ABSTRACT
As an alternative means of convenient and smart transportation, mobility-on-demand (MOD), typified by online ride-sharing and connected taxicabs, has been rapidly growing and spreading worldwide. The large volume of complex traffic and the uncertainty of market supplies/demands have made it essential for many MOD service providers to proactively dispatch vehicles towards ride-seekers.
To meet this need effectively, we propose STRide, an MOD coordination-learning mechanism reinforced spatio-temporally with capsules. We formalize the adaptive coordination of vehicles into a reinforcement learning framework. STRide incorporates spatial and temporal distributions of supplies (vehicles) and demands (ride requests), customers' preferences and other external factors. A novel spatio-temporal capsule neural network is designed to predict the provider's rewards based on MOD network states, vehicles and their dispatch actions. This way, the MOD platform adapts itself to the supply-demand dynamics with the best potential rewards. We have conducted extensive data analytics and experimental evaluation with three large-scale datasets (~ 21 million rides from Uber, Yellow Taxis and Didi). STRide is shown to outperform state-of-the-arts, substantially reducing request-rejection rate and passenger waiting time, and also increasing the service provider's profits, often making 30% improvement over state-of-the-arts.
- 2018. Global Petrol Prices. https://www.globalpetrolprices.com/gasoline_prices/.Google Scholar
- 2018. National Centers for Environmental Information, National Oceanic and Atmospheric Association (NOAA) - Data Tools: Local Climatological Data (LCD). https://www.ncdc.noaa.gov/cdo-web/datatools/lcd.Google Scholar
- 2018. Open Street Map. https://www.openstreetmap.org/.Google Scholar
- 2018. TLC Trip Record Data. http://www.nyc.gov/html/tlc/html/about/trip_record_ data.shtml.Google Scholar
- 2018. Uber pickups in New York City.https://www.kaggle.com/fivethirtyeight/uber-pickups-in-new-york-city/data.Google Scholar
- 2019. Didi Chuxing Technology Co.www.didiglobal.com.Google Scholar
- Niels Agatz, Alan Erera, Martin Savelsbergh, and Xing Wang. 2012. Optimization for dynamic ride-sharing: A review. European Journal of Operational Research 223, 2 (2012), 295 - 303.Google ScholarCross Ref
- Siddhartha Banerjee, Ramesh Johari, and Carlos Riquelme. 2015. Pricing in Ride-Sharing Platforms: A Queueing-Theoretic Approach. In Proc. ACM EC. 639-639. Google ScholarDigital Library
- Qingpeng Cai, Aris Filos-Ratsikas, Pingzhong Tang, and Yiwei Zhang. 2018. Reinforcement Mechanism Design for e-Commerce. In Proc. WWW. 1339-1348. Google ScholarDigital Library
- Rachel Dovey. 2017. 5 Florida Cities Team Up to Subsidize Uber Rides. https://nextcity.org/daily/entry/five-florida-cities-subsidize-uber-rides.Google Scholar
- Zhixuan Fang, Longbo Huang, and Adam Wierman. 2017. Prices and Subsidies in the Sharing Economy. In Proc. WWW. 53-62. Google ScholarDigital Library
- Yong Gao, Dan Jiang, and Yan Xu. 2018. Optimize taxi driving strategies based on reinforcement learning. IJGIS 32, 8 (2018), 1677-1696.Google ScholarCross Ref
- Ian Goodfellow, Yoshua Bengio, Aaron Courville, and Yoshua Bengio. 2016. Deep Learning. Vol. 1. MIT Press Cambridge. Google ScholarDigital Library
- Jiawei Han, Jian Pei, and Micheline Kamber. 2011. Data mining: Concepts and techniques. Elsevier. Google ScholarDigital Library
- Suining He and Kang G. Shin. 2018. (Re)Configuring Bike Station Network via Crowdsourced Information Fusion and Joint Optimization. In Proc. ACM MobiHoc. 1-10. Google ScholarDigital Library
- Geoffrey E Hinton, Sara Sabour, and Nicholas Frosst. 2018. Matrix capsules with EM routing. In Proc. ICLR.Google Scholar
- Josh Horwitz. 2017. One year after the Uber-Didi merger, it's only getting harder to hail a ride in China. Respondent: https://qz.com/1045268/one-year-after-the-uber-didi-merger-its-only-getting-harder-to-hail-a-ride-in-china/and waiting time: http://news.sina.com.cn/c/2017-07-26/doc-ifyinryq6222913.shtml.Google Scholar
- Yaguang Li, Kun Fu, Zheng Wang, Cyrus Shahabi, Jieping Ye, and Yan Liu. 2018. Multi-task Representation Learning for Travel Time Estimation. In Proc. ACM KDD. 1695-1704. Google ScholarDigital Library
- Yaguang Li, Han Su, Ugur Demiryurek, Bolong Zheng, Tieke He, and Cyrus Shahabi. 2017. PaRE: A System for Personalized Route Guidance. In Proc. WWW. 637-646. Google ScholarDigital Library
- Yexin Li, Yu Zheng, and Qiang Yang. 2018. Dynamic Bike Reposition: A Spatio-Temporal Reinforcement Learning Approach. In Proc. ACM KDD. 1724-1733. Google ScholarDigital Library
- Kaixiang Lin, Renyu Zhao, Zhe Xu, and Jiayu Zhou. 2018. Efficient Large-Scale Fleet Management via Multi-Agent Deep Reinforcement Learning. In Proc. ACM KDD. 1774-1783. Google ScholarDigital Library
