# Optimizing Taxi Carpool Policies via Reinforcement Learning and Spatio-Temporal Mining

arXiv: https://arxiv.org/abs/1811.04345 · HTML: https://mlsanity.com/paper/1811.04345/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Nov 2018.

## Built on by (most cited first, 8 of 8)

- [Reinforcement Learning for Ridesharing: An Extended Survey](https://mlsanity.com/paper/2105.01099/) (2021)
- [Deep Reinforcement Learning for Demand Driven Services in Logistics and Transportation Systems: A Survey](https://mlsanity.com/paper/2108.04462/) (2021)
- [A Survey of Machine Learning-Based Ride-Hailing Planning](https://mlsanity.com/paper/2303.14646/) (2023)
- [Interval Deep Learning for Uncertainty Quantification in Safety Applications](https://mlsanity.com/paper/2105.06438/) (2021)
- [An Explainable Stacked Ensemble Model for Static Route-Free Estimation of Time of Arrival](https://mlsanity.com/paper/2203.09438/) (2022)
- [CoordiQ : Coordinated Q-learning for Electric Vehicle Charging Recommendation](https://mlsanity.com/paper/2102.00847/) (2021)
- [Modelling and Reasoning Techniques for Context Aware Computing in Intelligent Transportation System](https://mlsanity.com/paper/2107.14374/) (2021)
- [Non-myopic Matching and Rebalancing in Large-Scale On-Demand Ride-Pooling Systems Using Simulation-Informed Reinforcement Learning](https://mlsanity.com/paper/2510.25796/) (2025)

## Abstract (arXiv)

In this paper, we develop a reinforcement learning (RL) based system to learn an effective policy for carpooling that maximizes transportation efficiency so that fewer cars are required to fulfill the given amount of trip demand. For this purpose, first, we develop a deep neural network model, called ST-NN (Spatio-Temporal Neural Network), to predict taxi trip time from the raw GPS trip data. Secondly, we develop a carpooling simulation environment for RL training, with the output of ST-NN and using the NYC taxi trip dataset. In order to maximize transportation efficiency and minimize traffic congestion, we choose the effective distance covered by the driver on a carpool trip as the reward. Therefore, the more effective distance a driver achieves over a trip (i.e. to satisfy more trip demand) the higher the efficiency and the less will be the traffic congestion. We compared the performance of RL learned policy to a fixed policy (which always accepts carpool) as a baseline and obtained promising results that are interpretable and demonstrate the advantage of our RL approach. We also compare the performance of ST-NN to that of state-of-the-art travel time estimation methods and observe that ST-NN significantly improves the prediction performance and is more robust to outliers.