Forecasting short-term ridership of different origin-destination pairs (i.e., OD matrix) is crucial to the real-time operation of a metro system. However, this problem is notoriously difficult due to the large-scale, high-dimensional, noisy, and highly skewed nature of OD matrices. In this paper, we address the short-term OD matrix forecasting problem by estimating a low-rank high-order vector autoregression (VAR) model. We reconstruct this problem as a data-driven reduced-order regression model and estimate it using dynamic mode decomposition (DMD). The VAR coefficients estimated by DMD are the best-fit (in terms of Frobenius norm) linear operator for the rank-reduced full-size data. To address the practical issue that metro OD matrices cannot be observed in real time, we use the boarding demand to replace the unavailable OD matrices. Moreover, we consider the time-evolving feature of metro systems and improve the forecast by exponentially reducing the weights for historical data. A tailored online update algorithm is then developed for the high-order weighted DMD model (HW-DMD) to update the model coefficients at a daily level, without storing historical data or retraining. Experiments on data from two large-scale metro systems show that the proposed HW-DMD is robust to noisy and sparse data, and significantly outperforms baseline models in forecasting both OD matrices and boarding flow. The online update algorithm also shows consistent accuracy over a long time, allowing us to maintain an HW-DMD model at much low costs.

Smartcard automated fare collection systems (SCAFC) for transit have been considered primarily for their administrative function of controlling access to the service and for revenue management. However, it is likely that data from these systems also can be used to describe both transport supply and demand. This article illustrates the use of smartcard data to estimate various transit performance measures. Combined with well-established evaluation processes, such measures can help operators monitor their networks in greater detail. The performance of the network supply (vehicle-kilometers, vehicle-hours, commercial speed, etc.) and the statistics on passenger service (passenger-kilometers, passenger-hours, average trip length, etc.) can be calculated from these datasets for any spatial or temporal level of resolution, including route and bus stop levels.

While many cities around the world qualify themselves as “smart cities”, there is no comprehensive way to evaluate to what extent they are “smart”. This article proposes a framework for comparison of the level of “smartness” of the urban mobility systems. The most relevant indicators that have the greatest impact on smart mobility systems were selected in the course of literature review. The impact of indicators on smart mobility systems is variable. Evaluating smart mobility systems, different authors distinguish between different indicators, which usually do not duplicate. The paper categorizes the indicators of the smart mobility system into five groups, called “factors”: motor travel and congestion reduction measures; pollution reduction measures; travel safety and accident reduction measures; traffic management tools and services; smart infrastructure measures. A number of indicators are attributed to each of the listed groups. A Multiple Criteria Decision-Making method, namely, the Analytic Hierarchy Process (AHP) method, has been used to evaluate the significance of the smartness level used in the research. This method bases the weighting of subjective criteria on expert judgement. Rank correlation is used to determine the consistency of expert opinions. A model has been developed to compare smart mobility systems of individual cities and their infrastructure.

In many cities, worldwide public transit companies use smart card system to manage fare collection. Analysis of this acquisitive information provides a comprehensive insight of user's influence in the interactive public transit network. In this regard, analysis of temporal data, describing the time of entering to the public transit network is considered as the most substantial component of the data gathered from the smart cards. Classical distance-based techniques are not always suitable to analyze this time series data. A novel projection with intuitive visual map from higher dimension into a three-dimensional clock-like space is suggested to reveal the underlying temporal pattern of public transit users. This projection retains the temporal distance between any arbitrary pair of time-stamped data with meaningful visualization. Consequently, this information is fed into a hierarchical clustering algorithm as a method of data segmentation to discover the pattern of users.

Car sharing systems are gaining new members every month. However, few researches are conducted to better understand how these systems are used. In this paper, typical patterns of use of the car sharing system are identified using a transaction database covering a full year of operation. Data mining techniques are used to classify users according to their temporal patterns of car use frequency, traveled distance, and week use variability. The experiments reveal various classes of users. With respect to number of transactions throughout the year, users are segmented in two large classes: the regular and occasional ones, the majority of users belonging to the latter. The study of average trip length leads to the identification of 5 clusters of users. Finally, 8 types of typical weeks of use are described. Information about users' patterns could help the car sharing managers to optimize the use of the cars. It can also assist users in selecting the most advantageous subscription offer.