Travel Behavior Modeling: Taxonomy, Challenges, and Opportunities

Smartphone-based travel surveys have attracted much attention recently, for their potential to improve data quality and response rate. One of the first such survey systems, Future Mobility Sensing (FMS), leverages sensors on smartphones, and machine learning techniques to collect detailed personal travel data. The main purpose of this research is to compare data collected by FMS and traditional methods, and study the implications of using FMS data for travel behavior modeling. Since its initial field test in Singapore, FMS has been used in several large-scale household travel surveys, including one in Tel Aviv, Israel. We present comparative analyses that make use of the rich datasets from Singapore and Tel Aviv, focusing on three main aspects: (1) richness in activity behaviors observed, (2) completeness of travel and activity data, and (3) data accuracy. Results show that FMS has clear advantages over traditional travel surveys: it has higher resolution and better accuracy of times, locations, and paths; FMS represents out-of-work and leisure activities well; and reveals large variability in day-to-day activity pattern, which is inadequately captured in a one-day snapshot in typical traditional surveys. FMS also captures travel and activities that tend to be under-reported in traditional surveys such as multiple stops in a tour and work-based sub-tours. These richer and more complete and accurate data can improve future activity-based modeling.

Download Full-text

Blue-collared Workers’ Travel Behavior Modeling using “exPlainable” Machine Learning Model: The Case of Qatar

10.29117/quarfe.2021.0198 ◽

2021 ◽

Author(s):

Aya Alkhereibi ◽

Ali AbuZaid ◽

Tadesse Wakjira

Keyword(s):

Machine Learning ◽

Travel Behavior ◽

Total Population ◽

Performance Metrics ◽

Predictive Accuracy ◽

Mode Choice ◽

Behavior Modeling ◽

Significant Feature ◽

Machine Learning Model ◽

Occupation Level

This paper presents a novel study on the examination of explainable machine learning (ML) technique to predict the mode choice for communities with a majority of blue-collared workers. A total of 4875 trip records for 1050 blue-collared workers have been used to predict their travel mode choices based on 11 trips and socio-economic attributes. The data used in this paper are obtained from the Ministry of Transportation and Communication (MoTC), which targeted blue-collared workers as they represent 89% of the total population in the State of Qatar. A total of four ML models are evaluated to propose the best predictive model. The four models were examined using different performance metrics. The models’ prediction results showed that the random forest (RF) model had the highest accuracy with a predictive accuracy of 0.97. Moreover, SHapley Additive exPlanation (SHAP) approach is used to investigate the significance of the input features and explain the output of the RF model. The results of SHAP analysis revealed that occupation level is the most significant feature that influences the mode choice followed by occupation section, arrival time, and arrival municipality.

Download Full-text

Individual Travel Behavior Modeling of Public Transport Passenger Based on Graph Construction

Journal of Advanced Transportation ◽

10.1155/2018/3859830 ◽

2018 ◽

Vol 2018 ◽

pp. 1-13 ◽

Cited By ~ 3

Author(s):

Quan Liang ◽

Jiancheng Weng ◽

Wei Zhou ◽

Selene Baez Santamaria ◽

Jianming Ma ◽

...

Keyword(s):

Public Transport ◽

Travel Behavior ◽

Behavior Modeling ◽

The Public ◽

Demand Prediction ◽

Study Results ◽

Characteristics Analysis ◽

Significant Change ◽

The Individual ◽

Behavior Choice

This paper presents a novel method for mining the individual travel behavior regularity of different public transport passengers through constructing travel behavior graph based model. The individual travel behavior graph is developed to represent spatial positions, time distributions, and travel routes and further forecasts the public transport passenger’s behavior choice. The proposed travel behavior graph is composed of macronodes, arcs, and transfer probability. Each macronode corresponds to a travel association map and represents a travel behavior. A travel association map also contains its own nodes. The nodes of a travel association map are created when the processed travel chain data shows significant change. Thus, each node of three layers represents a significant change of spatial travel positions, travel time, and routes, respectively. Since a travel association map represents a travel behavior, the graph can be considered a sequence of travel behaviors. Through integrating travel association map and calculating the probabilities of the arcs, it is possible to construct a unique travel behavior graph for each passenger. The data used in this study are multimode data matched by certain rules based on the data of public transport smart card transactions and network features. The case study results show that graph based method to model the individual travel behavior of public transport passengers is effective and feasible. Travel behavior graphs support customized public transport travel characteristics analysis and demand prediction.

Download Full-text

Personalized Choice Model for Managed Lane Travel Behavior

Transportation Research Record Journal of the Transportation Research Board ◽

10.1177/0361198120923355 ◽

2020 ◽

Vol 2674 (7) ◽

pp. 442-455

Author(s):

Yifei Xie ◽

Yundi Zhang ◽

Arun Prakash Akkinepally ◽

Moshe Ben-Akiva

Keyword(s):

Travel Behavior ◽

Choice Model ◽

Explanatory Power ◽

Revealed Preference ◽

Discrete Choice Models ◽

Hierarchical Bayes ◽

Behavior Modeling ◽

Parameter Distribution ◽

Explanatory Variables ◽

Prediction Capability

This paper presents a methodology for enhancing discrete choice models for managed lane travel behavior with personal trip history. We refer to this process as personalization and the enhanced model as a personalized choice model. With the objective of better understanding managed lane choices and improving the model’s prediction capability, personalization was carried out at two levels. First, we used each traveler’s habits and travel experiences before each trip for constructing a set of explanatory variables that could be used with any model structure. Second, under a logit mixture framework, the distribution of random parameters was updated with Bayesian inference according to personal trip history. The structure of the parameter distribution explicitly considered preference variations across individuals (interpersonal heterogeneity), as well as preference variations across trips performed by the same individual (intrapersonal heterogeneity). The proposed methodology is especially relevant for modeling revealed preference (RP) data from automatic vehicle identification sensors, for which limited socioeconomic characteristics of travelers are available. An empirical study was conducted on an operational managed lane corridor near Dallas/Fort Worth Airport in Texas. Available trip records over a 5-month period were utilized. A hierarchical Bayes estimator was adopted for efficient model estimation. The results suggest significant inter- and intrapersonal heterogeneity and that the proposed personalization method improves the model’s explanatory power and prediction capability. To the best of our knowledge, this paper represents the first introduction of personalization in managed lane choice behavior modeling and the first attempt to estimate intrapersonal heterogeneity with RP data.

Download Full-text