Smartphone-based Travel Survey Research Articles

Recruitment, participant motivation and response rates in a smartphone-based travel survey: Mobility Panel in aspern Seestadt

The aspern Seestadt Mobility Panel is a longitudinal smartphone-based mobility survey that has been running since 2019 in the urban development zone of aspern Seestadt, Vienna. Although it is easy to record routes using smartphones a comprehensive strategy is still needed to recruit and activate sufficient participants to ensure representativeness of the survey. Key elements of this strategy include understanding the motivation of participants (both to participate and drop-out). This paper describes the strategies used to activate and keep aspern Seestadt Mobility Panel participants during the first two survey years and recommendations for changes in future years.

Online and in-person activity logging using a smartphone-based travel, activity, and time-use survey

With increased internet connectivity, there are more opportunities to substitute physical, or “in-person”, activities with online activities. With this, changes to household and individual travel/activity patterns are to be expected. For example, the increased adoption of e-commerce can affect individual travel destinations and frequency. To study these shifts, we enhance an existing smartphone-based travel/activity diary to capture online activities, with a focus on activities that can replace passenger trips. Furthermore, we expand the options to report shopping details, to better understand the influence of e-commerce on in-person shopping. In this paper, we detail the data collection tool development strategy and user-experience findings. Moreover, we provide insights from a pilot deployment in Singapore during 2020, which allowed measuring travel/activity behavior, expenditure, and time allocation changes during various COVID-19 pandemic travel restriction periods.

Insights on data quality from a large-scale application of smartphone-based travel survey technology in the Phoenix metropolitan area, Arizona, USA

Collecting accurate travel data is vital for transportation planning purposes. Regional travel demand forecasts as well as transportation system analyses depend on datasets that provide origins and destinations of travel for various modes, purposes of travel, socio-economic characteristics of the system users, and other attributes critical for understanding travel demand. GPS-based household travel surveys emerged as a state-of-the-practice method to collect travel data with increased accuracy and detail. The Maricopa Association of Governments conducted a survey utilizing Future Mobility Sensing (FMS) technology. One hundred percent of the sample was collected with the FMS technology platform that combines mobile sensing through a smartphone app with machine learning and a user interface. The technology enables detailed, multi-day, multimodal, user-verified travel and activity behavior data to be obtained with a reduced burden on participants. The data collected through the survey was analyzed together with a comparable dataset obtained through traditional recall-based collection methods during the same time period. The broad conclusions are that the 100% GPS-based surveys with the FMS technology platform provide greater accuracy, detail and completeness of data, as well as greater flexibility than traditional data collection approaches that rely on participant recall. Emphasis was made on comparative analyses between traditionally collected data and the GPS survey with the FMS technology. The paper systematically identifies and explains differences and provides original analyses that can inform future decision making relevant to similar data collection exercises. The method is particularly applicable for monitoring mobility in the ongoing conditions of rapidly changing travel behavior, especially due to the COVID-19 pandemic.

Introducing a smart phone-based travel survey application

Travel demand models, which are simplified representations of the real-world trip decision making processes, involve a series of steps that attempt to simulate complex human behaviour.

Stop detection for smartphone-based travel surveys using geo-spatial context and artificial neural networks

The problem of stop detection is at the base of many current and upcoming smartphone-based travel survey technologies and directly impacts the quality of many downstream operations. The inference of departure/arrival time, mode, and purpose of a trip, for example, rely on the stop/motion patterns represented by smartphone sensor-data. As users handle smartphones for various purposes and their preferences determine different device positions while traveling, accelerometer and gyroscope, for instance, often present ambiguities that prevent accurate stop detection.To mitigate the impact of these ambiguities, we combine spatial time-series, i.e. GPS, with spatial context information retrieved from Open Street Map, which we represent as multi-dimension tensors. This project explores simple representations, such as dummy variables, and novel multidimensional representations, which are bench-marked through the classification performance of specialized artificial neural network (ANN), as well as other machine learning (ML) baselines. Our main contribution stems from this novel multidimensional representation of time-series fusion with spatial context, combined with the corresponding specialized ANN classifier. The results show a stop detection score improvement on the baselines between 3% and 6.5%.

