Search

Incorporating mobile phone data, known for its high spatial and temporal resolution and extensive population coverage, into Activity-Based Models (ABM) for understanding individual travel and activity behaviors is a current research hotspot. However, applying them in ABM building is not straightforward because they miss key information — activity type. In this paper, we present an activity types detection method named the Teacher–Student Cycling model based on a self-training framework. Our model can combine travel survey data with extensive prior knowledge and mobile phone data. We introduce two different resolutions of mobile phone datasets to test the model performance. Our results show that our proposed model can achieve good performance on datasets of different resolutions. Our model, one of the semi-supervised learning models that uses a mixture of labeled and unlabeled knowledge performs better than the supervised learning model that uses labeled knowledge alone. In addition, our model improves the overall detection accuracy by 7% over the second-best semi-supervised learning model and improves the detection accuracy of secondary activities by up to 17%. Our model can be valuable in generating daily activity schedules for agent-based models. • Using a semi-supervised self-learning model to explore the activity types hidden in the mobile phone. • Activity-travel survey is introduced as prior knowledge in the detection model. • The model's performance is evaluated using various resolution data. [ABSTRACT FROM AUTHOR]

Up-to-date information on different modes of travel to monitor transport traffic and evaluate rapid urban transport planning interventions is often lacking. Transport systems typically rely on traditional data sources providing outdated mode-of-travel data due to their data latency, infrequent data collection and high cost. To address this issue, we propose a method that leverages mobile phone data as a cost-effective and rich source of geospatial information to capture current human mobility patterns at unprecedented spatiotemporal resolution. Our approach employs mobile phone application usage traces to infer modes of transportation that are challenging to identify (bikes and ride-hailing/taxi services) based on mobile phone location data. Using data fusion and matrix factorisation techniques, we integrate official data sources (household surveys and census data) with mobile phone application usage data. This integration enables us to reconstruct the official data and create an updated dataset that incorporates insights from digital footprint data from application usage. We illustrate our method using a case study focused on Santiago, Chile successfully inferring four modes of transportation: mass-transit (all public transportation), motorised (cars and similar vehicles), active (pedestrian and cycle trips), and taxi (traditional taxi and ride-hailing services). Our analysis revealed significant changes in transportation patterns between 2012 and 2020. We quantify a reduction in mass-transit usage across municipalities in Santiago, except where metro/rail lines have been more recently introduced, highlighting added resilience to the public transport network of these infrastructure enhancements. Additionally, we evidence an overall increase in motorised transport throughout Santiago, revealing persistent challenges in promoting urban sustainable transportation. Findings also point to a rise in the share of taxi usage, and a drop in active mobility, suggesting a modal shift towards less sustainable modes of travel. We validate our findings comparing our updated estimates with official smart card transaction data. The consistency of findings with expert domain knowledge from the literature and historical transport usage trends further support the robustness of our approach. • Leveraging mobile phone data and application usage traces to infer mode split. • Using data fusion to combine mobile phone data with official data. • Updated mode split includes hard-to-identify modes (e.g., taxi, bikes). • The method was validated against official smart card transaction data. [ABSTRACT FROM AUTHOR]

Highlights • We process two months mobile network trajectories from the Greater Paris region. • The transport mode is inferred from all trajectories, using few labeled data. • Total road and rail OD flows are estimated over time at different resolutions. • The estimates are validated against survey and travel cards flows. Abstract Fast urbanization generates increasing amounts of travel flows, urging the need for efficient transport planning policies. In parallel, mobile phone data have emerged as the largest mobility data source, but are not yet integrated to transport planning models. Currently, transport authorities are lacking a global picture of daily passenger flows on multimodal transport networks. In this work, we propose the first methodology to infer dynamic Origin-Destination flows by transport modes using mobile network data e.g., Call Detail Records. For this study, we pre-process 360 million trajectories for more than 2 million devices from the Greater Paris as our case study region. The model combines mobile network geolocation with transport network geospatial data, travel survey, census and travel card data. The transport modes of mobile network trajectories are identified through a two-steps semi-supervised learning algorithm. The later involves clustering of mobile network areas and Bayesian inference to generate transport probabilities for trajectories. After attributing the mode with highest probability to each trajectory, we construct Origin-Destination matrices by transport mode. Flows are up-scaled to the total population using state-of-the-art expansion factors. The model generates time variant road and rail passenger flows for the complete region. From our results, we observe different mobility patterns for road and rail modes and between Paris and its suburbs. The resulting transport flows are extensively validated against the travel survey and the travel card data for different spatial scales. [ABSTRACT FROM AUTHOR]

