Search

Spatiotemporal data, and more specifically origin-destination matrices, are critical inputs to mobility studies for transportation planning and urban management purposes. Traditionally, high-cost and hard-to-update household travel surveys are used to produce large-scale origin-destination flow information of individuals' whereabouts. In this paper, we propose a methodology to estimate origin-destination (O-D) matrices based on passively-collected cellular network signalling data of millions of anonymous mobile phone users in the Rhone-Alpes region, France. Unlike Call Detail Record (CDR) data which rely only on phone usage, signalling data include all network-based records providing higher spatiotemporal granularity. The explored dataset, which consists of time-stamped traces from 2G and 3G cellular networks with users' unique identifier and cell tower locations, is used to first analyse the cell phone activity degree indicators of each user in order to qualify the mobility information involved in these records. These indicators serve as filtering criteria to identify users whose device transactions are sufficiently distributed over the analysed period to allow studying their mobility. Trips are then extracted from the spatiotemporal traces of users for whom the home location could be detected. Trips have been derived based on a minimum stationary time assumption that enables to determine activity (stop) zones for each user. As a large, but still partial, fraction of the population is observed, scaling is required to obtain an O-D matrix for the full population. We propose a method to perform this scaling and we show that signalling data-based O-D matrix carries similar estimations as those that can be obtained via travel surveys.

Correctly identifying the home location is crucial for human mobility analysis with telecom data, more specifically call detail record (CDR) data. To that end, multiple heuristics have been developed in literature. Nevertheless, due to the lack of ground truth home location data, no study has thoroughly validated these widely used methods so far. We present a detailed performance analysis of existing home detection heuristics, using a unique dataset that enables this important validation on the lowest level, being the level of the cell tower. Our research indicates that simple heuristics surprisingly outperform their more complex counterparts. The benchmark study revealed that the best heuristic is able to identify the home location with an average error of approximately 4.5 km and selects the correct home tower in 60.69% of the cases. Based on the insights provided by our study, we propose a new heuristic that increases the accuracy to 61% and lowers the average distance error to 4.365 km. Secondly, if the home location is known for possibly only a fraction of the instances, we propose a labelled predictive modelling approach. Adding social network based variables in this predictive model further enhances the predictive performance. Our best model reduces the average distance error to 2.848 km and selects the correct home location in 72.08% of the cases. Furthermore, this result provides an indication of the upper bound for home detection with CDR data. Finally, models that only make use of social network based data are developed as well. Results show that even without using data of the focal individual, these models are able to select the correct home tower in 37.65% of the cases and achieve an average distance error of 8.1 km.