Investigating response, data sending, and completion rates of a smartphone-based travel survey conducted in Kabul, Afghanistan

This study concentrates on the response, data sending, and survey completion rates of a smartphone-based travel survey in a developing country. A smartphone-based survey reinforced by a web-based survey was conducted in 2015 in Kabul, Afghanistan. The aims of this study are: 1) to explore the effects of age and gender on the response rate of the smartphone-based travel survey, and 2) to examine how gender, age, smartphone ownership history, and occupation influence the data sending and completion rates of the smartphone-based travel survey. The results show that gender had no effects on the response rate, while age had a significant impact. In addition, gender had effect on data sending rate, and the respondents who owned smartphones for longer time were more likely to complete the survey successfully. Furthermore, the survey withdraw rate of respondents was higher on weekends in comparison to weekdays.

Technical requirements for organising successful mobility campaigns in citizen observatories

Traditional travel surveys are usually organised in a top-down manner, when data collection is commissioned by transport authorities. Recently, however, there has been an increasing tendency towards bottom-up initiatives when citizens and stakeholders initiate and implement data collection campaigns in citizen observatories. Citizen observatories are usually based on participatory sensing, i.e. data collection by volunteers using smart mobile devices. The ubiquity of smartphones and hardware developments make it a potentially useful instrument to have travel behaviour registered. However, creating a new citizen observatory remains technically difficult and labour-intensive. Today, citizen observatories have to be developed from scratch, which acts as a significant barrier to their implementation. Recently, an open, reusable and reconfigurable citizen observatory platform has been proposed. Through such a platform, ICT-agnostic stakeholders are able to initiate new citizen observatories in various domains.In this paper we have developed a general technical framework for a platform that can allow stakeholders to create citizen observatories, with the aim of collecting data on travel behaviour. The framework is based on the available functionalities of travel survey applications, elaborating on the potential limitations and enhancements of a citizen observatory platform compared to travel survey apps. First, we compared the functionalities of smartphone-based travel surveys and citizen observatories, which exposed the main areas of concern of citizen observatories compared to traditional travel surveys. Next, we developed a framework for the technical requirements to organise a citizen observatory for mobility. The three main stages in this framework are the collection of raw data, processing it into valuable behavioural and environmental information (e.g. travel modes, air pollution) and the feedback of this data to all stakeholders. Data validation should be applied in each stage and it is important to ensure user acceptance during the entire lifespan of the campaign. We highlighted points for attention that should be considered when developing a citizen observatory for mobility. Defining the required accuracy, representativeness and reliability in a specific campaign are key questions in the context of an acceptable rate of battery drain. The technical requirements developed in this paper reflect a ‘best-case scenario’ that will later be tested with specific use cases.

Improvement of Data Sending Rate and Survey Completion Rate of Smartphone-based Travel Surveys in Two Afghanistan Cities

Improvement of Data Sending Rate and Survey Completion Rate of Smartphone-based Travel Surveys in Two Afghanistan Cities

Detecting travel modes from smartphone-based travel surveys with continuous hidden Markov models

In the last decades, studies on travel mode detection from location data have been increasing exponentially. However, these studies have struggled with three limitations: data collection-, feature ...

Context-aware stated preferences with smartphone-based travel surveys

Stated preferences surveys are most commonly used to provide behavioral insights on hypothetical travel scenarios such as new transportation services or attribute ranges beyond those observed in existing conditions. When designing SP surveys, considerable care is needed to balance the statistical objectives with the realism of the experiment. This paper presents an innovative method for smartphone-based stated preferences (SP) surveys leveraging state-of-the-art smartphone-based survey platforms and their revealed preferences sensing capabilities. A random experimental design generates context-aware SP profiles using user specific socioeconomic characteristics and past travel data along with relevant web data for scenario generation. The generated choice tasks are automatically validated to reduce the number of dominant or inferior alternatives in real-time, then validated using Monte-Carlo simulations offline. In this paper we focus our attention on mode choice and design an experiment that considers a wide range of possible existing mode alternatives along with a new alternative on-demand mobility service that does not exist in real life. This experiment is then used to collect SP data or a sample of 224 respondents in the Greater Boston Area. A discrete mode choice model is estimated to illustrate the benefit of the proposed method in capturing current context-specific preferences in response to the new scenario.