New mobility data sources like mobile phone traces have been shown to reveal individuals’ movements in space and time. However, socioeconomic attributes of travellers are missing in those data. Consequently, it is not possible to partition the population and have an in-depth understanding of the socio-demographic factors influencing travel behaviour. Aiming at filling this gap, we use mobile internet usage behaviour, including one’s preferred type of website and application (app) visited through mobile internet as well as the level of usage frequency, as a distinguishing element between different population segments. We compare the travel behaviour of each segment in terms of the preference for types of trip destinations. The point of interest (POI) data are used to cluster grid cells of a city according to the main function of a grid cell, serving as a reference to determine the type of trip destination. The method is tested for the city of Shanghai, China, by using a special mobile phone dataset that includes not only the spatial-temporal traces but also the mobile internet usage behaviour of the same users. We identify statistically significant relationships between a traveller’s favourite category of mobile internet content and more frequent types of trip destinations that he/she visits. For example, compared to others, people whose favourite type of app/website is in the “tourism” category significantly preferred to visit touristy areas. Moreover, users with different levels of internet usage intensity show different preferences for types of destinations as well. We found that people who used mobile internet more intensively were more likely to visit more commercial areas, and people who used it less preferred to have activities in predominantly residential areas. [ABSTRACT FROM AUTHOR]

The practice of estimating origin–destination (OD) demand usually requires large-scale travel surveys. To reduce the cost and time spent on surveys, individual trajectory data obtained from mobile devices has been used as an alternative dataset since the last two decades for OD estimation but also constrained in practice in some countries. To estimate OD matrices while protecting privacy, this study uses aggregated data of mobile phone traces to estimate work-related trips. The proposed approach is a sequential updater based on the maximum entropy principle. Trip production and attraction are firstly calculated by a non-linear programming problem followed by a matrix fitting problem to distribute trips to each OD pair. Numerical study shows that updated values are much closer to the synthesize real values than the referred ones. The case study in Tokyo further demonstrates that the proposed updating approach can track the change of travel pattern. [ABSTRACT FROM AUTHOR]

Rapid urbanization is placing increasing stress on already burdened transportation infrastructure. Ubiquitous mobile computing and the massive data it generates presents new opportunities to measure the demand for this infrastructure, diagnose problems, and plan for the future. However, before these benefits can be realized, methods and models must be updated to integrate these new data sources into existing urban and transportation planning frameworks for estimating travel demand and infrastructure usage. While recent work has made great progress extracting valid and useful measurements from new data resources, few present end-to-end solutions that transform and integrate raw, massive data into estimates of travel demand and infrastructure performance. Here we present a flexible, modular, and computationally efficient software system to fill this gap. Our system estimates multiple aspects of travel demand using call detail records (CDRs) from mobile phones in conjunction with open- and crowdsourced geospatial data, census records, and surveys. We bring together numerous existing and new algorithms to generate representative origin–destination matrices, route trips through road networks constructed using open and crowd-sourced data repositories, and perform analytics on the system’s output. We also present an online, interactive visualization platform to communicate these results to researchers, policy makers, and the public. We demonstrate the flexibility of this system by performing analyses on multiple cities around the globe. We hope this work will serve as unified and comprehensive guide to integrating new big data resources into customary transportation demand modeling. [ABSTRACT FROM AUTHOR]

In this work, we present methods to estimate average daily origin–destination trips from triangulated mobile phone records of millions of anonymized users. These records are first converted into clustered locations at which users engage in activities for an observed duration. These locations are inferred to be home , work , or other depending on observation frequency, day of week, and time of day, and represent a user’s origins and destinations. Since the arrival time and duration at these locations reflect the observed (based on phone usage) rather than true arrival time and duration of a user, we probabilistically infer departure time using survey data on trips in major US cities. Trips are then constructed for each user between two consecutive observations in a day. These trips are multiplied by expansion factors based on the population of a user’s home Census Tract and divided by the number of days on which we observed the user, distilling average daily trips. Aggregating individuals’ daily trips by Census Tract pair, hour of the day, and trip purpose results in trip matrices that form the basis for much of the analysis and modeling that inform transportation planning and investments. The applicability of the proposed methodology is supported by validation against the temporal and spatial distributions of trips reported in local and national surveys. [ABSTRACT FROM AUTHOR]

