To reduce the vehicle relocation rate considering relieving disequilibrium of the supply-demand ratios across regions for car-sharing systems, in this paper, we propose a data-driven optimization framework by integrating the non-parametric learning algorithm and two-stage stochastic programming modeling technique to address the one-way station-based car-sharing relocation problem. In contrast with the most existing work that deals with demand uncertainty using predefined probability distributions, the learning-based framework is capable of handling demand uncertainty by learning the intrinsic pattern from large-scale historical data and computing high quality solutions. To validate the performance of our proposed approach, we conduct a group of numerical experiments based on New York taxicab trip record data set. The experimental results show that our proposed data-driven approach outperforms the parametric approaches and deterministic model in terms of business profit, relocation rate, and value of stochastic solution (VSS). Most significantly, compared with the deterministic approach, the vehicle relocation rates are reduced by approximate 80%, 70% and 40% under small fleet size, medium fleet size and large fleet size, respectively. In addition, the VSS of our approach is more than 3 times higher than the one of Poisson distribution by average.

@en

We propose a learning-based approach for open driver guidance and rebalancing in ride-hailing platforms. The objective is to further enhance the wait time reduction benefit of batched matching by incorporating learning-based open driver guidance and rebalancing. By leveraging the rider demand data, the guidance solutions are computed through the integration of machine learning techniques with a two-stage stochastic programming model. To validate the performance of the proposed approach, we conduct numerical experiments using the New York taxi trip data sets. Our results show that the proposed approach outperforms the single value estimation model and the parametric model using Poisson distribution in terms of average wait time. When assuming the open drivers are randomly located before the batching time window, the proposed approach reduces more than 70% of average wait time compared to batched matching without guidance. @en

In this paper, we propose a data-driven robust optimization model to reduce total travel cost in ride-sharing systems under travel time uncertainty. Instead of using a pre-defined uncertainty set, we study a data-driven robust optimization approach that integrates gated recurrent units (GRUs) predictions with a one-stage robust optimization model. The proposed approach has the ability to compute high quality solutions by leveraging the large-scale historical data to derive the uncertainty set for the designed robust optimization model. To evaluate the proposed approach, we conduct a group of simulations based on the New York taxi trip record data sets. The validation results show that our data-driven robust optimization approach outperforms the robust optimization approach with a pre-defined uncertainty set in terms of travellers' total travel cost. Most importantly, the total travel cost under the proposed approach is reduced by up to 31.7%, and by 26.7% on average compared with the robust model with a pre-defined uncertainty set.

@en

In ride-sharing services, travel time uncertainty significantly impacts the quality of matching solutions for both the drivers and the riders. This paper studies a one-to-many ride-sharing matching problem where travel time between locations is uncertain. The goal is to generate robust ride-sharing matching solutions that minimize the total driver detour cost and the number of unmatched riders. To this end, we formulate the ride-sharing matching problem as a robust vehicle routing problem with time window (RVRPTW). To effectively capture the travel time uncertainty, we propose a deep learning-based data-driven approach that can dynamically estimate the uncertainty sets of travel times. Given the NP-hard nature of the optimization problem, we design a hybrid meta-heuristic algorithm that can handle large-scale instances in a time-efficient manner. To evaluate the performance of the proposed method, we conduct a set of numeric experiments based on real traffic data. The results confirm that the proposed approach outperforms the non-data-driven one in several important performance metrics, including a proper balance between robustness and inclusiveness of the matching solution. Specifically, by applying the proposed data-driven approach, the matching solution violation rate can be reduced up to 85.8%, and the valid serving rate can be increased up to 42.3% compared to the non-data-driven benchmark.@en

Freelance drivers in ride-hailing systems may strategically accept or reject ride requests based on their projection of the profitability of the assigned rides. This driver acceptance uncertainty is mainly caused by the flat rate payment and the blind ride acceptance rule adopted by most ride-hailing platforms. As a result, a high driver rejection rate has been observed, causing a negative impact on the service quality and matching efficiency for the ride-hailing systems. In this paper, we propose a pricing mechanism to improve drivers’ average ride acceptance rate by offering personalized payments computed based on the characteristics of individual riders and the estimated acceptance rates of the drivers. Specifically, we model and predict the drivers’ ride acceptance rates through a binary choice model and incorporate it into the stochastic optimization problem for the ride-hailing system. This provides personalized payment for each driver in connection with the characteristics of the assigned ride and the preferences of the drivers. We then evaluate the performance of the proposed pricing mechanism through extensive numerical experiments based on RideAustin trip data from June 2016 to April 2017. The results suggest that our proposed pricing mechanism improves the drivers’ average acceptance rate by an average of 60% compared to some commonly used pricing schemes. It also significantly increases the platform’s expected profit and matching rate. This implies a strong potential for the proposed pricing mechanism to improve service reliability and quality in ride-hailing systems.@en

This paper proposes an integrated dispatching framework for matching drivers with riders in ride-hailing systems. The goal is to compute matching solutions that maximize social welfare and benefit both sides of the market, such that the sustainable growth of the ride-hailing system is ensured. The proposed framework integrates data-driven proactive guidance strategies with batched matching optimization to increase social welfare, improve matching rate and reduce rider wait time. Proactive guidance strategies are computed by leveraging short-term demand forecasts based on historical data. Taken the resulting guidance strategies as inputs, the batched matching algorithm computes optimal bipartite matching between drivers and riders in a batch. Using New York City taxi data from 2016 March 1st to March 31st as input, we conduct a numerical study to evaluate the performance of the proposed framework and compare it with existing approaches in the literature. Our results show that the proposed framework improves social welfare for up to 50%. It also increases the matching rate by an average of 20% and reduces the average rider wait time by over 15%. This implies a strong potential for the proposed dispatching framework to improve service quality in ride-hailing systems.@en

Vehicle proactive guidance strategies are used by ride-hailing platforms to mitigate supply–demand imbalance across regions by directing idle vehicles to high-demand regions before the demands are realized. This article presents a data-driven stochastic optimization framework for computing idle vehicle guidance strategies. The objective is to minimize drivers’ idle travel distance, riders’ wait time, and the oversupply costs (OSCs) and undersupply costs (USCs) of the platform. Specifically, we design a novel neural network that integrates gated recurrent units (GRUs) with mixture density networks (MDNs) to capture the spatial-temporal features of the rider demand distribution.@en