An activity scheduling and multi‐agent micro‐simulation platform (ASMMSP) based on long‐term cellular data and considering multi‐mode transfer

Liu B., Li F., Hou Y., Antonio Biancardo S., Ma X.

Understanding the associations between the built environment and road traffic CO2 emissions is crucial for developing strategies to mitigate carbon emissions. However, previous research struggled to capture complex spatial relationships accurately due to classical geospatial models' limitations and the challenges of estimating CO2 emissions from operational vehicle data or limited sample sizes. Therefore, we introduce a novel model that leverages extensive vehicle trajectory data for estimating road traffic CO2 emissions. Furthermore, we develop a geographically convolutional neural network weighted regression (GCNNWR) model to analyze the correlation between the built environment and these emissions. This model employs convolutional neural networks to effectively capture non-linear spatial relationships. An empirical analysis was conducted in Beijing, China, demonstrating the superiority of the GCNNWR model in accommodating spatial heterogeneity compared to conventional geospatial models. Our findings provide critical insights into optimizing the built environment to minimize CO2 emissions.

Guo Y., Yang F., Yan H., Xie S., Liu H., Dai Z.

AbstractCellular data is a sequence of base station‐interaction data that records user ID, timestamp, location area code (LAC), and cell identity (CID). With long observation periods, the data allows traffic planners to analyze coarse‐granularity user travel behaviours at low costs. However, utilizing cellular data for urban planning is not an easy task as the data lacks user socioeconomic attributes due to privacy issues. The data is also challenging to recognize user activity types. This paper proposed an activity‐based model (ABM) with skeleton schedule constraints for multi‐day cellular data. The model first infers the activity pattern and home location. Then it predicts start time, duration, and locations separately for primary and secondary activities. Next, the model infers the travel mode and path considering user multi‐day travel behaviour, path non‐linear coefficient, and transfers. Finally, a time adjustment module is proposed to avoid time conflicts in consecutive activities. The proposed activity‐based model is validated at activity, travel, and path levels. Results show that the proposed model can effectively predict activities and has much higher stability than existing ABMs based on travel surveys.

Wang Y., Yang F., He L., Liu H., Tan L., Wang C.

Cellular signaling data have become increasingly indispensable in analyzing residents’ travel characteristic. Especially with the enhancement of positioning quality in 4G-LTE and 5G wireless communication systems, it is expected that the identification accuracy of fine-grained travel modes will achieve an optimal level. However, due to data privacy issues, the empirical evaluation of the performance of different identification methods is not yet sufficient. This paper builds a travel mode identification model that utilizes the gated recurrent unit (GRU) neural network. With 24 features as input, this method can identify four traffic modes, including walking, bicycle, car, and bus. Moreover, in cooperation with the operator, we organized an experiment collecting cellular signaling data, as well as the corresponding GPS data. Using the collected dataset as ground-truth data, the performance of the method presented in this paper and other popular methods is verified and compared. The results indicate that the GRU-based method has a better performance, with a precision, recall, and F score of 90.5%. Taking F score as an example, the outcome of the GRU-based method is about 6% to 7% higher than methods based on other machine learning algorithms. Considering the identification accuracy and model training time comprehensively, the method suggested in this paper outperforms the other three deep learning-based methods, namely, recurrent neural network (RNN), long short-term memory network (LSTM), and bidirectional long short-term memory network (Bi-LSTM). This study may provide some insights for the application and development of cellular signaling-based travel information collection technology for residents in the future.

Liu X., Cathy Liu X., Liu Z., Shi R., Ma X.

Photovoltaic and energy storage system (PESS) offers a compelling pathway towards boosting green transportation due to its low carbon emissions. This study investigates a solar-powered bus charging infrastructure location problem by considering PESS. A two-stage robust optimization model is formulated to handle the uncertainty of charging service degradation. The first-stage decision is to determine which bus depots are to be upgraded with PESS. The second-stage decision is to conduct emergent bus and energy scheduling when the charging service degrades. Two objectives are optimized simultaneously. The first objective is to maximize the net benefits of PESS during day-to-day operations. The second objective is to minimize the unmet passenger demands during the charging service degradation. We implement a case study using a sub-network of Beijing public transport. The results present pieces of evidence that PESS can lower the daily bus charging costs and improve the service capacity of passengers when the charging service degrades.

Guo Y., Yang F., Xie S., Yao Z.

Zwick F., Kuehnel N., Hörl S.

