Modeling taxi services with smartphone-based e-hailing applications

He, Fang; Shen, Zuo‐Jun Max

doi:10.1016/j.trc.2015.06.023

Cited by 155 publications

(69 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The taxi operator might coordinate supply and demand via the on-demand service platform and thus influence the equilibrium state by regulating the entry of taxi and determining the taxi fare structure, such as non-linear pricing (Yang et al, 2010a). Apart from the traditional taxi market, some emerging market structures, like the ride-sourcing market (Zha et al, 2016), e-hailing taxi market (He and Shen, 2015;Wang et al, 2016b), were examined under the same equilibrium modeling framework. Recently, the on-demand ride-sharing market and the optimal assignment strategies have also attracted researchers' attentions (Alonso-Mora et al, 2017).…”

Section: Literature Reviewmentioning

confidence: 99%

Short-term forecasting of passenger demand under on-demand ride services: A spatio-temporal deep learning approach

Zheng

Chen

2017

Transportation Research Part C: Emerging Technologies

578

261

View full text Add to dashboard Cite

Short-term passenger demand forecasting is of great importance to the ondemand ride service platform, which can incentivize vacant cars moving from over-supply regions to over-demand regions. The spatial dependences, temporal dependences, and exogenous dependences need to be considered simultaneously, however, which makes short-term passenger demand forecasting challenging. We propose a novel deep learning (DL) approach, named the fusion convolutional long short-term memory network (FCL-Net), to address these three dependences within one end-to-end learning architecture. The model is stacked and fused by multiple convolutional long short-term memory (LSTM) layers, standard LSTM layers, and convolutional layers. The fusion of convolutional techniques and the LSTM network enables the proposed DL approach to better capture the spatiotemporal characteristics and correlations of explanatory variables. A tailored spatially aggregated random forest is employed to rank the importance of the explanatory variables. The ranking is then used for feature selection. The proposed DL approach is applied to the short-term forecasting of passenger demand under an on-demand ride service platform in Hangzhou, China. Experimental results, validated on real-world data provided by DiDi Chuxing, show that the FCL-Net achieves better predictive performance than traditional approaches in- * cluding both classical time-series prediction models and neural network based algorithms (e.g., artificial neural network and LSTM). Furthermore, the consideration of exogenous variables in addition to passenger demand itself, such as the travel time rate, time-of-day, day-of-week, and weather conditions, is proven to be promising, since it reduces the root mean squared error (RMSE) by 50.9%. It is also interesting to find that the feature selection reduces 30% in the dimension of predictors and leads to only 0.6% loss in the forecasting accuracy measured by RMSE in the proposed model. This paper is one of the first DL studies to forecast the short-term passenger demand of an on-demand ride service platform by examining the spatio-temporal correlations.Keywords: On-demand ride services, short-term demand forecasting, deep learning (DL), fusion convolutional long short-term memory network (FCL-Net), long short-term memory (LSTM), convolutional neural network (CNN)

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Short-term forecasting of passenger demand under on-demand ride services: A spatio-temporal deep learning approach

Zheng

Chen

2017

Transportation Research Part C: Emerging Technologies

578

261

View full text Add to dashboard Cite

show abstract

“…In other words, an idle driver's searching process ends only when this driver sees a passenger and the passenger accepts the ride (see Figure (1a)). E-hailing applications (such as Didi and Uber), on the other hand, offer an online platform to match a driver with a passenger even when they are not present in the same space at the same time (He and Shen, 2015;Qian and Ukkusuri, 2017). In other words, even when an idle driver sees a passenger waiting on the roadside, as long as the e-hailing platform does not match them, the driver cannot give a ride to the passenger.…”

Section: Motivationmentioning

confidence: 99%

Optimal passenger-seeking policies on E-hailing platforms using Markov decision process and imitation learning

