Search

Ride-sourcing services have become an integral part of urban transportation systems. In a cooperative relationship with the platform, ride-sourcing drivers have the flexibility to work according to their habits or preferences, which could constitute heterogeneous working pattern(s) among drivers. Based on real-world ride-sourcing data, we analyze drivers' working pattern(s) via their work slices (WSs). A WS is defined as a time window during which a driver is continuously online to take ride-sourcing orders. We use the Gaussian mixture model to classify the WSs according to the start time/start place attributes and obtain 9 clusters with significant spatiotemporal differences. It is found that WSs starting from rural areas that are closed to urban areas show no significant spatiotemporal grouping and mediocre performance in terms of fare income and fare-cost ratio, while WSs starting from distant rural areas show the worst performance; WSs starting from urban/suburban areas reflect more diverse spatiotemporal groupings and significant differences in attributes of delivery time and fares earned. We also analyze drivers' working pattern(s) by summarizing their habitual WS clusters. Findings include that drivers with suburban-dominant/rural-dominant patterns have a disadvantage in working efficiency in terms of the shorter order distance and longer cruising time. If geometrical distance allows, they can improve their income by switching to urban WS clusters occasionally. However, frequently switching between WS clusters can lead to a decrease in fares income. According to our results, the platform can introduce differentiated policies, including subsidies and order-matching algorithms, to better handle the spatiotemporal inequality, ultimately improving the service performance and overall benefits. • We analyze ride-sourcing drivers' working pattern(s) via their work slices (WSs) based on real-world data. • We use the Gaussian mixture model to classify the WSs according to their attributes and obtain 9 clusters. • WSs starting from distant rural areas show lowest earnings, but WSs from urban/suburban areas reflect higher earnings. • Drivers with suburban/rural working patterns have a disadvantage in efficiency with short order distance and long cruise. [ABSTRACT FROM AUTHOR]

• Propose a Bayesian supervised learning tensor factorization model for the analysis of travel choice. • The BSTF model induces a hierarchical probabilistic structure between the predictor variables and the choice. • The BSTF model offers good performance in both classification accuracy and the interpretability of choice behavior. • Apply the BSTF model to analyze passengers' choices between diverse service options on a ride-sourcing platform. In the era of transportation big data, the analysis of mobility patterns generally involves large quantities of datasets with high-dimensional variables recording individual travelers' activities and socio-economic attributes, bringing new challenges to researchers. Conventional regression-based models commonly require complex structures in depicting random or fixed effects with a considerable number of parameters to estimate, and state-of-the-art machine learning models are regarded as black-boxes that are not clear in interpreting the mechanism in human mobility. To overcome the challenges of capturing complex high-order relationships among variables of interest, this paper proposes a Bayesian supervised learning tensor factorization (BSTF) model for the classification of travel choices in the mobility pattern analysis. The BSTF model induces a hierarchical probabilistic structure between predictor variables and the dependent variable, which offers a nature supervised learning foundation via Bayesian inference. Latent class (LC) variables are considered in the BSTF model to discover hidden preferences/states among travelers associated with their mobility patterns. We apply the BSTF model to analyze passenger-side choice patterns between diverse service options on a ride-sourcing platform, drawing increasing attention during recent years. A case study with a real-world dynamic ridesharing dataset in Hangzhou, China, is conducted. Different cases of training sizes are utilized to fit the proposed BSTF model as well as some other state-of-the-art machine learning models. By identifying significant variables and derive their probabilistic relationship between service types (i.e., ridesharing, non-sharing, and taxi), the proposed BSTF model offers good performance in both classification accuracy and the interpretability of shared mobility. [ABSTRACT FROM AUTHOR]

• Model multi-homing ride-sourcing drivers' decision sequence on platform switching. • Develop dynamic decision model based on high-order hidden Markov model (HO-HMM). • Present the merits of HO-HMM in exploring individual characteristics and preferences. • Reveal interdependencies of hidden platform preferences of multi-homing ride-sourcing drivers. • Verify superiority of HO-HMMs by city-wide real-wrold multiple platforms data. Ride-sourcing drivers enjoy flexibility in scheduling work hours and choosing platforms, which generates multi-homing behavior on multiple platforms. It is challenging to observe the labor supply of multi-homing ride-sourcing drivers due to data limitations, which motivates our research on modeling drivers' dynamic decisions on labor supply in the competitive ride-sourcing market with multiple platforms. We propose a dynamic discrete choice framework by modeling drivers' high-frequency decision sequences on platform switching as high-order hidden Markov processes and considering time-varying factors. The high-order hidden Markov framework relaxes the first-order Markovian assumption and builds interdependencies of unobserved states across multiple time periods. The estimation of the proposed model takes advantage of city-scale multiple platforms datasets in Hangzhou, China, including more than 16 million records of the order payment information and more than 46 thousand records of active ride-sourcing drivers' information. The case study results indicate that the high-order hidden Markov model (HO-HMM) has superior explanatory power in fitting multiple platforms datasets than the multinomial logit model and the first-order hidden Markov model. HO-HMM performs an advantage in modeling the extended historical dependency of drivers' decisions and individual driver behavior modeling with high interpretability. The results uncover the variations of drivers' attitudes in different hidden (unobservable) states and towards different platforms. In general, ride-sourcing drivers respond actively and positively to income and working time. The findings support the platforms' decision-making on pricing, reward, and personalized management of ride-sourcing drivers, and provide beneficial suggestions for improving income in the competitive ride-sourcing market. [ABSTRACT FROM AUTHOR]