Highlights: [•] We examine monthly human spatial behaviour over a 12-month period. [•] We use mobile phone call detail records as a novel data source. [•] Our proposed approach is able to capture longitudinal human spatial behaviour. [•] We provide a set of new insights on monthly variances in human spatial behaviour. [ABSTRACT FROM AUTHOR]

Call detail records (CDR) collected by mobile phone network providers have been largely used to model and analyze human-centric mobility. Despite their potential, they are limited in terms of both spatial and temporal accuracy thus being unable to capture detailed human mobility information. Network Signaling Data (NSD) represent a much richer source of spatio-temporal information currently collected by network providers, but mostly unexploited for fine-grained reconstruction of human-centric trajectories. In this paper, we present TRANSIT, TRAjectory inference from Network SIgnaling daTa, a novel framework capable of processing NSD to accurately distinguish mobility phases from stationary activities for individual mobile devices, and reconstruct, at scale, fine-grained human mobility trajectories, by exploiting, with a DBSCAN-based clustering approach, the inherent recurrence of human mobility and the higher sampling rate of NSD. The validation on a ground-truth dataset of GPS trajectories showcases the superior performance of TRANSIT (80% precision and 96% recall) with respect to state-of-the-art solutions in the identification of movement periods, as well as an average 190 m spatial accuracy in the estimation of the trajectories. We also leverage TRANSIT to process a unique large-scale NSD dataset of more than 10 millions of individuals and perform an exploratory analysis of city-wide transport mode shares, recurrent commuting paths, urban attractivity and analysis of mobility flows. • Our study addresses locational uncertainty and temporal sparsity using mobile phone network signaling data. • We provide the TRANSIT (TRAjectory inference from Network Signaling daTa) framework for processing network signaling data, able to accurately tell apart movement intervals from stationary activity periods for each mobile device and infer fine-grained human mobility trajectories during the associated movement intervals. • We validate TRANSIT with ground truth GPS data and show that our framework exceeds the performance of the existing methods. • We implement and discuss large-scale human-mobility applications of TRANSIT. [ABSTRACT FROM AUTHOR]

• We propose a framework for estimating traffic speeds from temporally-biased trips. • We propose a temporal bias model linked to user-activity-dependent positioning data. • We apply the methodology to a set of GPS trips downsampled according to this model. • We investigate the effect of successive data degradation on the results. • We investigate the effect of statistical representativity on the results. Over the past few decades, the digitalization of services and infrastructures has led to the emergence of a broad set of new information sources to characterize human mobility. These sources usually offer valuable significant population penetration rates but may also suffer from important temporal sparsity. Data generated by user activity, such as social networks or mobile phone data, especially fit this description. Although this temporal sparsity might prevent estimating individual travel speeds, we state that such low-frequency positioning data enable estimating the average urban traffic speed dynamics when considering an adequate network partitioning. In this sense, this article proposes a new method, based on the division of the urban area of a given city into regions and on the analysis of a limited set of basic characteristics of individual vehicle trips, such as the regional path. Our solution first involves estimating robust travel times from travelers sharing similar trip features and then jointly analyzing these travel times to deduce the underlying regional traffic speeds, using regression analysis. We apply this methodology on a set of trips derived from a large GPS dataset of vehicle tracks covering the city of Lyon. These data are purposely downsampled to reduce the sampling rate and reproduce bias and temporal features that are proper to sparser but larger-scale, mobility data sources dependent on user's communication activities. Controlling the data downsampling process allows us to evaluate the impacts of the progressive information loss on the speed estimation, while the raw GPS data provide the ground truth speed reference against which to compare our results. Provided that the amount of observed individual trips is sufficient, the analysis returns satisfying speed estimation results, both at low and high downsampling levels. Thus, we successfully demonstrate that it is possible to estimate zonal traffic speeds from degraded trip data without evaluating individual travel speeds. [ABSTRACT FROM AUTHOR]

• Privacy regulations avoid mobile companies to share users geocoded data. • Mobile phone data is useful for transport planning simulations. • Framework to synthesise realistic individual travel demand. • Input restricted to user-aggregated mobile phone. • Extensive validation of synthetic data results vs real one. Mobile phone data generated in mobile communication networks has the potential to improve current travel demand models and in general, how we plan for better urban transportation systems. However, due to its high-dimensionality, even if anonymised there still exists the possibility to re-identify the users behind the mobile phone traces. This risk makes its usage outside the telecommunication network incompatible with recent data privacy regulations, hampering its adoption in transportation-related applications. To address this issue, we propose a framework designed only with user-aggregated mobile phone data to synthesise realistic daily individual mobility — Digital Twin Travellers. We explore different strategies built around modified Markov models and an adaption of the Rejection Sampling algorithm to recreate realistic daily schedules and locations. We also define a one-day mobility population score to measure the similarity between the population of generated agents and the real mobile phone user population. Ultimately, we show how with a series of histograms provided by the telecommunication service provider (TSP) it is possible and plausible to disaggregate them into new synthetic and useful individual-level information, building in this way a big data travel demand framework that is designed in accordance with current data privacy regulations. [ABSTRACT FROM AUTHOR]