On-demand ride-pooling systems have gained increasing attention in science and practice in recent years. Simulation studies have shown an enormous potential to reduce fleet sizes and vehicle kilometers traveled if private car trips are replaced with ride-pooling services. However, existing simulation studies assume operation with autonomous vehicles, with no restrictions on operational tasks required when the vehicles are operated by manual drivers. In this article, we simulate and evaluate the operational challenges of non-autonomous ride-pooling systems through driver shifts and breaks and compare their capacity and efficiency to autonomous on-demand services. Based on the existing ride-pooling service MOIA in Hamburg, Germany, we introduce shift and break schedules and implement a new hub return logic to perform the respective tasks at different types of vehicle hubs. This way, currently operating on-demand services are modeled more realistically and the efficiency gains of such services through autonomous vehicles are quantified. The results suggest that operational challenges substantially limit the ride-pooling capacity in terms of served rides with a given number of vehicles. While results largely depend on the chosen shift plan, the presented operational factors should be considered for the assessment of current operational real-world services. The contribution of this study is threefold: From a technical perspective, it is shown that the explicit simulation of operational constraints of current services is crucial to assess ride-pooling services. From a policy perspective, the study shows the operational challenges of a ride-pooling service with non-autonomous vehicles and the potential of future autonomous services. Lastly, the paper adds to the literature a practical ride-pooling simulation use case based on observed real-world demand and shift data.

Yan H., Ma X., Pu Z.

Traffic forecasting has attracted considerable attention due to its importance in proactive urban traffic control and management. Scholars and engineers have exerted considerable efforts in improving the performance of traffic forecasting algorithms in terms of accuracy, reliability, and efficiency. Spatial feature representation of traffic flow is a core component that greatly influences traffic forecasting performance. In previous studies, several spatial attributes of traffic flow are ignored due to the following issues: a) traffic flow propagation does not comply with the road network, b) the spatial pattern of traffic flow varies over time, and c) single adjacent matrix cannot handle the complex and hierarchical urban traffic flow. To address the abovementioned issues, this study proposes a novel traffic forecasting algorithm called traffic transformer, which achieves great success in natural language processing. The multihead attention mechanism and stacking layers enable the transformer to learn dynamic and hierarchical features in sequential data. Two components, namely, global encoder and global–local decoder, are proposed to extract and fuse the spatial patterns globally and locally. Experimental results indicate that the proposed traffic transformer outperforms state-of-the-art methods. The learned dynamic and hierarchical features of traffic flow can help achieve a better understanding of spatial dependency of traffic flow for effective and efficient traffic control and management strategies.

Jiang H., Yang F., Su W., Yao Z., Dai Z.

Existing studies on activity location recognition based on mobile phone data has made great progresses. However, current studies generally assume constant distance threshold when performing activity location clustering, and ignore the influence of base station layout on positioning accuracies of mobile phone data. Given different recognition accuracy requirements, the authors propose two methods to recognise activity locations: (1) An improved hierarchical agglomerative clustering algorithm that integrates a genetic algorithm component to search and dynamically adjust optimal distance thresholds based on base station densities; (2) The recognition method based on Bi-directional long short-term memory network that classifies travel statuses of mobile phone traces. Results show that, compared with existing methods, the activity location recognition accuracy of the proposed hierarchical agglomerative clustering algorithm increases by about 5%. The Bi-directional long short-term memory network model further outperforms the improved hierarchical agglomerative clustering, especially in the aspect of recognising non-commuting activity locations. However, the Bi-directional long short-term memory network model training requires the users’ actual travel information, so there are certain obstacles in popularising Bi-directional long short-term memory network in practice.

Wang L., Yang F., Jin P.J., Zhou T., Guo Y.

Signaling positioning technology provides a new opportunity to understand an individual’s travel characteristics. In recent studies, the travel parameters obtained are mainly macroscopic travel information. However, extracting detailed trip chain information, such as the trip mode and mode-switching time point, remains a challenge. Furthermore, because of the iterative development of wireless networks, existing communication operators usually store different frequencies and accuracy (2G/3G and 4G) of signaling data simultaneously, making the refined identification of travel information more difficult. Therefore, this paper proposes a new method. First, we use the shortest distance algorithm to match the signaling data with the road network. Second, a wavelet transform modulus maximum (WTMM) algorithm is proposed to divide multimodal travel trajectories into single-mode trip segments; thus, spatiotemporal information related to mode transfer can be obtained. Finally, an unsupervised fuzzy kernel c-means clustering (FKCM) algorithm is proposed to distinguish travel modes. As comparison data, smartphone GPS and travel log data are also collected to analyze the detection result and improve the method. The identification errors of mode-switching time points at different frequencies are all less than 360 s. The average correct rate of traffic mode identification for 2G is 65.1%, and the average correct rate of traffic mode identification for 3G is 78.2%. 4G intensive cellular positioning data has a significantly better recognition effect than low-frequency data; the average trip mode detection accuracy reaches 89.6%, and the mode-switching time point detection errors are within 300 s.

Liu X., Qu X., Ma X.

• Electric bus charging infrastructure and vehicle flow are jointly optimized. • Power matching and seasonality effects on bus battery are considered. • A surrogate-based optimization approach is proposed to solve the model. • An empirical study with bus operational data in Beijing is analyzed. In this research, a novel optimization model for electric bus charging station location, charger configuration, charging time and vehicle flow is developed considering power matching and seasonality. The seasonality highlights the effect of air temperature on the battery performances of electric buses. Power matching between batteries and chargers jointly determines the maximum battery acceptance rates of electric buses, and this consideration results in nonlinear constraints. A surrogate-based optimization approach is proposed to solve the mixed integer nonlinear program efficiently. The optimization model is demonstrated on a sub-transit network including 17 bus lines in Beijing. The results reveal significant performance differences regarding vehicle scheduling and charging among different bus fleets in the BEB-based transit system. The interesting findings on the distribution of vehicle flows for charging provide strong evidence to consider powering match in the bus charging infrastructure layout.

