A scalable heuristic algorithm for demand responsive transportation for first mile transit

First/last mile transit using public transport has consistently been a bottleneck for commuters due to the relatively higher time spent in these legs when compared to the overall journey. Recently, demand responsive transportation(DRT) services have been proposed for the first/last mile transit. However, in contrast to the requirements of a public transportation system, existing DRT services either match only a few passengers to a vehicle or require advance booking. Hence, in this paper we propose a DRT system, specifically for the First mile transit, by matching multiple passengers to a vehicle in real-time. We first model the problem as a convergent graph and obtain an exact solution. Next, a scalable heuristic algorithm has been proposed that not only provides near optimal solution, but also does that in real-time (≈ms) as opposed to minutes/hours taken for the exact solution.

[1]  Sophie N. Parragh.  Introducing heterogeneous users and vehicles into models and algorithms for the dial-a-ride problem , 2011, Transportation research. Part C, Emerging technologies.

[2]  Fan Zhang,et al.  Last-Mile Transit Service with Urban Infrastructure Data , 2016, ACM Trans. Cyber Phys. Syst..

[3]  David Koffman,et al.  Operational Experiences with Flexible Transit Services , 2004 .

[4]  Angelos Amditis,et al.  Public transport optimisation based on traveller requests and network efficiency , 2014 .

[5]  E. Osaba,et al.  An Asymmetric Multiple Traveling Salesman Problem with Backhauls to solve a Dial-a-Ride problem , 2015, 2015 IEEE 13th International Symposium on Applied Machine Intelligence and Informatics (SAMI).

[6]  Guy Desaulniers,et al.  The Vehicle Routing Problem with Time Windows , 2014, Vehicle Routing.

[7]  E. Deakin,et al.  Markets for Dynamic Ridesharing? , 2010 .

[8]  Fernando Ordóñez,et al.  Ridesharing: The state-of-the-art and future directions , 2013 .

[9]  Martin W. P. Savelsbergh,et al.  Optimization for dynamic ride-sharing: A review , 2012, Eur. J. Oper. Res..

[10]  Vinay Kolar,et al.  Singapore in Motion: Insights on Public Transport Service Level Through Farecard and Mobile Data Analytics , 2016, KDD.

[11]  Gilbert Laporte,et al.  A Tabu Search Heuristic for the Static Multi-Vehicle Dial-a-Ride Problem , 2002 .

[12]  J. Preston,et al.  The demand for public transport: The effects of fares, quality of service, income and car ownership , 2006 .

[13]  Bernhard Nebel,et al.  A Mechanism for Dynamic Ride Sharing Based on Parallel Auctions , 2011, IJCAI.

[14]  Philippe Canalda,et al.  Comparison of three algorithms for solving the convergent demand responsive transportation problem , 2006, 2006 IEEE Intelligent Transportation Systems Conference.

[15]  Niels A. H. Agatz,et al.  The Value of Optimization in Dynamic Ride-Sharing: A Simulation Study in Metro Atlanta , 2010 .

[16]  Sarvapali D. Ramchurn,et al.  Algorithms for Electric Vehicle Scheduling in Mobility-on-Demand Schemes , 2015, 2015 IEEE 18th International Conference on Intelligent Transportation Systems.

[17]  E. Frazzoli,et al.  Autonomous personal vehicle for the first- and last-mile transportation services , 2011, 2011 IEEE 5th International Conference on Cybernetics and Intelligent Systems (CIS).