• Estimation of all parameters required for MFD-based simulation using large scale mobile phone data. • Estimation of time dependent OD-specific and origin-specific penetration rates. • Proposal of well-defined correlation between the detour ratio and the mean speeds. • Empirical estimation of path-flow coeffcients and UE gaps. The present work proposes a global framework to estimate all MFD model parameters using mobile phone data. The three major components that are estimated in the present context are MFD shapes, regional trip lengths and path flow distribution. A trip enrichment scheme based on the map matching process is proposed for the trips that have sparser records. Time dependent penetration rates are estimated by fusing the OD matrix and the Loop Detector Data (LDD). Two different types of penetration rates of vehicles are proposed based on the OD flow and the trips starting within an origin, respectively. The estimated MFDs based on two types of penetration rates are stable with very low scatter. In the following step, macro-paths and their corresponding trip lengths are estimated. This work is the first to present empirical evidences of the dynamic evolution of mean trip lengths over the day, which is very difficult to capture with other types of data sources. The last component is the time dependent path flow distributions between the different macro-paths for a given OD pair. The manuscript is concluded by presenting the time evolution of the User Equilibrium (UE) gap for different macroscopic OD pairs. It is noticed that UE principle holds true most of the time, except for OD pairs that have macro-paths transversing through congested reservoirs, especially during peak hours. [ABSTRACT FROM AUTHOR]

• First definition of mobile phone "Track & Trace" (T&T) personal mobility data. • Limitations of conventional data and models within transportation are summarised. • Mobility content and individual context ceonceptualise New and Emerging Data Forms. • The unique potential (and challenges) of GPS-based T&T data are characterised. • We propose that T&T data can be integrated into existing planning and policy methods. High quality, reliable data and robust models are central to the development and appraisal of transportation planning and policy. Although conventional data may offer good 'content', it is widely observed that it lacks context i.e. who and why people are travelling. Transportation modelling has developed within these boundaries, with implications for the planning, design and management of transportation systems and policy-making. This paper establishes the potential of passively collected GPS-based "Track & Trace" (T&T) datasets of individual mobility profiles towards enhancing transportation modelling and policy-making. T&T is a type of New and Emerging Data Form (NEDF), lying within the broader 'Big Data' paradigm, and is typically collected using mobile phone sensors and related technologies. These capture highly grained mobility content and can be linked to the phone owner/user behavioural choices and other individual context. Our meta-analysis of existing literature related to spatio-temporal mobile phone data demonstrates that NEDF's, and in particular T&T data, have had little mention to date within an applied transportation planning and policy context. We thus establish there is an opportunity for policy-makers, transportation modellers, researchers and a wide range of stakeholders to collaborate in developing new analytic approaches, revise existing models and build the skills and related capacity needed to lever greatest value from the data, as well as to adopt new business models that could revolutionise citizen participation in policy-making. This is of particular importance due to the growing awareness in many countries for a need to develop and monitor efficient cross-sectoral policies to deliver sustainable communities. [ABSTRACT FROM AUTHOR]

• Planning bike trips with novel data sources. • Using percolation theory to optimize cost and maximize global connectivity. • Network Science for detecting affinity of trips by income level and further inform local interventions. Cities around the world are turning to non-motorized transport alternatives to help solve congestion and pollution issues. This paradigm shift demands on new infrastructure that serves and boosts local cycling rates. This creates the need for novel data sources, tools, and methods that allow us to identify and prioritize locations where to intervene via properly planned cycling infrastructure. Here, we define potential demand as the total trips of the population that could be supported by bicycle paths. To that end, we use information from a phone-based travel demand and the trip distance distribution from bike apps. Next, we use percolation theory to prioritize paths with high potential demand that benefit overall connectivity if a bike path would be added. We use Bogotá as a case study to demonstrate our methods. The result is a data science framework that informs interventions and improvements to an urban cycling infrastructure. [ABSTRACT FROM AUTHOR]