger car (PC), and motorized rickshaw (MR), respectively. It also showed that a significant deviation was found between new and IHCM PCEs, as the source of error was partially due to the vehicle compositions. The present traffic characteristics were also substantially different from the prevailing conditions of IHCM 1997. Therefore, the proposed PCEs enhanced the accuracy of base saturation flow prediction, provided support for traffic operation design, alleviated congestion, and reduced delay within the city, which in turn improved the estimation of signalized intersection capacity.

Influence of personality traits on aberrant driving behaviors: A comparison of Japanese, Chinese, and Vietnamese drivers.

INTRODUCTION: This study aims to explore the influence of Big Five personality traits in combination with various socio-demographic factors and experiences of accident involvement on aberrant driving behaviors. The study also compares the effects of the level of development (i.e., developed or developing) of three countries on the personality traits and driving behaviors. METHOD: The four-factor Driver Behavior Questionnaire was used to collect data on aberrant driving behaviors, while a short version of the 10-item Big Five Inventory was used to collect data on personality traits. Responses were collected from Japan (1,250 responses), China (1,250), and Vietnam (1,000). A latent variable model was applied after controlling data in each category (e.g., age). RESULTS: This study revealed that respondents who experienced accidents in the past and scored higher on Agreeableness were less likely to commit aggressive violations in Japan, China, and Vietnam. Further, Japanese and Vietnamese female drivers who scored high on Conscientiousness were found to be less likely to commit ordinary violations. Neuroticism was positively correlated with aggressive violations only in the case of Vietnamese drivers, irrespective of the history of accident involvement. CONCLUSIONS: Drivers with particular personality types that are linked with aberrant driving behavior may need to receive additional training on behavior management. Practical Applications: This study may help road traffic policymakers predict future driving behaviors of Vietnamese and Chinese drivers based on those of Chinese and Japanese drivers, respectively, and act accordingly.

Dataset on commuting patterns and mode-switching behavior under prospective policy scenarios for public transport.

This paper covers a broadly used methodology used in travel behavior research aiming at determining individual and alternative-specific variables that influence the choice of the transportation mode for commuting trips. Data used in the analysis were obtained in July 2015 by means of a computer-assisted telephonic interview survey conducted in Cluj Metropolitan Area, Romania. The survey collected a wide range of day-by-day travel patterns, socioeconomic data, and attitudes and perceptions toward urban transportation services. Given the lack of studies from emerging, post-socialist countries, the survey assigned a section dedicated to an alternative ticketing policy for public transport services in order to evaluate the willingness of commuters to switch to a more sustainable transportation through non-coercive interventions. A revealed preference - stated preference modelling methodology was adopted in order to reveal the role of socioeconomic characteristics, along with features of transport supply and built environment in explaining commuting patterns and forecast sustainable modal splits. Both the survey and the methodology are scalable and flexible to be used, adapted, and applied in a wide range of transport policies regarding modal shifting strategies.