Reasons for Non-response to Smartphone-based Travel Survey in Two Afghanistan Cities

Reasons for Non-response to Smartphone-based Travel Survey in Two Afghanistan Cities

Enriching Activity-Based Models using Smartphone-Based Travel Surveys

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.

Estimating a Toronto pedestrian route choice model using smartphone GPS data

This study examines the feasibility of using revealed preference GPS data collected through a smartphone-based travel survey and discrete choice modeling techniques to determine pedestrians’ preferences towards street infrastructure, built environment, and land use. Smartphone GPS points were collected after 50 m of travel and had a horizontal spatial accuracy of 30 m or less. A path size logit model with stochastic route choice generation choice set was used for this model. The results of the model showed that distance, the number of turns, the number of signalized intersections, and distance along links with sidewalks on both sides of the street were significant variables in the route choice model. Turns are found to be equivalent to an additional 32 m, signalized intersections are equivalent to a reduction of 34 m, and travel along streets with sidewalks on both sides of the road is evaluated as 33% shorter than streets with other sidewalk conditions. While the dataset used for this study was relatively small (776 trip observations), these results are consistent with other pedestrian route choice studies which support the viability of using smartphone GPS data for future pedestrian route choice studies.

Intrapersonal mode choice variation: Evidence from a four-week smartphone-based travel survey in the Netherlands

This paper examines mode choice variation in the Netherlands based on the trip data of 432 respondents from a four-week smartphone-based travel survey. Trip characteristics, including origin and destination location, arrival and departure time, mode and trip purpose, were automatically recorded, but checked and if necessary revised in a web-based prompted recall survey. Statistical analyses and mixed logit mode choice models were used to explore intrapersonal variation and its effect on mode choice. We found relatively much intrapersonal variation for short trips (<2 km) as respondents who usually travel by car also regularly walk and/or cycle. By contrast, intrapersonal variation was significantly smaller in trips longer than 10 km, suggesting that people choose the same mode when they repeat long journeys. The intrapersonal variation is also relatively small for commute trips, implying a high level of habituation. In addition, the results from the mixed logit mode choice models clearly show that including a classification of travellers determined by the degree of intrapersonal variation significantly explains mode choice.

Travel Mode Detection Using GPS Data and Socioeconomic Attributes Based on a Random Forest Classifier

The past few years have witnessed the rapid growth in the collection of large-scale GPS data via smartphone-based travel surveys around the world, following which transportation modes detection received significant attention. A mass of methods varying from Criteria-based rules to Machine Learning technology were employed to recognize the travel modes. However, the limited sample size, deficient feature selection and the less emphasis on addressing confusion modes, which leave room for improvement. This paper therefore sought to develop and evaluate a Random Forest classifier combined with a rule-based method to detect six travel modes (subway, walking, bicycle, e-bike, bus and car). Seven GPS-related variables are selected as feature set from the initial list of 22 variables. Consequently, more than 98% subway trips were correctly identified and the overall accuracy of the rest five modes classification is obtained as high as 93.11%. More than 85% trips were successfully identified for each mode except for the bus. More importantly, results show that socioeconomic attributes data could significantly improve the prediction of e-bike and address the confusion between bus and car modes. The employment of ROC curve provides a statistical proof to the excellent classification capacity of Random Forest in this study. Besides, the comparison with two representative classifiers demonstrates the applicability of Random Forest classifier for travel modes detection incorporating multi-source attributes.