- Zhidan Liu, Zhenjiang Li, Kaishun Wu, and Mo Li. 2018. Urban Traffic Prediction from Mobility Data Using Deep Learning. IEEE Network 32, 4 (July 2018), 40-46.Google ScholarDigital Library
- Chenglin Miao, Qi Li, Lu Su, Mengdi Huai, Wenjun Jiang, and Jing Gao. 2018. Attack Under Disguise: An Intelligent Data Poisoning Attack Mechanism in Crowdsourcing. In Proc. WWW. 13-22. Google ScholarDigital Library
- Fei Miao, Shuo Han, Shan Lin, John A Stankovic, Desheng Zhang, Sirajum Munir, Hua Huang, Tian He, and George J Pappas. 2016. Taxi Dispatch With Real-Time Sensing Data in Metropolitan Areas: A Receding Horizon Control Approach. IEEE Trans. Automation Science and Engineering 13, 2 (April 2016), 463-478.Google ScholarCross Ref
- Takuma Oda and Carlee Joe-Wong. 2018. MOVI: A Model-Free Approach to Dynamic Fleet Management. In Proc. IEEE INFOCOM. 2708-2716.Google Scholar
- Ling Pan, Qingpeng Cai, Zhixuan Fang, Pingzhong Tang, and Longbo Huang. 2018. A Deep Reinforcement Learning Framework for Rebalancing Dockless Bike Sharing Systems. In Proc. AAAI.Google Scholar
- Lisa Rayle, Danielle Dai, Nelson Chan, Robert Cervero, and Susan Shaheen. 2016. Just a better Taxi? A survey-based comparison of taxis, transit, and ridesourcing services in San Francisco. Transport Policy 45(2016), 168-178.Google ScholarCross Ref
- Sara Sabour, Nicholas Frosst, and Geoffrey E Hinton. 2017. Dynamic routing between capsules. In Proc. NIPS. 3856-3866. Google ScholarDigital Library
- Richard S Sutton and Andrew G Barto. 1998. Introduction to Reinforcement Learning. Vol. 135. MIT Press Cambridge. Google ScholarDigital Library
- Jian Tang, Meng Qu, Mingzhe Wang, Ming Zhang, Jun Yan, and Qiaozhu Mei. 2015. LINE: Large-scale Information Network Embedding. In Proc. WWW. 1067-1077. Google ScholarDigital Library
- Hado Van Hasselt, Arthur Guez, and David Silver. 2016. Deep Reinforcement Learning with Double Q-Learning.. In Proc. AAAI, Vol. 2. 5. Google ScholarDigital Library
- Erwin Walraven, Matthijs TJ Spaan, and Bram Bakker. 2016. Traffic flow optimization: A reinforcement learning approach. Engineering Applications of Artificial Intelligence 52 (2016), 203-212. Google ScholarDigital Library
- Dong Wang, Wei Cao, Jian Li, and Jieping Ye. 2017. DeepSD: Supply-demand prediction for online car-hailing services using deep neural networks. In Proc. IEEE ICDE. 243-254.Google ScholarCross Ref
- Zheng Wang, Kun Fu, and Jieping Ye. 2018. Learning to Estimate the Travel Time. In Proc. ACM KDD. 858-866. Google ScholarDigital Library
- Hua Wei, Guanjie Zheng, Huaxiu Yao, and Zhenhui Li. 2018. Intellilight: A reinforcement learning approach for intelligent traffic light control. In Proc. ACM SIGKDD. 2496-2505. Google ScholarDigital Library
- Jian Wen, Jinhua Zhao, and Patrick Jaillet. 2017. Rebalancing shared mobility-on-demand systems: A reinforcement learning approach. In Proc. IEEE ITSC. 220-225.Google Scholar
- Zhe Xu, Zhixin Li, Qingwen Guan, Dingshui Zhang, Qiang Li, Junxiao Nan, Chunyang Liu, Wei Bian, and Jieping Ye. 2018. Large-Scale Order Dispatch in On-Demand Ride-Hailing Platforms: A Learning and Planning Approach. In Proc. ACM KDD. 905-913. Google ScholarDigital Library
- Junbo Zhang, Yu Zheng, and Dekang Qi. 2017. Deep Spatio-Temporal Residual Networks for Citywide Crowd Flows Prediction.. In Proc. AAAI. 1655-1661. Google ScholarDigital Library
- Junbo Zhang, Yu Zheng, Dekang Qi, Ruiyuan Li, Xiuwen Yi, and Tianrui Li. 2018. Predicting citywide crowd flows using deep spatio-temporal residual networks. Artificial Intelligence 259 (2018), 147 - 166.Google ScholarCross Ref
- Huanyang Zheng and Jie Wu. 2017. Online to Offline Business: Urban Taxi Dispatching with Passenger-Driver Matching Stability. In Proc. IEEE ICDCS. 816-825.Google Scholar
Recommendations
Spatio-temporal Adaptive Pricing for Balancing Mobility-on-Demand Networks
Survey Papers and Regular PapersPricing in mobility-on-demand (MOD) networks, such as Uber, Lyft, and connected taxicabs, is done adaptively by leveraging the price responsiveness of drivers (supplies) and passengers (demands) to achieve such goals as maximizing drivers’ incomes, ...
Demand-responsive rebalancing zone generation for reinforcement learning-based on-demand mobility
Agents in Traffic and Transportation (ATT 2020)Enabling Ride-sharing (RS) in Mobility-on-demand (MoD) systems allows reduction in vehicle fleet size while preserving the level of service. This, however, requires an efficient vehicle to request assignment, and a vehicle rebalancing strategy, which ...
dFDA-VeD: A Dynamic Future Demand Aware Vehicle Dispatching System
MobiQuitous '20: MobiQuitous 2020 - 17th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and ServicesWith the rising demand of smart mobility, ride-hailing service is getting popular in the urban regions. These services maintain a system for serving the incoming trip requests by dispatching available vehicles to the pickup points. As the process ...
Comments