Shou

Ye³

et al. 2020

Transportation Research Part C: Emerging Technologies

View full text Add to dashboard Cite

Vacant taxi drivers' passenger seeking process in a road network generates additional vehicle miles traveled, adding congestion and pollution into the road network and the environment. This paper aims to employ a Markov Decision Process (MDP) to model idle e-hailing drivers' optimal sequential decisions in passenger-seeking. Transportation network companies (TNC) or e-hailing (e.g., Didi, Uber) drivers exhibit different behaviors from traditional taxi drivers because e-hailing drivers do not need to actually search for passengers. Instead, they reposition themselves so that the matching platform can match a passenger. Accordingly, we incorporate e-hailing drivers' new features into our MDP model. The reward function used in the MDP model is uncovered by leveraging an inverse reinforcement learning technique. We then use 44,160 Didi drivers' 3-day trajectories to train the model. To validate the effectiveness of the model, a Monte Carlo simulation is conducted to simulate the performance of drivers under the guidance of the optimal policy, which is then compared with the performance of drivers following one baseline heuristic, namely, the local hotspot strategy. The results show that our model is able to achieve a 17.5% improvement over the local hotspot strategy in terms of the rate of return. The proposed MDP model captures the supply-demand ratio considering the fact that the number of drivers in this study is sufficiently large and thus the number of unmatched orders is assumed to be negligible. To better incorporate the competition among multiple drivers into the model, we have also devised and calibrated a dynamic adjustment strategy of the order matching probability.

show abstract

“…Their conclusion was that the fleet size and pricing policy are closely associated with the level of competition in the market and can have significant impacts on the total passenger cost, average waiting time, and fleet utilization. He and Shen [18] proposed a spatial equilibrium model for balancing the supply and demand of taxi services and depicted the possible adoption of e-hailing applications for drivers and customers in a well-regulated taxi market. Garza et al [19] analyzed two different government support programs within a dynamic equilibrium model of imperfectly competitive industries in order to investigate their effectiveness.…”

Section: E-hailing Services Of Taxicabsmentioning

confidence: 99%

Understanding the Effect of an E-Hailing App Subsidy War on Taxicab Operation Zones

Fang

Lian³

2018

Journal of Advanced Transportation

View full text Add to dashboard Cite

Understanding taxicab operation behaviors under various management or market policies (i.e., subsidies) is critical to making informed operating decisions for e-hailing companies and for government surveillance. This paper investigates the change of taxicab operation zones in context of an e-hailing app subsidy war in China, which is an important perspective that reflects changes in taxicab behavior, such as how the operation zones of taxicabs under the e-hailing app subsidy war change and how this change affects their trip distance and cruising time. To investigate this issue, this paper utilizes three indexes to elucidate the change of taxicab operation zones, namely, the repetition ratio of operation zone pairs, the area, and the degree of dispersion in the spatial distribution. A case study using taxicab trajectories during all of the important periods of the e-hailing app subsidy war in Shenzhen, China, was conducted and produced several valuable findings; for example, with respect to taxicabs as a whole, the proportion of habitual operation zone pairs among operation zone pairs in neighboring periods is relatively stable under any subsidy policy, and changes in the operation zones have little effect on changes in the average daily trip distance and average daily cruising time. Four groups of taxicabs divided according to initial change patterns in the operation zones present different change patterns during the subsidy war. By comparing these changes before and after the subsidy war, this paper finds that the subsidy war influences the taxicabs in groups I and II, while it has little influence on the taxicabs in groups III and IV, although all groups were affected during the subsidy war. For the taxicab groups in the period with the highest subsidy, the average daily trip distance and average daily cruising time decreased, whereas, in other periods, they presented different patterns.

show abstract

Modeling taxi services with smartphone-based e-hailing applications

Cited by 155 publications

References 21 publications

Short-term forecasting of passenger demand under on-demand ride services: A spatio-temporal deep learning approach

Short-term forecasting of passenger demand under on-demand ride services: A spatio-temporal deep learning approach

Optimal passenger-seeking policies on E-hailing platforms using Markov decision process and imitation learning

Understanding the Effect of an E-Hailing App Subsidy War on Taxicab Operation Zones

Contact Info

Product

Resources

About