Automatic Trip Detection with the Dutch Mobile Mobility Panel: Towards Reliable Multiple-Week Trip Registration for Large Samples

ABSTRACTThis paper examines the accuracy of trip and mode choice detection of the last wave of the Dutch Mobile Mobility Panel, a large-scale three-year, smartphone-based travel survey. Departure and arrival times, origins, destinations, modes, and travel purposes were recorded during a four week period in 2015, using the MoveSmarter app for a representative sample of 615 respondents, yielding over 60 thousand trips. During the monitoring period, respondents also participated in a web-based prompted recall survey and answered additional questions. This enables a comparison between automatic detected and reported trips. Most trips were detected with no clear biases in trip length or duration, and transport modes were classified correctly for over 80 percent of these trips. There is strong evidence that smartphone-based trip detection helps to reduce underreporting of trips, which is a common phenomenon in travel surveys. In the Dutch Mobile Mobility Panel, trip rates are substantially higher than trip-diary based travel surveys in the Netherlands, in particular for business and leisure trips which are often irregular. The rate of reporting also hardly decreased during the four-week period, which is a promising result for the use of smartphones in long duration travel surveys.

Assessment of trip validation interfaces for smartphone-based travel surveys

Abstract Solutions for smartphone-based travel surveys are currently mushrooming, yet methods to assess systematic validation of data are still sparse. We contribute to the field by selecting usability parameters for post-processing automated generated trip data relying on task completion time, errors and success rate as well as subjective ratings of trip validation interfaces. As a proof of concept, we apply this method to three different solutions for smartphone-based travel surveys. Our approach can serve as a template to assess interfaces that aim at achieving an effective, efficient, and user-friendly acquisition of human-entered data for validation of automatic trip data.

Automatic Trip End Imputation from Smartphone-based Travel Surveys

Automatic Trip End Imputation from Smartphone-based Travel Surveys

Enabling Bus Transit Service Quality Co-Monitoring Through Smartphone-Based Platform

The growing ubiquity of smartphones offers public transit agencies an opportunity to transform ways to measure, monitor, and manage service performance. The potential of a new tool is demonstrated for engaging customers in measuring satisfaction and co-monitoring [Editor’s note: This is the authors’ word, meaning “agencies using public feedback to supplement official monitoring and regulation.”] bus service quality. The pilot project adapted a smartphone-based travel survey system, Future Mobility Sensing, to collect real-time customer feedback and objective operational measurements on specific bus trips. The system used a combination of GPS, Wi-Fi, Bluetooth, and accelerometer data to track transit trips while soliciting users’ feedback on trip experience. Though not necessarily intended to replace traditional monitoring channels and processes, these data can complement official performance monitoring through a more real-time, customer-centric perspective. The pilot project operated publicly for 3 months on the Silver Line bus rapid transit in Boston, Massachusetts. Seventy-six participants completed the entrance survey; half of them actively participated and completed more than 500 questionnaires while on board either at the end of a trip, at the end of a day, or both. Participation was biased toward frequent Silver Line users, the majority of whom were white and of higher income. Indicative models of user-reported satisfaction reveal some interesting relationships, but the models can be improved by fusing the app-collected data with actual performance characteristics. Broader and more sustained user engagement remains a critical future challenge.

Detecting trip purposes from smartphone-based travel surveys with artificial neural networks and particle swarm optimization

Travel surveys based on global positioning system (GPS) data have exponentially increased over the past decades. Trip characteristics, including trip ends, travel modes, and trip purposes need to be detected from GPS data. Compared with other trip characteristics, studies on trip purpose detection are limited. These studies struggle with three types of limitations, namely, data validation, classification approach-related issues, and result comparison under multiple scenarios. Therefore, we attempt to obtain full understanding and improve these three aspects when detecting trip purposes in the current study. First, a smartphone-based travel survey is employed to collect GPS data, and a surveyor-intervened prompted recall survey is used to validate trip characteristics automatically detected from the GPS data. Second, artificial neural networks combined with particle swarm optimization are used to detect trip purposes from the GPS data. Third, four scenarios are constructed by employing two methods for land-use type coding, i.e., polygon-based information and point of interest, and two methods for selecting training dataset, i.e., equal proportion selection and equal number selection. The accuracy of trip purpose detection is then compared under these scenarios. The highest accuracies of 97.22% for the training dataset and 96.53% for the test dataset are achieved under the scenario of polygon-based information and equal proportion selection by comparing the detected and validated trip purposes. Promising results indicate that a smartphone-based travel survey can complement conventional travel surveys.