Real-Time Driver-Request Assignment in Ridesourcing

Authors: Hao Wang, Xiaohui Bei3840-3849

AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We prove that our algorithm is optimal ex-ante in the single-request setting, and demonstrate its effectiveness in the general multi-request setting through experiments on both synthetic and real-world datasets.
Researcher Affiliation Academia School of Physical and Mathematical Sciences, Nanyang Technological University wang1242@e.ntu.edu.sg, xhbei@ntu.edu.sg
Pseudocode Yes The pseudocode can be found in Algorithm 1.
Open Source Code No The paper does not include an unambiguous statement or link indicating that the source code for their methodology is publicly available.
Open Datasets Yes We use a publicly available New York City taxi dataset (Donovan and Work 2014) (under CC0 license) and extract 100, 000 taxi trip records as our real-world data.
Dataset Splits No The paper mentions using synthetic and real-world data for experiments but does not provide specific train/validation/test dataset splits, percentages, or absolute counts.
Hardware Specification Yes We use a computer with 2.2 GHz Intel Core i7 processor, 16 GB 1600 MHz DDR3 memory and Intel Iris Pro 1536 MB Graphics to run all the experiments.
Software Dependencies Yes We use (Gurobi Optimization 2021) as our linear program solver in our code.
Experiment Setup Yes We introduce two parameters: the matching factor CL and the waiting factor TL. CL influences the value of edge cost crd and represents the relative importance of matching cost and waiting cost. TL is only used by synthetic data and influences the quitting distribution of requests. Larger TL means requests are more likely to wait for a longer time.