• Propose an interpretable and high-accuracy model for ride-sourcing demand forecasting. • Propose a systematic process to identify heterogeneous predictability of different regions. • Reveal heteroscedasticity of demand uncertainty across regions and categorize region types. • Exploit correlation between local supply and demand with bidirectional attention mechanism. • Verify outperformance and interpretability of Bi-STTNP by large-scale real-world data. Understanding the spatial–temporal stochasticity in shared mobility is crucial for ride-sourcing demand forecasting, supply–demand management, and vehicle dispatch optimization. In contrast to conventional deep learning methods that typically provide point predictions or deterministic predictions, this paper introduces the bidirectional spatial–temporal Transformer neural processes (Bi-STTNP) prediction model, which stands out from conventional deep learning methods by providing probabilistic predictions and uncertainty estimations for ride-sourcing demand. Bi-STTNP captures the multivariate spatial–temporal Gaussian distribution of demand, offering not only demand expectations but also comprehensive uncertainty representations. We propose a predictability identification process based on predictive distributions to assess varying predictability across time slots and regions, improving interpretability. Our model, consisting of the bidirectional supply–demand attention module and spatial–temporal Transformer module, maintains interpretability while ensuring accurate demand expectation predictions. Extensive experiments on a real-world dataset of 15 million ride-sourcing orders in Hangzhou, China, demonstrate that Bi-STTNP outperforms baseline models in predicting demand expectation and quantifying demand uncertainty. Furthermore, we compute loose spatial–temporal predictability lower bounds and categorize regions by predictability, providing insights for optimizing passenger pricing strategies, driver incentives, and vehicle dispatching in ride-sourcing platforms. [ABSTRACT FROM AUTHOR]

With the rapid development of electric vehicles, how to improve the charging efficiency of electric vehicles has become a challenge. The Chinese government has made great efforts to build charging piles. At present, the most popular construction mode is to build charging piles on a fixed proportion of spaces in existing parking lots. The proportions of charging piles recommended by the government, which is known as a one-size-fits-all strategy. Challenges have arisen with the growth of electric vehicles, such as how to keep hot charging piles from overheating. This paper mainly simulates the actual demand and optimizes the configuration of charging piles to reduce the uneven spatial distribution of charging demand, to improve the overall utilization efficiency of regional charging stations. In particular, based on agent-based simulation technology, a multi-agent system of the road network, vehicles, and charging stations is constructed to simulate the charging behavior in the mixed scenario of electric vehicles and traditional fuel vehicles. The optimization model aims to design the configuration of charging piles to minimize the sum of electric vehicle queueing time, gasoline vehicle queueing time, and vehicle transfer time to idle parking lots. The model is solved by the genetic algorithm. This paper takes the Wulin Square business district in Hangzhou as a real-world example. The simulation results show that by optimizing the number of charging piles, the objective function is reduced by 17.1% compared with the initial number of charging piles, which effectively improves the operation efficiency of the parking system. To this end, this paper predicts the electric vehicle charging demand in the city and optimizes the configuration of charging facilities, which is conducive to maximizing social benefits in the transportation system and has good practical engineering application value. • ABM is established to present individual charging and parking behavior. • The process of cruising for available charging facilities or spaces is described. • The fixed proportion of setting charging piles on spaces can be optimized. • Optimizing proportion configurations can improve system travel efficiency. [ABSTRACT FROM AUTHOR]

• Propose a bottom-up CO 2 emissions calculation approach based on sparse individual trip chain data; • Propose a novel multimodal individual trip chain identification and reconstruction method; • Propose two city-scale trip information expansion methods based on population and checkpoint data; • Develop a multi-scale high-resolution transportation carbon emission calculation platform; • Explore potential implementations of carbon reduction policies in the transportation sector. Passenger transportation is one of the primary sources of urban carbon emissions. Travel data acquisition and appropriate emission inventory availability make estimating high-resolution urban passenger transportation carbon emissions challenging. This paper aims to establish a method to estimate and analyze urban passenger transportation carbon emissions based on sparse trip trajectory data. First, a trip chain identification and reconstruction method is proposed to extract travelers' trip information from sparse trip trajectory data. Meanwhile, a city-scale trip sampling expansion method based on population and checkpoint data is proposed to estimate population movements. Second, the identified trip information (e.g., trip origin and destination, and travel modes) is used to calculate multimodal passenger transportation CO 2 emissions based on a bottom-up CO 2 emissions calculation approach. Third, we develop a multi-scale high-resolution transportation carbon emission calculation and monitoring platform and take the city of Hangzhou, one of China's leading cities, as our case study, with around 10 million daily trips data and a quarter million road links. Five modes of passenger transportation are identified, i.e., walking, cycling, buses, metro, and cars. Hourly carbon emissions are calculated and attributed to corresponding road links, which build up passenger transportation carbon emissions from road links to region and city levels. Results show that a typical working day's total passenger transportation CO 2 emission is about 36,435 tonnes, equivalent to CO 2 emissions from 4 million gallons of gasoline consumed. According to our analysis of the carbon emissions produced by approximately 40,000 km of roadways, urban expressways have the most hourly carbon emissions at 194 kg/(h·km). Moreover, potential applications of the developed methods and platform linking to smart mobility management (e.g., Mobility as a Service, MaaS) and how to work in tandem to support green transportation policies (e.g., green travel rewards and carbon credits in transportation) have been discussed. [ABSTRACT FROM AUTHOR]