Community-Based Trip Sharing for Urban Commuting

Authors: Mohd. Hafiz Hasan, Pascal Van Hentenryck, Ceren Budak, Jiayu Chen, Chhavi Chaudhry

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results show that community-based trip sharing reduces daily car usage by up to 44%, thus producing significant environmental and traffic benefits and reducing parking pressure.
Researcher Affiliation Academia University of Michigan Ann Arbor, Michigan 48109
Pseudocode Yes Algorithm 1 Shareability Graph for Inbound Trips" and "Algorithm 2 All Feasible Inbound Trip Sharing Routes
Open Source Code No The paper does not provide any specific links or statements about releasing the source code for the described methodology.
Open Datasets No The dataset contains access information of 15 parking structures located in downtown Ann Arbor. Each entry contains the ID, access time, and direction (in/out) of each customer throughout April 2017. This information was joined with the home address of every customer to reconstruct their daily trips.
Dataset Splits No The paper describes using a dataset of 15,000 commuters for its study but does not provide explicit training, validation, and test dataset splits with percentages or sample counts for reproducibility.
Hardware Specification Yes All models were executed on a high-performance computing cluster with 8 cores of a 2.5 GHz Intel Xeon E5-2680v3 processor, 64 GB of RAM, and a time limit of 120 hours.
Software Dependencies Yes The clustering stage converts all GPS coordinates to local Cartesian coordinates before applying MATLAB 2016b s clusterdata function. The rest of the algorithm was implemented in C++, using GUROBI 6.5.2 for solving the MIPs.
Experiment Setup Yes The values σlimit = 2 miles and Δ = 20 mins were used for most experiments. We also include results for σlimit = 2 1 miles and Δ = 20 10 mins to demonstrate the algorithm s sensitivity to these parameters. All models were executed on a high-performance computing cluster with 8 cores of a 2.5 GHz Intel Xeon E5-2680v3 processor, 64 GB of RAM, and a time limit of 120 hours.