Hörl S., Balac M.

Synthetic populations of travelers and their detailed mobility behavior are an important basis for agent-based transport simulations, which are increasingly used in transport planning and research today. To date, research based on such simulations is rarely replicable as it is based on proprietary data and tools. To foster the discussion and steer research towards reproducible transport simulations, this paper introduces a process for generating a synthetic travel demand with individual households, persons, and their daily activity chains for Paris and its surrounding region Île-de-France — entirely based on open data and open software and replicable by any researcher. The resulting travel demand is published for others to use as a comprehensive data basis for agent-based transport simulations and as a test bed for population and demand synthesis algorithms. Furthermore, it is discussed how implicit correlation structures impact the potential use cases of the synthetic travel demand for simulation and analysis purposes and how the common practice of using population samples for downstream simulations affects the results. • Open-source pipeline for population and travel demand synthesis for Île-de-France. • Reproducible approach that utilizes only open data and software. • Modular approach that is transferable to any region in France. • Validation and sampling error analysis of the generated synthetic travel demand. • Discussion of correlations between variables impacting later practical use.

Guo Y., Yang F., Jin P.J., Liu H., Ma S., Yao Z.

Vehicle travel paths provide basic information for improving traffic forecasting models, tracking epidemics transmission, and road construction. Nevertheless, the challenge of recognition and verif...

Yang F., Wang Y., Jin P.J., Li D., Yao Z.

Cellular phone data has been proven to be valuable in the analysis of residents’ travel patterns. Existing studies mostly identify the trip ends through rule-based or clustering algorithms. These methods largely depend on subjective experience and users’ communication behaviors. Moreover, limited by privacy policy, the accuracy of these methods is difficult to assess. In this paper, points of interest data is applied to supplement cellular phone data’s missing information generated by users’ behaviors. Specifically, a random forest model for trip end identification is proposed using multi-dimensional attributes. A field data acquisition test is designed and conducted with communication operators to implement synchronized cellular phone data and real trip information collection. The proposed identification approach is empirically evaluated with real trip information. Results show that the overall trip end detection precision and recall reach 95.2% and 88.7% with an average distance error of 269 m, and the time errors of the trip ends are less than 10 min. Compared with the rule-based approach, clustering algorithm, naive Bayes method, and support vector machine, the proposed method has better performance in accuracy and consistency.

Ziemke D., Knapen L., Nagel K.

MATSim is an agent-based transport simulation model. In contrast to a pure dynamic traffic assignment (DTA) model, MATSim can react to more choice dimensions than route choice, which is the only choice treated by a typical DTA. Simulating the mobility and activity participation of individuals during the whole day, MATSim can additionally represent mode choice, departure time choice, and other decisions and is, therefore, policy-sensitive in terms of these choice dimensions. This allows for the analysis of a wide scope of policies. MATSim can, however, not model the choice of the sequence of activity participation nor the choice of activity participation as such. Also, choices of locations are not typically part of the modeling scope. Interventions into the transport and land-use systems may, however, be substantial such that they can effect changes in behavior in terms of these choices. To allow for the assessment of such reactions, the FEATHERS activity-based demand model is coupled with MATSim. This paper explores different options of integration and describes the development steps of the integration of FEATHERS and MATSim.

Adnan M., Nahmias Biran B., Baburajan V., Basak K., Ben-Akiva M.

Peak and off-peak pricing strategies are an important policy tool used to spread peak demand in public transportation systems. This study uses an agent-based simulator (SimMobility Mid-term) to examine the impact of pricing (off-peak fare discounts) strategies used in Singapore. The aim of the paper is to demonstrate the capabilities of the simulator, and types of detailed performance indicators it can provide, in order to examine the effects of complex public transport pricing policies. Behavioral models within the simulator are calibrated with relevant datasets such as household travel survey, smart card, GPS probe data from taxis and traffic counts for the Singapore network. Nine (09) time-based pricing strategies are examined that consist of a combination of free pre-peak travel on Mass Rapid Transit (MRT) and an off-peak discount for integrated transit (public buses, MRT and Light Rail Transit (LRT)). Changes in public transport ridership, mode shares, operator's revenue and denied boarding are used as indicators to examine the impacts of pricing strategies. The effects of these policies are also examined on segments of the population in terms of income level, person type and gender. Results indicate that off-peak discounts spread PM peak demand and attract individuals to public transportation. However, the availability of fare discounts in all off-peak periods results in adverse impacts during the AM peak because many commuters shift the return leg rather than the initial leg of their journey. The study concludes with suggestions on how to explore more effective pricing strategies, i.e. providing fare discounts only during off-peak periods that surround AM peak.